Hugging Face's logo

AutomatosX
/
AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit

Text Generation
MLX
Safetensors
mistral3
apple-silicon
quantized
mixed-precision
axquant
axq
development
4bit
4-bit precision
v2
vision
conversational
Model card Files
xet
 Community
AutomatosX commited on
Commit
a41e012
·
verified ·
1 Parent(s): 04046b2

Add development AXQ pack AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit

Browse files
Files changed (14) hide show
  1. .gitattributes +1 -0
  2. README.md +172 -0
  3. axquant_manifest.json +186 -0
  4. axquant_plan.json +0 -0
  5. axquant_quantizer_execution.json +2817 -0
  6. axquant_runtime.json +58 -0
  7. chat_template.jinja +216 -0
  8. config.json +2865 -0
  9. generation_config.json +7 -0
  10. model-00001-of-00002.safetensors +3 -0
  11. model-00002-of-00002.safetensors +3 -0
  12. model.safetensors.index.json +933 -0
  13. tokenizer.json +3 -0
  14. tokenizer_config.json +1015 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,172 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: mlx
4
+ base_model: mistralai/Ministral-3-14B-Instruct-2512
5
+ base_model_relation: quantized
6
+ pipeline_tag: text-generation
7
+ tags:
8
+ - mlx
9
+ - apple-silicon
10
+ - quantized
11
+ - mixed-precision
12
+ - axquant
13
+ - axq
14
+ - development
15
+ - mistral3
16
+ - 4bit
17
+ - 4-bit
18
+ ---
19
+
20
+ # AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit
21
+
22
+ An **AXQuant (AXQ)** mixed-precision MLX checkpoint for Apple Silicon, converted directly from
23
+ the BF16 source model. The language path is quantized under AXQuant protection floors (embeddings, norms, and other protected tensors remain higher precision).
24
+
25
+ > **Development evidence — not a certified AXQuant release.** This package has conversion and
26
+ > artifact-integrity records, but it does not publish measured quality, long-context, kernel-speed,
27
+ > or MTP-speed evidence. Do not interpret the AXQ product label as a benchmark claim.
28
+
29
+ ## Model details
30
+
31
+ | Property | Value |
32
+ | --- | --- |
33
+ | Base model | [mistralai/Ministral-3-14B-Instruct-2512](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) |
34
+ | Source revision | `unrecorded` |
35
+ | Product family | `mistral3` |
36
+ | Source architecture | `Mistral3ForConditionalGeneration` (dense); text path optimized |
37
+ | Main-model parameters | 13.51B logical parameters |
38
+ | Quantizer | AXQuant `1.0.1` |
39
+ | Hub budget class | `4bit` |
40
+ | AXQuant base precision class | `4bit` |
41
+ | Planned storage-adjusted BPW | 5.2798 |
42
+ | Measured main-model BPW | 5.2799 |
43
+ | Measured total BPW | **5.2799** |
44
+ | Safetensors weight size | 8.91 GB |
45
+ | Approximate complete download | 8.93 GB |
46
+ | Configured maximum context | 262,144 tokens; practical limits depend on unified memory |
47
+ | Primary runtime | AX Engine, compatibility level A |
48
+ | Compatible runtime | MLX-LM standard text inference, compatibility level B |
49
+ | MTP present | `False` |
50
+ | Vision sidecar present | `False` |
51
+
52
+ This repository contains MLX Safetensors. It does **not** contain PyTorch or GGUF weights.
53
+
54
+ ## Choosing an AXQ pack
55
+
56
+ AXQ names describe a **storage-budget product class**, not one uniform precision applied to every
57
+ tensor. Protected tensors remain at higher precision, so the exact measured BPW is authoritative.
58
+ In particular, a `6bit`-named mixed plan may retain `4bit` as its base precision while selecting
59
+ 6-bit, 8-bit, or BF16 for other tensors to meet an approximately 6-BPW total budget. Protection
60
+ floors can also raise a `4bit`-named pack close to (or above) a `6bit` budget on small or heavily
61
+ protected models.
62
+
63
+ | Sibling | Intended trade-off |
64
+ | --- | --- |
65
+ | [4bit sibling](https://huggingface.co/AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit) | Lower-storage AXQ budget; check its exact BPW |
66
+ | [6bit sibling](https://huggingface.co/AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-6bit) | Higher average precision near a 6-BPW budget |
67
+
68
+ See the [AutomatosX MLX model catalog](https://huggingface.co/collections/AutomatosX/automatosx-mlx-model-catalog)
69
+ for related MLX and OptiQ alternatives.
70
+
71
+ ## Download
72
+
73
+ ```bash
74
+ python -m pip install -U huggingface_hub
75
+ hf download AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit --local-dir ./AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit
76
+ ```
77
+
78
+ Allow at least 8.93 GB of free disk space. Pin the resulting Hub commit in reproducible
79
+ deployments rather than relying indefinitely on `main`.
80
+
81
+ ## Run with MLX-LM
82
+
83
+ ```bash
84
+ python -m pip install -U mlx-lm
85
+ mlx_lm.generate \
86
+ --model AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit \
87
+ --prompt "Explain mixed-precision quantization in three sentences." \
88
+ --max-tokens 128 \
89
+ --temp 0.0
90
+ ```
91
+
92
+ MLX-LM compatibility covers standard **text/backbone inference**. It may ignore AXQuant runtime
93
+ metadata and optional sidecars (`vision.safetensors`, `mtp.safetensors`); this command therefore
94
+ does not establish MTP acceleration or vision-language quality. The artifact records MLX
95
+ `0.32.0` and MLX-LM `0.31.3` from conversion.
96
+
97
+ ## Serve with AX Engine
98
+
99
+ After installing [AX Engine](https://github.com/defai-digital/ax-engine), download the complete
100
+ repository and serve the local directory:
101
+
102
+ ```bash
103
+ ax-engine serve ./AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit --port 31418
104
+ ```
105
+
106
+ AX Engine is the authority for the AXQ runtime contract.
107
+ This development package does not claim runtime speedups until identical-checkpoint benchmarks are
108
+ published. The artifact records AX Engine version `not recorded`. Native
109
+ `model-manifest.json` status: not included.
110
+
111
+ ## Quantization layout
112
+
113
+ | Main-weight precision | Parameters | Share |
114
+ | --- | ---: | ---: |
115
+ | `4bit` | 12.16B | 90.06% |
116
+ | `8bit` | 671.09M | 4.97% |
117
+ | `bf16` | 671.50M | 4.97% |
118
+
119
+ - Quantization methods: `affine, bf16`.
120
+ - Group sizes used by quantized assignments: `32, 64`.
121
+ - MTP sidecar: not included.
122
+ - Vision sidecar: not included.
123
+ - Optimization scope: `text-path`.
124
+ - Support tier: `convertible`.
125
+
126
+ BF16 sidecars, when present, are included in total download size. Their presence does not by itself
127
+ establish MTP acceleration or vision-language quality.
128
+
129
+ ## Evidence and validation status
130
+
131
+ | Check | Status |
132
+ | --- | --- |
133
+ | Planning evidence | `architecture_prior` |
134
+ | Calibration | none; the allocation is based on architecture priors |
135
+ | Quantizer execution | 281/281 recorded module conversions succeeded; 0 fallbacks |
136
+ | AX Engine native manifest | not included |
137
+ | Quality versus BF16 or uniform baselines | Not published; no quality-retention claim |
138
+ | MTP acceptance and speed | not measured; no MTP speedup claim |
139
+ | AX Engine kernel evidence | `unmeasured` |
140
+ | Vision-language quality | Not applicable (no vision sidecar in this package) |
141
+ | Long-context quality | 262,144-token capacity is config metadata, not a validated claim |
142
+ | Release certification | **Not certified**; formal AXQuant M0-M8 gates are not closed |
143
+
144
+ ## Intended use and limitations
145
+
146
+ - Intended for local development and evaluation on Apple Silicon with MLX-compatible runtimes.
147
+ - No minimum unified-memory figure is claimed; loadability depends on model size, context length,
148
+ KV-cache policy, runtime buffers, and other processes using unified memory.
149
+ - Architecture-prior allocation is not measured sensitivity. It must not be presented as measured
150
+ model quality.
151
+ - The configured context window can require substantially more memory as the KV cache grows.
152
+ - Upstream capabilities, limitations, biases, and responsible-use guidance still apply.
153
+
154
+ ## Provenance and audit files
155
+
156
+ - [`axquant_manifest.json`](axquant_manifest.json): package identity, byte accounting, runtime
157
+ contract, software versions, and file checksums.
158
+ - [`axquant_plan.json`](axquant_plan.json): per-tensor precision decisions and planning evidence.
159
+ - [`axquant_quantizer_execution.json`](axquant_quantizer_execution.json): conversion coverage and
160
+ fallback records.
161
+ - [`axquant_runtime.json`](axquant_runtime.json): AX Engine and MLX-LM compatibility contract.
162
+
163
+ All published provenance uses repository-relative paths. Local source paths are stripped before
164
+ publication. The checkpoint was converted from BF16 rather than re-quantized from an OptiQ
165
+ artifact. Parallel OptiQ repositories use a different quantizer and should not be assumed to have
166
+ identical BPW or quality.
167
+
168
+ ## License
169
+
170
+ The checkpoint follows the upstream model license where applicable (often Apache License 2.0). See
171
+ the [mistralai/Ministral-3-14B-Instruct-2512 model card](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) for license terms, model
172
+ limitations, and responsible-use guidance.
axquant_manifest.json ADDED
@@ -0,0 +1,186 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "axquant_version": "1.0.1",
3
+ "calibration": null,
4
+ "created_at": "2026-08-02T19:52:56.105649Z",
5
+ "effective_bpw": 5.279832291155292,
6
+ "files": [
7
+ {
8
+ "path": "README.md",
9
+ "sha256": "efee8e591496ed537a3fc7d0e8d2090a6fbefd03d670ef7fbaacb087b34befff",
10
+ "size_bytes": 7456
11
+ },
12
+ {
13
+ "path": "axquant_plan.json",
14
+ "sha256": "02b20463cc90ca81dcd03492b246ab9b7f4c97bed3f248052deee83105b74144",
15
+ "size_bytes": 347258
16
+ },
17
+ {
18
+ "path": "axquant_quantizer_execution.json",
19
+ "sha256": "c97f9283ba562ba735d4cefc191e35f7b7964318b9325c08d5ebd47fd0e2a76e",
20
+ "size_bytes": 72744
21
+ },
22
+ {
23
+ "path": "axquant_runtime.json",
24
+ "sha256": "5715430cce19e6ec6627c9d8e4d07c96cb935cf21582561db7734b22480f6042",
25
+ "size_bytes": 1643
26
+ },
27
+ {
28
+ "path": "chat_template.jinja",
29
+ "sha256": "2f545122222db8bb43ca0ea0c49e9185320a8670f7d35575b0da0eb48b1e8970",
30
+ "size_bytes": 11913
31
+ },
32
+ {
33
+ "path": "config.json",
34
+ "sha256": "d444f78fddd3e95b0f2a18725a0ac72a923b79deb3adbe9f75b9b2abd1a92342",
35
+ "size_bytes": 87042
36
+ },
37
+ {
38
+ "path": "generation_config.json",
39
+ "sha256": "e0923390059f84a9180b00e5501778acc45ea9856cd7f2fd68208b360927c677",
40
+ "size_bytes": 131
41
+ },
42
+ {
43
+ "path": "model-00001-of-00002.safetensors",
44
+ "sha256": "2581a3f8a5beb43f361c0dd2886cdff436bdbad90a2b46ca3e13864e3b1e1f41",
45
+ "size_bytes": 5361479864
46
+ },
47
+ {
48
+ "path": "model-00002-of-00002.safetensors",
49
+ "sha256": "7e6eba9b62bd4e3ed0035efc5254d961768e21915feb6c9adc84c52c6d91561d",
50
+ "size_bytes": 3552365268
51
+ },
52
+ {
53
+ "path": "model.safetensors.index.json",
54
+ "sha256": "56d962482ac45a61674b07033f2a8b5a9ef1106568eddc5148dfcaa789132e9e",
55
+ "size_bytes": 93118
56
+ },
57
+ {
58
+ "path": "tokenizer.json",
59
+ "sha256": "99cf274236c60277fcfad861a5a1007518687ad06ba8938760f50b55ffa0b1ef",
60
+ "size_bytes": 17077420
61
+ },
62
+ {
63
+ "path": "tokenizer_config.json",
64
+ "sha256": "790c3e17aefff6a4e0daca94a2dfbb17d4db65df0421ed3bc7759cda526b93c2",
65
+ "size_bytes": 21222
66
+ }
67
+ ],
68
+ "format": "mlx",
69
+ "logical_parameters": 13506073600,
70
+ "main_logical_parameters": 13506073600,
71
+ "main_weight_file_size_bytes": 8913845132,
72
+ "measured_main_bpw": 5.2799031878517235,
73
+ "measured_total_bpw": 5.2799031878517235,
74
+ "mtp_acceptance_retention": null,
75
+ "mtp_distribution": {},
76
+ "mtp_measured_speedup": null,
77
+ "mtp_policy": {
78
+ "candidate_bits": [
79
+ 8,
80
+ 16
81
+ ],
82
+ "min_bits": 8,
83
+ "mode": "protected",
84
+ "optimize_for_acceptance": true,
85
+ "preserve_external_sidecar": true,
86
+ "protect_norms": true,
87
+ "protect_output_head": true
88
+ },
89
+ "mtp_present": false,
90
+ "mtp_weight_file_size_bytes": 0,
91
+ "plan_sha256": "864e1704f89086e16e8e8b124ef72e348aea1b7ad863e8dd1a6af299af74e835",
92
+ "profile": "general",
93
+ "protected_weight_file_size_bytes": 0,
94
+ "quantizer": "axquant",
95
+ "runtime": {
96
+ "ax_engine": {
97
+ "decode_kernel": null,
98
+ "fused_mtp": null,
99
+ "kernel_evidence": "unmeasured",
100
+ "model_manifest": "model-manifest.json",
101
+ "preferred_group_size": 32
102
+ },
103
+ "compatible_runtimes": [
104
+ {
105
+ "compatibility_level": "B",
106
+ "manifest": "config.json",
107
+ "mtp_support": "runtime-dependent",
108
+ "name": "mlx-lm",
109
+ "notes": [
110
+ "Standard backbone inference is the compatibility target.",
111
+ "AXQuant MTP metadata may be ignored by MLX-LM."
112
+ ],
113
+ "standard_inference": true,
114
+ "standard_mlx_weights": true,
115
+ "support_level": "standard-inference"
116
+ }
117
+ ],
118
+ "created_at": "2026-08-02T19:52:51.472949Z",
119
+ "kv_cache": null,
120
+ "memory_policy": {
121
+ "kv_cache_precision": "runtime-default",
122
+ "mtp_buffers": "not-required",
123
+ "prefix_cache": "runtime-managed",
124
+ "unified_memory_safety_margin": "benchmark-required"
125
+ },
126
+ "mtp": {
127
+ "acceptance_retention": null,
128
+ "detected": false,
129
+ "draft_tokens": null,
130
+ "enabled_by_default": false,
131
+ "head_precision": null,
132
+ "measured_speedup": null,
133
+ "optimized": false,
134
+ "recommended_temperature_max": null,
135
+ "sidecar_file": null,
136
+ "verification_mode": null
137
+ },
138
+ "optimization_scope": "text-path",
139
+ "primary_runtime": {
140
+ "compatibility_level": "A",
141
+ "manifest": "model-manifest.json",
142
+ "mtp_support": "native",
143
+ "name": "ax-engine",
144
+ "notes": [
145
+ "Runtime claims require a passing AX Engine doctor and benchmark report."
146
+ ],
147
+ "standard_inference": true,
148
+ "standard_mlx_weights": true,
149
+ "support_level": "optimized"
150
+ },
151
+ "schema_version": "axquant.runtime.v1"
152
+ },
153
+ "schema_version": "axquant.artifact.v2",
154
+ "software_versions": {
155
+ "ax_engine": null,
156
+ "axquant": "1.0.1",
157
+ "mlx": "0.32.0",
158
+ "mlx_lm": "0.31.3",
159
+ "pydantic": "2.13.4",
160
+ "python": "3.13.14",
161
+ "safetensors": "0.8.0"
162
+ },
163
+ "source_model": {
164
+ "architecture": "Mistral3ForConditionalGeneration",
165
+ "format": "mlx",
166
+ "local_path": null,
167
+ "model_id": "mistralai/Ministral-3-14B-Instruct-2512",
168
+ "revision": null
169
+ },
170
+ "target_class": "4bit",
171
+ "weight_distribution": {
172
+ "4bit": {
173
+ "fraction": 0.9005934633733967,
174
+ "parameters": 12163481600
175
+ },
176
+ "8bit": {
177
+ "fraction": 0.049687915220601196,
178
+ "parameters": 671088640
179
+ },
180
+ "bf16": {
181
+ "fraction": 0.04971862140600211,
182
+ "parameters": 671503360
183
+ }
184
+ },
185
+ "weight_file_size_bytes": 8913845132
186
+ }
axquant_plan.json ADDED
The diff for this file is too large to render. See raw diff
 
axquant_quantizer_execution.json ADDED
@@ -0,0 +1,2817 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "created_at": "2026-08-02T19:52:51.464543Z",
3
+ "plan_sha256": "864e1704f89086e16e8e8b124ef72e348aea1b7ad863e8dd1a6af299af74e835",
4
+ "records": [
5
+ {
6
+ "bits": 8,
7
+ "fallback": false,
8
+ "group_size": 64,
9
+ "metadata": {},
10
+ "method": "affine",
11
+ "module_path": "language_model.model.embed_tokens",
12
+ "note": "MLX-LM affine packing",
13
+ "success": true
14
+ },
15
+ {
16
+ "bits": 4,
17
+ "fallback": false,
18
+ "group_size": 64,
19
+ "metadata": {},
20
+ "method": "affine",
21
+ "module_path": "language_model.model.layers.0.mlp.down_proj",
22
+ "note": "MLX-LM affine packing",
23
+ "success": true
24
+ },
25
+ {
26
+ "bits": 4,
27
+ "fallback": false,
28
+ "group_size": 64,
29
+ "metadata": {},
30
+ "method": "affine",
31
+ "module_path": "language_model.model.layers.0.mlp.gate_proj",
32
+ "note": "MLX-LM affine packing",
33
+ "success": true
34
+ },
35
+ {
36
+ "bits": 4,
37
+ "fallback": false,
38
+ "group_size": 64,
39
+ "metadata": {},
40
+ "method": "affine",
41
+ "module_path": "language_model.model.layers.0.mlp.up_proj",
42
+ "note": "MLX-LM affine packing",
43
+ "success": true
44
+ },
45
+ {
46
+ "bits": 4,
47
+ "fallback": false,
48
+ "group_size": 64,
49
+ "metadata": {},
50
+ "method": "affine",
51
+ "module_path": "language_model.model.layers.0.self_attn.k_proj",
52
+ "note": "MLX-LM affine packing",
53
+ "success": true
54
+ },
55
+ {
56
+ "bits": 4,
57
+ "fallback": false,
58
+ "group_size": 64,
59
+ "metadata": {},
60
+ "method": "affine",
61
+ "module_path": "language_model.model.layers.0.self_attn.o_proj",
62
+ "note": "MLX-LM affine packing",
63
+ "success": true
64
+ },
65
+ {
66
+ "bits": 4,
67
+ "fallback": false,
68
+ "group_size": 64,
69
+ "metadata": {},
70
+ "method": "affine",
71
+ "module_path": "language_model.model.layers.0.self_attn.q_proj",
72
+ "note": "MLX-LM affine packing",
73
+ "success": true
74
+ },
75
+ {
76
+ "bits": 4,
77
+ "fallback": false,
78
+ "group_size": 64,
79
+ "metadata": {},
80
+ "method": "affine",
81
+ "module_path": "language_model.model.layers.0.self_attn.v_proj",
82
+ "note": "MLX-LM affine packing",
83
+ "success": true
84
+ },
85
+ {
86
+ "bits": 4,
87
+ "fallback": false,
88
+ "group_size": 64,
89
+ "metadata": {},
90
+ "method": "affine",
91
+ "module_path": "language_model.model.layers.1.mlp.down_proj",
92
+ "note": "MLX-LM affine packing",
93
+ "success": true
94
+ },
95
+ {
96
+ "bits": 4,
97
+ "fallback": false,
98
+ "group_size": 64,
99
+ "metadata": {},
100
+ "method": "affine",
101
+ "module_path": "language_model.model.layers.1.mlp.gate_proj",
102
+ "note": "MLX-LM affine packing",
103
+ "success": true
104
+ },
105
+ {
106
+ "bits": 4,
107
+ "fallback": false,
108
+ "group_size": 64,
109
+ "metadata": {},
110
+ "method": "affine",
111
+ "module_path": "language_model.model.layers.1.mlp.up_proj",
112
+ "note": "MLX-LM affine packing",
113
+ "success": true
114
+ },
115
+ {
116
+ "bits": 4,
117
+ "fallback": false,
118
+ "group_size": 64,
119
+ "metadata": {},
120
+ "method": "affine",
121
+ "module_path": "language_model.model.layers.1.self_attn.k_proj",
122
+ "note": "MLX-LM affine packing",
123
+ "success": true
124
+ },
125
+ {
126
+ "bits": 4,
127
+ "fallback": false,
128
+ "group_size": 64,
129
+ "metadata": {},
130
+ "method": "affine",
131
+ "module_path": "language_model.model.layers.1.self_attn.o_proj",
132
+ "note": "MLX-LM affine packing",
133
+ "success": true
134
+ },
135
+ {
136
+ "bits": 4,
137
+ "fallback": false,
138
+ "group_size": 64,
139
+ "metadata": {},
140
+ "method": "affine",
141
+ "module_path": "language_model.model.layers.1.self_attn.q_proj",
142
+ "note": "MLX-LM affine packing",
143
+ "success": true
144
+ },
145
+ {
146
+ "bits": 4,
147
+ "fallback": false,
148
+ "group_size": 64,
149
+ "metadata": {},
150
+ "method": "affine",
151
+ "module_path": "language_model.model.layers.1.self_attn.v_proj",
152
+ "note": "MLX-LM affine packing",
153
+ "success": true
154
+ },
155
+ {
156
+ "bits": 4,
157
+ "fallback": false,
158
+ "group_size": 64,
159
+ "metadata": {},
160
+ "method": "affine",
161
+ "module_path": "language_model.model.layers.2.mlp.down_proj",
162
+ "note": "MLX-LM affine packing",
163
+ "success": true
164
+ },
165
+ {
166
+ "bits": 4,
167
+ "fallback": false,
168
+ "group_size": 64,
169
+ "metadata": {},
170
+ "method": "affine",
171
+ "module_path": "language_model.model.layers.2.mlp.gate_proj",
172
+ "note": "MLX-LM affine packing",
173
+ "success": true
174
+ },
175
+ {
176
+ "bits": 4,
177
+ "fallback": false,
178
+ "group_size": 64,
179
+ "metadata": {},
180
+ "method": "affine",
181
+ "module_path": "language_model.model.layers.2.mlp.up_proj",
182
+ "note": "MLX-LM affine packing",
183
+ "success": true
184
+ },
185
+ {
186
+ "bits": 4,
187
+ "fallback": false,
188
+ "group_size": 64,
189
+ "metadata": {},
190
+ "method": "affine",
191
+ "module_path": "language_model.model.layers.2.self_attn.k_proj",
192
+ "note": "MLX-LM affine packing",
193
+ "success": true
194
+ },
195
+ {
196
+ "bits": 4,
197
+ "fallback": false,
198
+ "group_size": 64,
199
+ "metadata": {},
200
+ "method": "affine",
201
+ "module_path": "language_model.model.layers.2.self_attn.o_proj",
202
+ "note": "MLX-LM affine packing",
203
+ "success": true
204
+ },
205
+ {
206
+ "bits": 4,
207
+ "fallback": false,
208
+ "group_size": 64,
209
+ "metadata": {},
210
+ "method": "affine",
211
+ "module_path": "language_model.model.layers.2.self_attn.q_proj",
212
+ "note": "MLX-LM affine packing",
213
+ "success": true
214
+ },
215
+ {
216
+ "bits": 4,
217
+ "fallback": false,
218
+ "group_size": 64,
219
+ "metadata": {},
220
+ "method": "affine",
221
+ "module_path": "language_model.model.layers.2.self_attn.v_proj",
222
+ "note": "MLX-LM affine packing",
223
+ "success": true
224
+ },
225
+ {
226
+ "bits": 4,
227
+ "fallback": false,
228
+ "group_size": 64,
229
+ "metadata": {},
230
+ "method": "affine",
231
+ "module_path": "language_model.model.layers.3.mlp.down_proj",
232
+ "note": "MLX-LM affine packing",
233
+ "success": true
234
+ },
235
+ {
236
+ "bits": 4,
237
+ "fallback": false,
238
+ "group_size": 64,
239
+ "metadata": {},
240
+ "method": "affine",
241
+ "module_path": "language_model.model.layers.3.mlp.gate_proj",
242
+ "note": "MLX-LM affine packing",
243
+ "success": true
244
+ },
245
+ {
246
+ "bits": 4,
247
+ "fallback": false,
248
+ "group_size": 64,
249
+ "metadata": {},
250
+ "method": "affine",
251
+ "module_path": "language_model.model.layers.3.mlp.up_proj",
252
+ "note": "MLX-LM affine packing",
253
+ "success": true
254
+ },
255
+ {
256
+ "bits": 4,
257
+ "fallback": false,
258
+ "group_size": 64,
259
+ "metadata": {},
260
+ "method": "affine",
261
+ "module_path": "language_model.model.layers.3.self_attn.k_proj",
262
+ "note": "MLX-LM affine packing",
263
+ "success": true
264
+ },
265
+ {
266
+ "bits": 4,
267
+ "fallback": false,
268
+ "group_size": 64,
269
+ "metadata": {},
270
+ "method": "affine",
271
+ "module_path": "language_model.model.layers.3.self_attn.o_proj",
272
+ "note": "MLX-LM affine packing",
273
+ "success": true
274
+ },
275
+ {
276
+ "bits": 4,
277
+ "fallback": false,
278
+ "group_size": 64,
279
+ "metadata": {},
280
+ "method": "affine",
281
+ "module_path": "language_model.model.layers.3.self_attn.q_proj",
282
+ "note": "MLX-LM affine packing",
283
+ "success": true
284
+ },
285
+ {
286
+ "bits": 4,
287
+ "fallback": false,
288
+ "group_size": 64,
289
+ "metadata": {},
290
+ "method": "affine",
291
+ "module_path": "language_model.model.layers.3.self_attn.v_proj",
292
+ "note": "MLX-LM affine packing",
293
+ "success": true
294
+ },
295
+ {
296
+ "bits": 4,
297
+ "fallback": false,
298
+ "group_size": 64,
299
+ "metadata": {},
300
+ "method": "affine",
301
+ "module_path": "language_model.model.layers.4.mlp.down_proj",
302
+ "note": "MLX-LM affine packing",
303
+ "success": true
304
+ },
305
+ {
306
+ "bits": 4,
307
+ "fallback": false,
308
+ "group_size": 64,
309
+ "metadata": {},
310
+ "method": "affine",
311
+ "module_path": "language_model.model.layers.4.mlp.gate_proj",
312
+ "note": "MLX-LM affine packing",
313
+ "success": true
314
+ },
315
+ {
316
+ "bits": 4,
317
+ "fallback": false,
318
+ "group_size": 64,
319
+ "metadata": {},
320
+ "method": "affine",
321
+ "module_path": "language_model.model.layers.4.mlp.up_proj",
322
+ "note": "MLX-LM affine packing",
323
+ "success": true
324
+ },
325
+ {
326
+ "bits": 4,
327
+ "fallback": false,
328
+ "group_size": 64,
329
+ "metadata": {},
330
+ "method": "affine",
331
+ "module_path": "language_model.model.layers.4.self_attn.k_proj",
332
+ "note": "MLX-LM affine packing",
333
+ "success": true
334
+ },
335
+ {
336
+ "bits": 4,
337
+ "fallback": false,
338
+ "group_size": 64,
339
+ "metadata": {},
340
+ "method": "affine",
341
+ "module_path": "language_model.model.layers.4.self_attn.o_proj",
342
+ "note": "MLX-LM affine packing",
343
+ "success": true
344
+ },
345
+ {
346
+ "bits": 4,
347
+ "fallback": false,
348
+ "group_size": 64,
349
+ "metadata": {},
350
+ "method": "affine",
351
+ "module_path": "language_model.model.layers.4.self_attn.q_proj",
352
+ "note": "MLX-LM affine packing",
353
+ "success": true
354
+ },
355
+ {
356
+ "bits": 4,
357
+ "fallback": false,
358
+ "group_size": 64,
359
+ "metadata": {},
360
+ "method": "affine",
361
+ "module_path": "language_model.model.layers.4.self_attn.v_proj",
362
+ "note": "MLX-LM affine packing",
363
+ "success": true
364
+ },
365
+ {
366
+ "bits": 4,
367
+ "fallback": false,
368
+ "group_size": 64,
369
+ "metadata": {},
370
+ "method": "affine",
371
+ "module_path": "language_model.model.layers.5.mlp.down_proj",
372
+ "note": "MLX-LM affine packing",
373
+ "success": true
374
+ },
375
+ {
376
+ "bits": 4,
377
+ "fallback": false,
378
+ "group_size": 64,
379
+ "metadata": {},
380
+ "method": "affine",
381
+ "module_path": "language_model.model.layers.5.mlp.gate_proj",
382
+ "note": "MLX-LM affine packing",
383
+ "success": true
384
+ },
385
+ {
386
+ "bits": 4,
387
+ "fallback": false,
388
+ "group_size": 64,
389
+ "metadata": {},
390
+ "method": "affine",
391
+ "module_path": "language_model.model.layers.5.mlp.up_proj",
392
+ "note": "MLX-LM affine packing",
393
+ "success": true
394
+ },
395
+ {
396
+ "bits": 4,
397
+ "fallback": false,
398
+ "group_size": 64,
399
+ "metadata": {},
400
+ "method": "affine",
401
+ "module_path": "language_model.model.layers.5.self_attn.k_proj",
402
+ "note": "MLX-LM affine packing",
403
+ "success": true
404
+ },
405
+ {
406
+ "bits": 4,
407
+ "fallback": false,
408
+ "group_size": 64,
409
+ "metadata": {},
410
+ "method": "affine",
411
+ "module_path": "language_model.model.layers.5.self_attn.o_proj",
412
+ "note": "MLX-LM affine packing",
413
+ "success": true
414
+ },
415
+ {
416
+ "bits": 4,
417
+ "fallback": false,
418
+ "group_size": 64,
419
+ "metadata": {},
420
+ "method": "affine",
421
+ "module_path": "language_model.model.layers.5.self_attn.q_proj",
422
+ "note": "MLX-LM affine packing",
423
+ "success": true
424
+ },
425
+ {
426
+ "bits": 4,
427
+ "fallback": false,
428
+ "group_size": 64,
429
+ "metadata": {},
430
+ "method": "affine",
431
+ "module_path": "language_model.model.layers.5.self_attn.v_proj",
432
+ "note": "MLX-LM affine packing",
433
+ "success": true
434
+ },
435
+ {
436
+ "bits": 4,
437
+ "fallback": false,
438
+ "group_size": 64,
439
+ "metadata": {},
440
+ "method": "affine",
441
+ "module_path": "language_model.model.layers.6.mlp.gate_proj",
442
+ "note": "MLX-LM affine packing",
443
+ "success": true
444
+ },
445
+ {
446
+ "bits": 4,
447
+ "fallback": false,
448
+ "group_size": 64,
449
+ "metadata": {},
450
+ "method": "affine",
451
+ "module_path": "language_model.model.layers.6.self_attn.k_proj",
452
+ "note": "MLX-LM affine packing",
453
+ "success": true
454
+ },
455
+ {
456
+ "bits": 4,
457
+ "fallback": false,
458
+ "group_size": 64,
459
+ "metadata": {},
460
+ "method": "affine",
461
+ "module_path": "language_model.model.layers.6.self_attn.o_proj",
462
+ "note": "MLX-LM affine packing",
463
+ "success": true
464
+ },
465
+ {
466
+ "bits": 4,
467
+ "fallback": false,
468
+ "group_size": 64,
469
+ "metadata": {},
470
+ "method": "affine",
471
+ "module_path": "language_model.model.layers.6.self_attn.q_proj",
472
+ "note": "MLX-LM affine packing",
473
+ "success": true
474
+ },
475
+ {
476
+ "bits": 4,
477
+ "fallback": false,
478
+ "group_size": 64,
479
+ "metadata": {},
480
+ "method": "affine",
481
+ "module_path": "language_model.model.layers.6.self_attn.v_proj",
482
+ "note": "MLX-LM affine packing",
483
+ "success": true
484
+ },
485
+ {
486
+ "bits": 4,
487
+ "fallback": false,
488
+ "group_size": 64,
489
+ "metadata": {},
490
+ "method": "affine",
491
+ "module_path": "language_model.model.layers.10.mlp.down_proj",
492
+ "note": "MLX-LM affine packing",
493
+ "success": true
494
+ },
495
+ {
496
+ "bits": 4,
497
+ "fallback": false,
498
+ "group_size": 64,
499
+ "metadata": {},
500
+ "method": "affine",
501
+ "module_path": "language_model.model.layers.10.mlp.gate_proj",
502
+ "note": "MLX-LM affine packing",
503
+ "success": true
504
+ },
505
+ {
506
+ "bits": 4,
507
+ "fallback": false,
508
+ "group_size": 64,
509
+ "metadata": {},
510
+ "method": "affine",
511
+ "module_path": "language_model.model.layers.10.mlp.up_proj",
512
+ "note": "MLX-LM affine packing",
513
+ "success": true
514
+ },
515
+ {
516
+ "bits": 4,
517
+ "fallback": false,
518
+ "group_size": 64,
519
+ "metadata": {},
520
+ "method": "affine",
521
+ "module_path": "language_model.model.layers.10.self_attn.k_proj",
522
+ "note": "MLX-LM affine packing",
523
+ "success": true
524
+ },
525
+ {
526
+ "bits": 4,
527
+ "fallback": false,
528
+ "group_size": 64,
529
+ "metadata": {},
530
+ "method": "affine",
531
+ "module_path": "language_model.model.layers.10.self_attn.o_proj",
532
+ "note": "MLX-LM affine packing",
533
+ "success": true
534
+ },
535
+ {
536
+ "bits": 4,
537
+ "fallback": false,
538
+ "group_size": 64,
539
+ "metadata": {},
540
+ "method": "affine",
541
+ "module_path": "language_model.model.layers.10.self_attn.q_proj",
542
+ "note": "MLX-LM affine packing",
543
+ "success": true
544
+ },
545
+ {
546
+ "bits": 4,
547
+ "fallback": false,
548
+ "group_size": 64,
549
+ "metadata": {},
550
+ "method": "affine",
551
+ "module_path": "language_model.model.layers.10.self_attn.v_proj",
552
+ "note": "MLX-LM affine packing",
553
+ "success": true
554
+ },
555
+ {
556
+ "bits": 4,
557
+ "fallback": false,
558
+ "group_size": 64,
559
+ "metadata": {},
560
+ "method": "affine",
561
+ "module_path": "language_model.model.layers.11.mlp.down_proj",
562
+ "note": "MLX-LM affine packing",
563
+ "success": true
564
+ },
565
+ {
566
+ "bits": 4,
567
+ "fallback": false,
568
+ "group_size": 64,
569
+ "metadata": {},
570
+ "method": "affine",
571
+ "module_path": "language_model.model.layers.11.mlp.gate_proj",
572
+ "note": "MLX-LM affine packing",
573
+ "success": true
574
+ },
575
+ {
576
+ "bits": 4,
577
+ "fallback": false,
578
+ "group_size": 64,
579
+ "metadata": {},
580
+ "method": "affine",
581
+ "module_path": "language_model.model.layers.11.mlp.up_proj",
582
+ "note": "MLX-LM affine packing",
583
+ "success": true
584
+ },
585
+ {
586
+ "bits": 4,
587
+ "fallback": false,
588
+ "group_size": 64,
589
+ "metadata": {},
590
+ "method": "affine",
591
+ "module_path": "language_model.model.layers.11.self_attn.k_proj",
592
+ "note": "MLX-LM affine packing",
593
+ "success": true
594
+ },
595
+ {
596
+ "bits": 4,
597
+ "fallback": false,
598
+ "group_size": 64,
599
+ "metadata": {},
600
+ "method": "affine",
601
+ "module_path": "language_model.model.layers.11.self_attn.o_proj",
602
+ "note": "MLX-LM affine packing",
603
+ "success": true
604
+ },
605
+ {
606
+ "bits": 4,
607
+ "fallback": false,
608
+ "group_size": 64,
609
+ "metadata": {},
610
+ "method": "affine",
611
+ "module_path": "language_model.model.layers.11.self_attn.q_proj",
612
+ "note": "MLX-LM affine packing",
613
+ "success": true
614
+ },
615
+ {
616
+ "bits": 4,
617
+ "fallback": false,
618
+ "group_size": 64,
619
+ "metadata": {},
620
+ "method": "affine",
621
+ "module_path": "language_model.model.layers.11.self_attn.v_proj",
622
+ "note": "MLX-LM affine packing",
623
+ "success": true
624
+ },
625
+ {
626
+ "bits": 4,
627
+ "fallback": false,
628
+ "group_size": 64,
629
+ "metadata": {},
630
+ "method": "affine",
631
+ "module_path": "language_model.model.layers.12.mlp.down_proj",
632
+ "note": "MLX-LM affine packing",
633
+ "success": true
634
+ },
635
+ {
636
+ "bits": 4,
637
+ "fallback": false,
638
+ "group_size": 64,
639
+ "metadata": {},
640
+ "method": "affine",
641
+ "module_path": "language_model.model.layers.12.mlp.gate_proj",
642
+ "note": "MLX-LM affine packing",
643
+ "success": true
644
+ },
645
+ {
646
+ "bits": 4,
647
+ "fallback": false,
648
+ "group_size": 64,
649
+ "metadata": {},
650
+ "method": "affine",
651
+ "module_path": "language_model.model.layers.12.mlp.up_proj",
652
+ "note": "MLX-LM affine packing",
653
+ "success": true
654
+ },
655
+ {
656
+ "bits": 4,
657
+ "fallback": false,
658
+ "group_size": 64,
659
+ "metadata": {},
660
+ "method": "affine",
661
+ "module_path": "language_model.model.layers.12.self_attn.k_proj",
662
+ "note": "MLX-LM affine packing",
663
+ "success": true
664
+ },
665
+ {
666
+ "bits": 4,
667
+ "fallback": false,
668
+ "group_size": 64,
669
+ "metadata": {},
670
+ "method": "affine",
671
+ "module_path": "language_model.model.layers.12.self_attn.o_proj",
672
+ "note": "MLX-LM affine packing",
673
+ "success": true
674
+ },
675
+ {
676
+ "bits": 4,
677
+ "fallback": false,
678
+ "group_size": 64,
679
+ "metadata": {},
680
+ "method": "affine",
681
+ "module_path": "language_model.model.layers.12.self_attn.q_proj",
682
+ "note": "MLX-LM affine packing",
683
+ "success": true
684
+ },
685
+ {
686
+ "bits": 4,
687
+ "fallback": false,
688
+ "group_size": 64,
689
+ "metadata": {},
690
+ "method": "affine",
691
+ "module_path": "language_model.model.layers.12.self_attn.v_proj",
692
+ "note": "MLX-LM affine packing",
693
+ "success": true
694
+ },
695
+ {
696
+ "bits": 4,
697
+ "fallback": false,
698
+ "group_size": 64,
699
+ "metadata": {},
700
+ "method": "affine",
701
+ "module_path": "language_model.model.layers.13.mlp.down_proj",
702
+ "note": "MLX-LM affine packing",
703
+ "success": true
704
+ },
705
+ {
706
+ "bits": 4,
707
+ "fallback": false,
708
+ "group_size": 64,
709
+ "metadata": {},
710
+ "method": "affine",
711
+ "module_path": "language_model.model.layers.13.mlp.gate_proj",
712
+ "note": "MLX-LM affine packing",
713
+ "success": true
714
+ },
715
+ {
716
+ "bits": 4,
717
+ "fallback": false,
718
+ "group_size": 64,
719
+ "metadata": {},
720
+ "method": "affine",
721
+ "module_path": "language_model.model.layers.13.mlp.up_proj",
722
+ "note": "MLX-LM affine packing",
723
+ "success": true
724
+ },
725
+ {
726
+ "bits": 4,
727
+ "fallback": false,
728
+ "group_size": 64,
729
+ "metadata": {},
730
+ "method": "affine",
731
+ "module_path": "language_model.model.layers.13.self_attn.k_proj",
732
+ "note": "MLX-LM affine packing",
733
+ "success": true
734
+ },
735
+ {
736
+ "bits": 4,
737
+ "fallback": false,
738
+ "group_size": 64,
739
+ "metadata": {},
740
+ "method": "affine",
741
+ "module_path": "language_model.model.layers.13.self_attn.o_proj",
742
+ "note": "MLX-LM affine packing",
743
+ "success": true
744
+ },
745
+ {
746
+ "bits": 4,
747
+ "fallback": false,
748
+ "group_size": 64,
749
+ "metadata": {},
750
+ "method": "affine",
751
+ "module_path": "language_model.model.layers.13.self_attn.q_proj",
752
+ "note": "MLX-LM affine packing",
753
+ "success": true
754
+ },
755
+ {
756
+ "bits": 4,
757
+ "fallback": false,
758
+ "group_size": 64,
759
+ "metadata": {},
760
+ "method": "affine",
761
+ "module_path": "language_model.model.layers.13.self_attn.v_proj",
762
+ "note": "MLX-LM affine packing",
763
+ "success": true
764
+ },
765
+ {
766
+ "bits": 4,
767
+ "fallback": false,
768
+ "group_size": 64,
769
+ "metadata": {},
770
+ "method": "affine",
771
+ "module_path": "language_model.model.layers.14.mlp.down_proj",
772
+ "note": "MLX-LM affine packing",
773
+ "success": true
774
+ },
775
+ {
776
+ "bits": 4,
777
+ "fallback": false,
778
+ "group_size": 64,
779
+ "metadata": {},
780
+ "method": "affine",
781
+ "module_path": "language_model.model.layers.14.mlp.gate_proj",
782
+ "note": "MLX-LM affine packing",
783
+ "success": true
784
+ },
785
+ {
786
+ "bits": 4,
787
+ "fallback": false,
788
+ "group_size": 64,
789
+ "metadata": {},
790
+ "method": "affine",
791
+ "module_path": "language_model.model.layers.14.mlp.up_proj",
792
+ "note": "MLX-LM affine packing",
793
+ "success": true
794
+ },
795
+ {
796
+ "bits": 4,
797
+ "fallback": false,
798
+ "group_size": 64,
799
+ "metadata": {},
800
+ "method": "affine",
801
+ "module_path": "language_model.model.layers.14.self_attn.k_proj",
802
+ "note": "MLX-LM affine packing",
803
+ "success": true
804
+ },
805
+ {
806
+ "bits": 4,
807
+ "fallback": false,
808
+ "group_size": 64,
809
+ "metadata": {},
810
+ "method": "affine",
811
+ "module_path": "language_model.model.layers.14.self_attn.o_proj",
812
+ "note": "MLX-LM affine packing",
813
+ "success": true
814
+ },
815
+ {
816
+ "bits": 4,
817
+ "fallback": false,
818
+ "group_size": 64,
819
+ "metadata": {},
820
+ "method": "affine",
821
+ "module_path": "language_model.model.layers.14.self_attn.q_proj",
822
+ "note": "MLX-LM affine packing",
823
+ "success": true
824
+ },
825
+ {
826
+ "bits": 4,
827
+ "fallback": false,
828
+ "group_size": 64,
829
+ "metadata": {},
830
+ "method": "affine",
831
+ "module_path": "language_model.model.layers.14.self_attn.v_proj",
832
+ "note": "MLX-LM affine packing",
833
+ "success": true
834
+ },
835
+ {
836
+ "bits": 4,
837
+ "fallback": false,
838
+ "group_size": 64,
839
+ "metadata": {},
840
+ "method": "affine",
841
+ "module_path": "language_model.model.layers.15.self_attn.k_proj",
842
+ "note": "MLX-LM affine packing",
843
+ "success": true
844
+ },
845
+ {
846
+ "bits": 4,
847
+ "fallback": false,
848
+ "group_size": 64,
849
+ "metadata": {},
850
+ "method": "affine",
851
+ "module_path": "language_model.model.layers.15.self_attn.o_proj",
852
+ "note": "MLX-LM affine packing",
853
+ "success": true
854
+ },
855
+ {
856
+ "bits": 4,
857
+ "fallback": false,
858
+ "group_size": 64,
859
+ "metadata": {},
860
+ "method": "affine",
861
+ "module_path": "language_model.model.layers.15.self_attn.q_proj",
862
+ "note": "MLX-LM affine packing",
863
+ "success": true
864
+ },
865
+ {
866
+ "bits": 4,
867
+ "fallback": false,
868
+ "group_size": 64,
869
+ "metadata": {},
870
+ "method": "affine",
871
+ "module_path": "language_model.model.layers.15.self_attn.v_proj",
872
+ "note": "MLX-LM affine packing",
873
+ "success": true
874
+ },
875
+ {
876
+ "bits": 4,
877
+ "fallback": false,
878
+ "group_size": 64,
879
+ "metadata": {},
880
+ "method": "affine",
881
+ "module_path": "language_model.model.layers.6.mlp.down_proj",
882
+ "note": "MLX-LM affine packing",
883
+ "success": true
884
+ },
885
+ {
886
+ "bits": 4,
887
+ "fallback": false,
888
+ "group_size": 64,
889
+ "metadata": {},
890
+ "method": "affine",
891
+ "module_path": "language_model.model.layers.6.mlp.up_proj",
892
+ "note": "MLX-LM affine packing",
893
+ "success": true
894
+ },
895
+ {
896
+ "bits": 4,
897
+ "fallback": false,
898
+ "group_size": 64,
899
+ "metadata": {},
900
+ "method": "affine",
901
+ "module_path": "language_model.model.layers.7.mlp.down_proj",
902
+ "note": "MLX-LM affine packing",
903
+ "success": true
904
+ },
905
+ {
906
+ "bits": 4,
907
+ "fallback": false,
908
+ "group_size": 64,
909
+ "metadata": {},
910
+ "method": "affine",
911
+ "module_path": "language_model.model.layers.7.mlp.gate_proj",
912
+ "note": "MLX-LM affine packing",
913
+ "success": true
914
+ },
915
+ {
916
+ "bits": 4,
917
+ "fallback": false,
918
+ "group_size": 64,
919
+ "metadata": {},
920
+ "method": "affine",
921
+ "module_path": "language_model.model.layers.7.mlp.up_proj",
922
+ "note": "MLX-LM affine packing",
923
+ "success": true
924
+ },
925
+ {
926
+ "bits": 4,
927
+ "fallback": false,
928
+ "group_size": 64,
929
+ "metadata": {},
930
+ "method": "affine",
931
+ "module_path": "language_model.model.layers.7.self_attn.k_proj",
932
+ "note": "MLX-LM affine packing",
933
+ "success": true
934
+ },
935
+ {
936
+ "bits": 4,
937
+ "fallback": false,
938
+ "group_size": 64,
939
+ "metadata": {},
940
+ "method": "affine",
941
+ "module_path": "language_model.model.layers.7.self_attn.o_proj",
942
+ "note": "MLX-LM affine packing",
943
+ "success": true
944
+ },
945
+ {
946
+ "bits": 4,
947
+ "fallback": false,
948
+ "group_size": 64,
949
+ "metadata": {},
950
+ "method": "affine",
951
+ "module_path": "language_model.model.layers.7.self_attn.q_proj",
952
+ "note": "MLX-LM affine packing",
953
+ "success": true
954
+ },
955
+ {
956
+ "bits": 4,
957
+ "fallback": false,
958
+ "group_size": 64,
959
+ "metadata": {},
960
+ "method": "affine",
961
+ "module_path": "language_model.model.layers.7.self_attn.v_proj",
962
+ "note": "MLX-LM affine packing",
963
+ "success": true
964
+ },
965
+ {
966
+ "bits": 4,
967
+ "fallback": false,
968
+ "group_size": 64,
969
+ "metadata": {},
970
+ "method": "affine",
971
+ "module_path": "language_model.model.layers.8.mlp.down_proj",
972
+ "note": "MLX-LM affine packing",
973
+ "success": true
974
+ },
975
+ {
976
+ "bits": 4,
977
+ "fallback": false,
978
+ "group_size": 64,
979
+ "metadata": {},
980
+ "method": "affine",
981
+ "module_path": "language_model.model.layers.8.mlp.gate_proj",
982
+ "note": "MLX-LM affine packing",
983
+ "success": true
984
+ },
985
+ {
986
+ "bits": 4,
987
+ "fallback": false,
988
+ "group_size": 64,
989
+ "metadata": {},
990
+ "method": "affine",
991
+ "module_path": "language_model.model.layers.8.mlp.up_proj",
992
+ "note": "MLX-LM affine packing",
993
+ "success": true
994
+ },
995
+ {
996
+ "bits": 4,
997
+ "fallback": false,
998
+ "group_size": 64,
999
+ "metadata": {},
1000
+ "method": "affine",
1001
+ "module_path": "language_model.model.layers.8.self_attn.k_proj",
1002
+ "note": "MLX-LM affine packing",
1003
+ "success": true
1004
+ },
1005
+ {
1006
+ "bits": 4,
1007
+ "fallback": false,
1008
+ "group_size": 64,
1009
+ "metadata": {},
1010
+ "method": "affine",
1011
+ "module_path": "language_model.model.layers.8.self_attn.o_proj",
1012
+ "note": "MLX-LM affine packing",
1013
+ "success": true
1014
+ },
1015
+ {
1016
+ "bits": 4,
1017
+ "fallback": false,
1018
+ "group_size": 64,
1019
+ "metadata": {},
1020
+ "method": "affine",
1021
+ "module_path": "language_model.model.layers.8.self_attn.q_proj",
1022
+ "note": "MLX-LM affine packing",
1023
+ "success": true
1024
+ },
1025
+ {
1026
+ "bits": 4,
1027
+ "fallback": false,
1028
+ "group_size": 64,
1029
+ "metadata": {},
1030
+ "method": "affine",
1031
+ "module_path": "language_model.model.layers.8.self_attn.v_proj",
1032
+ "note": "MLX-LM affine packing",
1033
+ "success": true
1034
+ },
1035
+ {
1036
+ "bits": 4,
1037
+ "fallback": false,
1038
+ "group_size": 64,
1039
+ "metadata": {},
1040
+ "method": "affine",
1041
+ "module_path": "language_model.model.layers.9.mlp.down_proj",
1042
+ "note": "MLX-LM affine packing",
1043
+ "success": true
1044
+ },
1045
+ {
1046
+ "bits": 4,
1047
+ "fallback": false,
1048
+ "group_size": 64,
1049
+ "metadata": {},
1050
+ "method": "affine",
1051
+ "module_path": "language_model.model.layers.9.mlp.gate_proj",
1052
+ "note": "MLX-LM affine packing",
1053
+ "success": true
1054
+ },
1055
+ {
1056
+ "bits": 4,
1057
+ "fallback": false,
1058
+ "group_size": 64,
1059
+ "metadata": {},
1060
+ "method": "affine",
1061
+ "module_path": "language_model.model.layers.9.mlp.up_proj",
1062
+ "note": "MLX-LM affine packing",
1063
+ "success": true
1064
+ },
1065
+ {
1066
+ "bits": 4,
1067
+ "fallback": false,
1068
+ "group_size": 64,
1069
+ "metadata": {},
1070
+ "method": "affine",
1071
+ "module_path": "language_model.model.layers.9.self_attn.k_proj",
1072
+ "note": "MLX-LM affine packing",
1073
+ "success": true
1074
+ },
1075
+ {
1076
+ "bits": 4,
1077
+ "fallback": false,
1078
+ "group_size": 64,
1079
+ "metadata": {},
1080
+ "method": "affine",
1081
+ "module_path": "language_model.model.layers.9.self_attn.o_proj",
1082
+ "note": "MLX-LM affine packing",
1083
+ "success": true
1084
+ },
1085
+ {
1086
+ "bits": 4,
1087
+ "fallback": false,
1088
+ "group_size": 64,
1089
+ "metadata": {},
1090
+ "method": "affine",
1091
+ "module_path": "language_model.model.layers.9.self_attn.q_proj",
1092
+ "note": "MLX-LM affine packing",
1093
+ "success": true
1094
+ },
1095
+ {
1096
+ "bits": 4,
1097
+ "fallback": false,
1098
+ "group_size": 64,
1099
+ "metadata": {},
1100
+ "method": "affine",
1101
+ "module_path": "language_model.model.layers.9.self_attn.v_proj",
1102
+ "note": "MLX-LM affine packing",
1103
+ "success": true
1104
+ },
1105
+ {
1106
+ "bits": 4,
1107
+ "fallback": false,
1108
+ "group_size": 64,
1109
+ "metadata": {},
1110
+ "method": "affine",
1111
+ "module_path": "language_model.model.layers.15.mlp.down_proj",
1112
+ "note": "MLX-LM affine packing",
1113
+ "success": true
1114
+ },
1115
+ {
1116
+ "bits": 4,
1117
+ "fallback": false,
1118
+ "group_size": 64,
1119
+ "metadata": {},
1120
+ "method": "affine",
1121
+ "module_path": "language_model.model.layers.15.mlp.gate_proj",
1122
+ "note": "MLX-LM affine packing",
1123
+ "success": true
1124
+ },
1125
+ {
1126
+ "bits": 4,
1127
+ "fallback": false,
1128
+ "group_size": 64,
1129
+ "metadata": {},
1130
+ "method": "affine",
1131
+ "module_path": "language_model.model.layers.15.mlp.up_proj",
1132
+ "note": "MLX-LM affine packing",
1133
+ "success": true
1134
+ },
1135
+ {
1136
+ "bits": 4,
1137
+ "fallback": false,
1138
+ "group_size": 64,
1139
+ "metadata": {},
1140
+ "method": "affine",
1141
+ "module_path": "language_model.model.layers.16.mlp.down_proj",
1142
+ "note": "MLX-LM affine packing",
1143
+ "success": true
1144
+ },
1145
+ {
1146
+ "bits": 4,
1147
+ "fallback": false,
1148
+ "group_size": 64,
1149
+ "metadata": {},
1150
+ "method": "affine",
1151
+ "module_path": "language_model.model.layers.16.mlp.gate_proj",
1152
+ "note": "MLX-LM affine packing",
1153
+ "success": true
1154
+ },
1155
+ {
1156
+ "bits": 4,
1157
+ "fallback": false,
1158
+ "group_size": 64,
1159
+ "metadata": {},
1160
+ "method": "affine",
1161
+ "module_path": "language_model.model.layers.16.mlp.up_proj",
1162
+ "note": "MLX-LM affine packing",
1163
+ "success": true
1164
+ },
1165
+ {
1166
+ "bits": 4,
1167
+ "fallback": false,
1168
+ "group_size": 64,
1169
+ "metadata": {},
1170
+ "method": "affine",
1171
+ "module_path": "language_model.model.layers.16.self_attn.k_proj",
1172
+ "note": "MLX-LM affine packing",
1173
+ "success": true
1174
+ },
1175
+ {
1176
+ "bits": 4,
1177
+ "fallback": false,
1178
+ "group_size": 64,
1179
+ "metadata": {},
1180
+ "method": "affine",
1181
+ "module_path": "language_model.model.layers.16.self_attn.o_proj",
1182
+ "note": "MLX-LM affine packing",
1183
+ "success": true
1184
+ },
1185
+ {
1186
+ "bits": 4,
1187
+ "fallback": false,
1188
+ "group_size": 64,
1189
+ "metadata": {},
1190
+ "method": "affine",
1191
+ "module_path": "language_model.model.layers.16.self_attn.q_proj",
1192
+ "note": "MLX-LM affine packing",
1193
+ "success": true
1194
+ },
1195
+ {
1196
+ "bits": 4,
1197
+ "fallback": false,
1198
+ "group_size": 64,
1199
+ "metadata": {},
1200
+ "method": "affine",
1201
+ "module_path": "language_model.model.layers.16.self_attn.v_proj",
1202
+ "note": "MLX-LM affine packing",
1203
+ "success": true
1204
+ },
1205
+ {
1206
+ "bits": 4,
1207
+ "fallback": false,
1208
+ "group_size": 64,
1209
+ "metadata": {},
1210
+ "method": "affine",
1211
+ "module_path": "language_model.model.layers.17.mlp.down_proj",
1212
+ "note": "MLX-LM affine packing",
1213
+ "success": true
1214
+ },
1215
+ {
1216
+ "bits": 4,
1217
+ "fallback": false,
1218
+ "group_size": 64,
1219
+ "metadata": {},
1220
+ "method": "affine",
1221
+ "module_path": "language_model.model.layers.17.mlp.gate_proj",
1222
+ "note": "MLX-LM affine packing",
1223
+ "success": true
1224
+ },
1225
+ {
1226
+ "bits": 4,
1227
+ "fallback": false,
1228
+ "group_size": 64,
1229
+ "metadata": {},
1230
+ "method": "affine",
1231
+ "module_path": "language_model.model.layers.17.mlp.up_proj",
1232
+ "note": "MLX-LM affine packing",
1233
+ "success": true
1234
+ },
1235
+ {
1236
+ "bits": 4,
1237
+ "fallback": false,
1238
+ "group_size": 64,
1239
+ "metadata": {},
1240
+ "method": "affine",
1241
+ "module_path": "language_model.model.layers.17.self_attn.k_proj",
1242
+ "note": "MLX-LM affine packing",
1243
+ "success": true
1244
+ },
1245
+ {
1246
+ "bits": 4,
1247
+ "fallback": false,
1248
+ "group_size": 64,
1249
+ "metadata": {},
1250
+ "method": "affine",
1251
+ "module_path": "language_model.model.layers.17.self_attn.o_proj",
1252
+ "note": "MLX-LM affine packing",
1253
+ "success": true
1254
+ },
1255
+ {
1256
+ "bits": 4,
1257
+ "fallback": false,
1258
+ "group_size": 64,
1259
+ "metadata": {},
1260
+ "method": "affine",
1261
+ "module_path": "language_model.model.layers.17.self_attn.q_proj",
1262
+ "note": "MLX-LM affine packing",
1263
+ "success": true
1264
+ },
1265
+ {
1266
+ "bits": 4,
1267
+ "fallback": false,
1268
+ "group_size": 64,
1269
+ "metadata": {},
1270
+ "method": "affine",
1271
+ "module_path": "language_model.model.layers.17.self_attn.v_proj",
1272
+ "note": "MLX-LM affine packing",
1273
+ "success": true
1274
+ },
1275
+ {
1276
+ "bits": 4,
1277
+ "fallback": false,
1278
+ "group_size": 64,
1279
+ "metadata": {},
1280
+ "method": "affine",
1281
+ "module_path": "language_model.model.layers.18.mlp.down_proj",
1282
+ "note": "MLX-LM affine packing",
1283
+ "success": true
1284
+ },
1285
+ {
1286
+ "bits": 4,
1287
+ "fallback": false,
1288
+ "group_size": 64,
1289
+ "metadata": {},
1290
+ "method": "affine",
1291
+ "module_path": "language_model.model.layers.18.mlp.gate_proj",
1292
+ "note": "MLX-LM affine packing",
1293
+ "success": true
1294
+ },
1295
+ {
1296
+ "bits": 4,
1297
+ "fallback": false,
1298
+ "group_size": 64,
1299
+ "metadata": {},
1300
+ "method": "affine",
1301
+ "module_path": "language_model.model.layers.18.mlp.up_proj",
1302
+ "note": "MLX-LM affine packing",
1303
+ "success": true
1304
+ },
1305
+ {
1306
+ "bits": 4,
1307
+ "fallback": false,
1308
+ "group_size": 64,
1309
+ "metadata": {},
1310
+ "method": "affine",
1311
+ "module_path": "language_model.model.layers.18.self_attn.k_proj",
1312
+ "note": "MLX-LM affine packing",
1313
+ "success": true
1314
+ },
1315
+ {
1316
+ "bits": 4,
1317
+ "fallback": false,
1318
+ "group_size": 64,
1319
+ "metadata": {},
1320
+ "method": "affine",
1321
+ "module_path": "language_model.model.layers.18.self_attn.o_proj",
1322
+ "note": "MLX-LM affine packing",
1323
+ "success": true
1324
+ },
1325
+ {
1326
+ "bits": 4,
1327
+ "fallback": false,
1328
+ "group_size": 64,
1329
+ "metadata": {},
1330
+ "method": "affine",
1331
+ "module_path": "language_model.model.layers.18.self_attn.q_proj",
1332
+ "note": "MLX-LM affine packing",
1333
+ "success": true
1334
+ },
1335
+ {
1336
+ "bits": 4,
1337
+ "fallback": false,
1338
+ "group_size": 64,
1339
+ "metadata": {},
1340
+ "method": "affine",
1341
+ "module_path": "language_model.model.layers.18.self_attn.v_proj",
1342
+ "note": "MLX-LM affine packing",
1343
+ "success": true
1344
+ },
1345
+ {
1346
+ "bits": 4,
1347
+ "fallback": false,
1348
+ "group_size": 64,
1349
+ "metadata": {},
1350
+ "method": "affine",
1351
+ "module_path": "language_model.model.layers.19.mlp.down_proj",
1352
+ "note": "MLX-LM affine packing",
1353
+ "success": true
1354
+ },
1355
+ {
1356
+ "bits": 4,
1357
+ "fallback": false,
1358
+ "group_size": 64,
1359
+ "metadata": {},
1360
+ "method": "affine",
1361
+ "module_path": "language_model.model.layers.19.mlp.gate_proj",
1362
+ "note": "MLX-LM affine packing",
1363
+ "success": true
1364
+ },
1365
+ {
1366
+ "bits": 4,
1367
+ "fallback": false,
1368
+ "group_size": 64,
1369
+ "metadata": {},
1370
+ "method": "affine",
1371
+ "module_path": "language_model.model.layers.19.mlp.up_proj",
1372
+ "note": "MLX-LM affine packing",
1373
+ "success": true
1374
+ },
1375
+ {
1376
+ "bits": 4,
1377
+ "fallback": false,
1378
+ "group_size": 64,
1379
+ "metadata": {},
1380
+ "method": "affine",
1381
+ "module_path": "language_model.model.layers.19.self_attn.k_proj",
1382
+ "note": "MLX-LM affine packing",
1383
+ "success": true
1384
+ },
1385
+ {
1386
+ "bits": 4,
1387
+ "fallback": false,
1388
+ "group_size": 64,
1389
+ "metadata": {},
1390
+ "method": "affine",
1391
+ "module_path": "language_model.model.layers.19.self_attn.o_proj",
1392
+ "note": "MLX-LM affine packing",
1393
+ "success": true
1394
+ },
1395
+ {
1396
+ "bits": 4,
1397
+ "fallback": false,
1398
+ "group_size": 64,
1399
+ "metadata": {},
1400
+ "method": "affine",
1401
+ "module_path": "language_model.model.layers.19.self_attn.q_proj",
1402
+ "note": "MLX-LM affine packing",
1403
+ "success": true
1404
+ },
1405
+ {
1406
+ "bits": 4,
1407
+ "fallback": false,
1408
+ "group_size": 64,
1409
+ "metadata": {},
1410
+ "method": "affine",
1411
+ "module_path": "language_model.model.layers.19.self_attn.v_proj",
1412
+ "note": "MLX-LM affine packing",
1413
+ "success": true
1414
+ },
1415
+ {
1416
+ "bits": 4,
1417
+ "fallback": false,
1418
+ "group_size": 64,
1419
+ "metadata": {},
1420
+ "method": "affine",
1421
+ "module_path": "language_model.model.layers.20.mlp.down_proj",
1422
+ "note": "MLX-LM affine packing",
1423
+ "success": true
1424
+ },
1425
+ {
1426
+ "bits": 4,
1427
+ "fallback": false,
1428
+ "group_size": 64,
1429
+ "metadata": {},
1430
+ "method": "affine",
1431
+ "module_path": "language_model.model.layers.20.mlp.gate_proj",
1432
+ "note": "MLX-LM affine packing",
1433
+ "success": true
1434
+ },
1435
+ {
1436
+ "bits": 4,
1437
+ "fallback": false,
1438
+ "group_size": 64,
1439
+ "metadata": {},
1440
+ "method": "affine",
1441
+ "module_path": "language_model.model.layers.20.mlp.up_proj",
1442
+ "note": "MLX-LM affine packing",
1443
+ "success": true
1444
+ },
1445
+ {
1446
+ "bits": 4,
1447
+ "fallback": false,
1448
+ "group_size": 64,
1449
+ "metadata": {},
1450
+ "method": "affine",
1451
+ "module_path": "language_model.model.layers.20.self_attn.k_proj",
1452
+ "note": "MLX-LM affine packing",
1453
+ "success": true
1454
+ },
1455
+ {
1456
+ "bits": 4,
1457
+ "fallback": false,
1458
+ "group_size": 64,
1459
+ "metadata": {},
1460
+ "method": "affine",
1461
+ "module_path": "language_model.model.layers.20.self_attn.o_proj",
1462
+ "note": "MLX-LM affine packing",
1463
+ "success": true
1464
+ },
1465
+ {
1466
+ "bits": 4,
1467
+ "fallback": false,
1468
+ "group_size": 64,
1469
+ "metadata": {},
1470
+ "method": "affine",
1471
+ "module_path": "language_model.model.layers.20.self_attn.q_proj",
1472
+ "note": "MLX-LM affine packing",
1473
+ "success": true
1474
+ },
1475
+ {
1476
+ "bits": 4,
1477
+ "fallback": false,
1478
+ "group_size": 64,
1479
+ "metadata": {},
1480
+ "method": "affine",
1481
+ "module_path": "language_model.model.layers.20.self_attn.v_proj",
1482
+ "note": "MLX-LM affine packing",
1483
+ "success": true
1484
+ },
1485
+ {
1486
+ "bits": 4,
1487
+ "fallback": false,
1488
+ "group_size": 64,
1489
+ "metadata": {},
1490
+ "method": "affine",
1491
+ "module_path": "language_model.model.layers.21.mlp.down_proj",
1492
+ "note": "MLX-LM affine packing",
1493
+ "success": true
1494
+ },
1495
+ {
1496
+ "bits": 4,
1497
+ "fallback": false,
1498
+ "group_size": 64,
1499
+ "metadata": {},
1500
+ "method": "affine",
1501
+ "module_path": "language_model.model.layers.21.mlp.gate_proj",
1502
+ "note": "MLX-LM affine packing",
1503
+ "success": true
1504
+ },
1505
+ {
1506
+ "bits": 4,
1507
+ "fallback": false,
1508
+ "group_size": 64,
1509
+ "metadata": {},
1510
+ "method": "affine",
1511
+ "module_path": "language_model.model.layers.21.mlp.up_proj",
1512
+ "note": "MLX-LM affine packing",
1513
+ "success": true
1514
+ },
1515
+ {
1516
+ "bits": 4,
1517
+ "fallback": false,
1518
+ "group_size": 64,
1519
+ "metadata": {},
1520
+ "method": "affine",
1521
+ "module_path": "language_model.model.layers.21.self_attn.k_proj",
1522
+ "note": "MLX-LM affine packing",
1523
+ "success": true
1524
+ },
1525
+ {
1526
+ "bits": 4,
1527
+ "fallback": false,
1528
+ "group_size": 64,
1529
+ "metadata": {},
1530
+ "method": "affine",
1531
+ "module_path": "language_model.model.layers.21.self_attn.o_proj",
1532
+ "note": "MLX-LM affine packing",
1533
+ "success": true
1534
+ },
1535
+ {
1536
+ "bits": 4,
1537
+ "fallback": false,
1538
+ "group_size": 64,
1539
+ "metadata": {},
1540
+ "method": "affine",
1541
+ "module_path": "language_model.model.layers.21.self_attn.q_proj",
1542
+ "note": "MLX-LM affine packing",
1543
+ "success": true
1544
+ },
1545
+ {
1546
+ "bits": 4,
1547
+ "fallback": false,
1548
+ "group_size": 64,
1549
+ "metadata": {},
1550
+ "method": "affine",
1551
+ "module_path": "language_model.model.layers.21.self_attn.v_proj",
1552
+ "note": "MLX-LM affine packing",
1553
+ "success": true
1554
+ },
1555
+ {
1556
+ "bits": 4,
1557
+ "fallback": false,
1558
+ "group_size": 64,
1559
+ "metadata": {},
1560
+ "method": "affine",
1561
+ "module_path": "language_model.model.layers.22.mlp.down_proj",
1562
+ "note": "MLX-LM affine packing",
1563
+ "success": true
1564
+ },
1565
+ {
1566
+ "bits": 4,
1567
+ "fallback": false,
1568
+ "group_size": 64,
1569
+ "metadata": {},
1570
+ "method": "affine",
1571
+ "module_path": "language_model.model.layers.22.mlp.gate_proj",
1572
+ "note": "MLX-LM affine packing",
1573
+ "success": true
1574
+ },
1575
+ {
1576
+ "bits": 4,
1577
+ "fallback": false,
1578
+ "group_size": 64,
1579
+ "metadata": {},
1580
+ "method": "affine",
1581
+ "module_path": "language_model.model.layers.22.mlp.up_proj",
1582
+ "note": "MLX-LM affine packing",
1583
+ "success": true
1584
+ },
1585
+ {
1586
+ "bits": 4,
1587
+ "fallback": false,
1588
+ "group_size": 64,
1589
+ "metadata": {},
1590
+ "method": "affine",
1591
+ "module_path": "language_model.model.layers.22.self_attn.k_proj",
1592
+ "note": "MLX-LM affine packing",
1593
+ "success": true
1594
+ },
1595
+ {
1596
+ "bits": 4,
1597
+ "fallback": false,
1598
+ "group_size": 64,
1599
+ "metadata": {},
1600
+ "method": "affine",
1601
+ "module_path": "language_model.model.layers.22.self_attn.o_proj",
1602
+ "note": "MLX-LM affine packing",
1603
+ "success": true
1604
+ },
1605
+ {
1606
+ "bits": 4,
1607
+ "fallback": false,
1608
+ "group_size": 64,
1609
+ "metadata": {},
1610
+ "method": "affine",
1611
+ "module_path": "language_model.model.layers.22.self_attn.q_proj",
1612
+ "note": "MLX-LM affine packing",
1613
+ "success": true
1614
+ },
1615
+ {
1616
+ "bits": 4,
1617
+ "fallback": false,
1618
+ "group_size": 64,
1619
+ "metadata": {},
1620
+ "method": "affine",
1621
+ "module_path": "language_model.model.layers.22.self_attn.v_proj",
1622
+ "note": "MLX-LM affine packing",
1623
+ "success": true
1624
+ },
1625
+ {
1626
+ "bits": 4,
1627
+ "fallback": false,
1628
+ "group_size": 64,
1629
+ "metadata": {},
1630
+ "method": "affine",
1631
+ "module_path": "language_model.model.layers.23.mlp.down_proj",
1632
+ "note": "MLX-LM affine packing",
1633
+ "success": true
1634
+ },
1635
+ {
1636
+ "bits": 4,
1637
+ "fallback": false,
1638
+ "group_size": 64,
1639
+ "metadata": {},
1640
+ "method": "affine",
1641
+ "module_path": "language_model.model.layers.23.mlp.gate_proj",
1642
+ "note": "MLX-LM affine packing",
1643
+ "success": true
1644
+ },
1645
+ {
1646
+ "bits": 4,
1647
+ "fallback": false,
1648
+ "group_size": 64,
1649
+ "metadata": {},
1650
+ "method": "affine",
1651
+ "module_path": "language_model.model.layers.23.self_attn.k_proj",
1652
+ "note": "MLX-LM affine packing",
1653
+ "success": true
1654
+ },
1655
+ {
1656
+ "bits": 4,
1657
+ "fallback": false,
1658
+ "group_size": 64,
1659
+ "metadata": {},
1660
+ "method": "affine",
1661
+ "module_path": "language_model.model.layers.23.self_attn.o_proj",
1662
+ "note": "MLX-LM affine packing",
1663
+ "success": true
1664
+ },
1665
+ {
1666
+ "bits": 4,
1667
+ "fallback": false,
1668
+ "group_size": 64,
1669
+ "metadata": {},
1670
+ "method": "affine",
1671
+ "module_path": "language_model.model.layers.23.self_attn.q_proj",
1672
+ "note": "MLX-LM affine packing",
1673
+ "success": true
1674
+ },
1675
+ {
1676
+ "bits": 4,
1677
+ "fallback": false,
1678
+ "group_size": 64,
1679
+ "metadata": {},
1680
+ "method": "affine",
1681
+ "module_path": "language_model.model.layers.23.self_attn.v_proj",
1682
+ "note": "MLX-LM affine packing",
1683
+ "success": true
1684
+ },
1685
+ {
1686
+ "bits": 4,
1687
+ "fallback": false,
1688
+ "group_size": 64,
1689
+ "metadata": {},
1690
+ "method": "affine",
1691
+ "module_path": "language_model.model.layers.23.mlp.up_proj",
1692
+ "note": "MLX-LM affine packing",
1693
+ "success": true
1694
+ },
1695
+ {
1696
+ "bits": 4,
1697
+ "fallback": false,
1698
+ "group_size": 64,
1699
+ "metadata": {},
1700
+ "method": "affine",
1701
+ "module_path": "language_model.model.layers.24.mlp.down_proj",
1702
+ "note": "MLX-LM affine packing",
1703
+ "success": true
1704
+ },
1705
+ {
1706
+ "bits": 4,
1707
+ "fallback": false,
1708
+ "group_size": 64,
1709
+ "metadata": {},
1710
+ "method": "affine",
1711
+ "module_path": "language_model.model.layers.24.mlp.gate_proj",
1712
+ "note": "MLX-LM affine packing",
1713
+ "success": true
1714
+ },
1715
+ {
1716
+ "bits": 4,
1717
+ "fallback": false,
1718
+ "group_size": 64,
1719
+ "metadata": {},
1720
+ "method": "affine",
1721
+ "module_path": "language_model.model.layers.24.mlp.up_proj",
1722
+ "note": "MLX-LM affine packing",
1723
+ "success": true
1724
+ },
1725
+ {
1726
+ "bits": 4,
1727
+ "fallback": false,
1728
+ "group_size": 64,
1729
+ "metadata": {},
1730
+ "method": "affine",
1731
+ "module_path": "language_model.model.layers.24.self_attn.k_proj",
1732
+ "note": "MLX-LM affine packing",
1733
+ "success": true
1734
+ },
1735
+ {
1736
+ "bits": 4,
1737
+ "fallback": false,
1738
+ "group_size": 64,
1739
+ "metadata": {},
1740
+ "method": "affine",
1741
+ "module_path": "language_model.model.layers.24.self_attn.o_proj",
1742
+ "note": "MLX-LM affine packing",
1743
+ "success": true
1744
+ },
1745
+ {
1746
+ "bits": 4,
1747
+ "fallback": false,
1748
+ "group_size": 64,
1749
+ "metadata": {},
1750
+ "method": "affine",
1751
+ "module_path": "language_model.model.layers.24.self_attn.q_proj",
1752
+ "note": "MLX-LM affine packing",
1753
+ "success": true
1754
+ },
1755
+ {
1756
+ "bits": 4,
1757
+ "fallback": false,
1758
+ "group_size": 64,
1759
+ "metadata": {},
1760
+ "method": "affine",
1761
+ "module_path": "language_model.model.layers.24.self_attn.v_proj",
1762
+ "note": "MLX-LM affine packing",
1763
+ "success": true
1764
+ },
1765
+ {
1766
+ "bits": 4,
1767
+ "fallback": false,
1768
+ "group_size": 64,
1769
+ "metadata": {},
1770
+ "method": "affine",
1771
+ "module_path": "language_model.model.layers.25.mlp.down_proj",
1772
+ "note": "MLX-LM affine packing",
1773
+ "success": true
1774
+ },
1775
+ {
1776
+ "bits": 4,
1777
+ "fallback": false,
1778
+ "group_size": 64,
1779
+ "metadata": {},
1780
+ "method": "affine",
1781
+ "module_path": "language_model.model.layers.25.mlp.gate_proj",
1782
+ "note": "MLX-LM affine packing",
1783
+ "success": true
1784
+ },
1785
+ {
1786
+ "bits": 4,
1787
+ "fallback": false,
1788
+ "group_size": 64,
1789
+ "metadata": {},
1790
+ "method": "affine",
1791
+ "module_path": "language_model.model.layers.25.mlp.up_proj",
1792
+ "note": "MLX-LM affine packing",
1793
+ "success": true
1794
+ },
1795
+ {
1796
+ "bits": 4,
1797
+ "fallback": false,
1798
+ "group_size": 64,
1799
+ "metadata": {},
1800
+ "method": "affine",
1801
+ "module_path": "language_model.model.layers.25.self_attn.k_proj",
1802
+ "note": "MLX-LM affine packing",
1803
+ "success": true
1804
+ },
1805
+ {
1806
+ "bits": 4,
1807
+ "fallback": false,
1808
+ "group_size": 64,
1809
+ "metadata": {},
1810
+ "method": "affine",
1811
+ "module_path": "language_model.model.layers.25.self_attn.o_proj",
1812
+ "note": "MLX-LM affine packing",
1813
+ "success": true
1814
+ },
1815
+ {
1816
+ "bits": 4,
1817
+ "fallback": false,
1818
+ "group_size": 64,
1819
+ "metadata": {},
1820
+ "method": "affine",
1821
+ "module_path": "language_model.model.layers.25.self_attn.q_proj",
1822
+ "note": "MLX-LM affine packing",
1823
+ "success": true
1824
+ },
1825
+ {
1826
+ "bits": 4,
1827
+ "fallback": false,
1828
+ "group_size": 64,
1829
+ "metadata": {},
1830
+ "method": "affine",
1831
+ "module_path": "language_model.model.layers.25.self_attn.v_proj",
1832
+ "note": "MLX-LM affine packing",
1833
+ "success": true
1834
+ },
1835
+ {
1836
+ "bits": 4,
1837
+ "fallback": false,
1838
+ "group_size": 64,
1839
+ "metadata": {},
1840
+ "method": "affine",
1841
+ "module_path": "language_model.model.layers.26.mlp.down_proj",
1842
+ "note": "MLX-LM affine packing",
1843
+ "success": true
1844
+ },
1845
+ {
1846
+ "bits": 4,
1847
+ "fallback": false,
1848
+ "group_size": 64,
1849
+ "metadata": {},
1850
+ "method": "affine",
1851
+ "module_path": "language_model.model.layers.26.mlp.gate_proj",
1852
+ "note": "MLX-LM affine packing",
1853
+ "success": true
1854
+ },
1855
+ {
1856
+ "bits": 4,
1857
+ "fallback": false,
1858
+ "group_size": 64,
1859
+ "metadata": {},
1860
+ "method": "affine",
1861
+ "module_path": "language_model.model.layers.26.mlp.up_proj",
1862
+ "note": "MLX-LM affine packing",
1863
+ "success": true
1864
+ },
1865
+ {
1866
+ "bits": 4,
1867
+ "fallback": false,
1868
+ "group_size": 64,
1869
+ "metadata": {},
1870
+ "method": "affine",
1871
+ "module_path": "language_model.model.layers.26.self_attn.k_proj",
1872
+ "note": "MLX-LM affine packing",
1873
+ "success": true
1874
+ },
1875
+ {
1876
+ "bits": 4,
1877
+ "fallback": false,
1878
+ "group_size": 64,
1879
+ "metadata": {},
1880
+ "method": "affine",
1881
+ "module_path": "language_model.model.layers.26.self_attn.o_proj",
1882
+ "note": "MLX-LM affine packing",
1883
+ "success": true
1884
+ },
1885
+ {
1886
+ "bits": 4,
1887
+ "fallback": false,
1888
+ "group_size": 64,
1889
+ "metadata": {},
1890
+ "method": "affine",
1891
+ "module_path": "language_model.model.layers.26.self_attn.q_proj",
1892
+ "note": "MLX-LM affine packing",
1893
+ "success": true
1894
+ },
1895
+ {
1896
+ "bits": 4,
1897
+ "fallback": false,
1898
+ "group_size": 64,
1899
+ "metadata": {},
1900
+ "method": "affine",
1901
+ "module_path": "language_model.model.layers.26.self_attn.v_proj",
1902
+ "note": "MLX-LM affine packing",
1903
+ "success": true
1904
+ },
1905
+ {
1906
+ "bits": 4,
1907
+ "fallback": false,
1908
+ "group_size": 64,
1909
+ "metadata": {},
1910
+ "method": "affine",
1911
+ "module_path": "language_model.model.layers.27.mlp.down_proj",
1912
+ "note": "MLX-LM affine packing",
1913
+ "success": true
1914
+ },
1915
+ {
1916
+ "bits": 4,
1917
+ "fallback": false,
1918
+ "group_size": 64,
1919
+ "metadata": {},
1920
+ "method": "affine",
1921
+ "module_path": "language_model.model.layers.27.mlp.gate_proj",
1922
+ "note": "MLX-LM affine packing",
1923
+ "success": true
1924
+ },
1925
+ {
1926
+ "bits": 4,
1927
+ "fallback": false,
1928
+ "group_size": 64,
1929
+ "metadata": {},
1930
+ "method": "affine",
1931
+ "module_path": "language_model.model.layers.27.mlp.up_proj",
1932
+ "note": "MLX-LM affine packing",
1933
+ "success": true
1934
+ },
1935
+ {
1936
+ "bits": 4,
1937
+ "fallback": false,
1938
+ "group_size": 64,
1939
+ "metadata": {},
1940
+ "method": "affine",
1941
+ "module_path": "language_model.model.layers.27.self_attn.k_proj",
1942
+ "note": "MLX-LM affine packing",
1943
+ "success": true
1944
+ },
1945
+ {
1946
+ "bits": 4,
1947
+ "fallback": false,
1948
+ "group_size": 64,
1949
+ "metadata": {},
1950
+ "method": "affine",
1951
+ "module_path": "language_model.model.layers.27.self_attn.o_proj",
1952
+ "note": "MLX-LM affine packing",
1953
+ "success": true
1954
+ },
1955
+ {
1956
+ "bits": 4,
1957
+ "fallback": false,
1958
+ "group_size": 64,
1959
+ "metadata": {},
1960
+ "method": "affine",
1961
+ "module_path": "language_model.model.layers.27.self_attn.q_proj",
1962
+ "note": "MLX-LM affine packing",
1963
+ "success": true
1964
+ },
1965
+ {
1966
+ "bits": 4,
1967
+ "fallback": false,
1968
+ "group_size": 64,
1969
+ "metadata": {},
1970
+ "method": "affine",
1971
+ "module_path": "language_model.model.layers.27.self_attn.v_proj",
1972
+ "note": "MLX-LM affine packing",
1973
+ "success": true
1974
+ },
1975
+ {
1976
+ "bits": 4,
1977
+ "fallback": false,
1978
+ "group_size": 64,
1979
+ "metadata": {},
1980
+ "method": "affine",
1981
+ "module_path": "language_model.model.layers.28.mlp.down_proj",
1982
+ "note": "MLX-LM affine packing",
1983
+ "success": true
1984
+ },
1985
+ {
1986
+ "bits": 4,
1987
+ "fallback": false,
1988
+ "group_size": 64,
1989
+ "metadata": {},
1990
+ "method": "affine",
1991
+ "module_path": "language_model.model.layers.28.mlp.gate_proj",
1992
+ "note": "MLX-LM affine packing",
1993
+ "success": true
1994
+ },
1995
+ {
1996
+ "bits": 4,
1997
+ "fallback": false,
1998
+ "group_size": 64,
1999
+ "metadata": {},
2000
+ "method": "affine",
2001
+ "module_path": "language_model.model.layers.28.mlp.up_proj",
2002
+ "note": "MLX-LM affine packing",
2003
+ "success": true
2004
+ },
2005
+ {
2006
+ "bits": 4,
2007
+ "fallback": false,
2008
+ "group_size": 64,
2009
+ "metadata": {},
2010
+ "method": "affine",
2011
+ "module_path": "language_model.model.layers.28.self_attn.k_proj",
2012
+ "note": "MLX-LM affine packing",
2013
+ "success": true
2014
+ },
2015
+ {
2016
+ "bits": 4,
2017
+ "fallback": false,
2018
+ "group_size": 64,
2019
+ "metadata": {},
2020
+ "method": "affine",
2021
+ "module_path": "language_model.model.layers.28.self_attn.o_proj",
2022
+ "note": "MLX-LM affine packing",
2023
+ "success": true
2024
+ },
2025
+ {
2026
+ "bits": 4,
2027
+ "fallback": false,
2028
+ "group_size": 64,
2029
+ "metadata": {},
2030
+ "method": "affine",
2031
+ "module_path": "language_model.model.layers.28.self_attn.q_proj",
2032
+ "note": "MLX-LM affine packing",
2033
+ "success": true
2034
+ },
2035
+ {
2036
+ "bits": 4,
2037
+ "fallback": false,
2038
+ "group_size": 64,
2039
+ "metadata": {},
2040
+ "method": "affine",
2041
+ "module_path": "language_model.model.layers.28.self_attn.v_proj",
2042
+ "note": "MLX-LM affine packing",
2043
+ "success": true
2044
+ },
2045
+ {
2046
+ "bits": 4,
2047
+ "fallback": false,
2048
+ "group_size": 64,
2049
+ "metadata": {},
2050
+ "method": "affine",
2051
+ "module_path": "language_model.model.layers.29.mlp.down_proj",
2052
+ "note": "MLX-LM affine packing",
2053
+ "success": true
2054
+ },
2055
+ {
2056
+ "bits": 4,
2057
+ "fallback": false,
2058
+ "group_size": 64,
2059
+ "metadata": {},
2060
+ "method": "affine",
2061
+ "module_path": "language_model.model.layers.29.mlp.gate_proj",
2062
+ "note": "MLX-LM affine packing",
2063
+ "success": true
2064
+ },
2065
+ {
2066
+ "bits": 4,
2067
+ "fallback": false,
2068
+ "group_size": 64,
2069
+ "metadata": {},
2070
+ "method": "affine",
2071
+ "module_path": "language_model.model.layers.29.mlp.up_proj",
2072
+ "note": "MLX-LM affine packing",
2073
+ "success": true
2074
+ },
2075
+ {
2076
+ "bits": 4,
2077
+ "fallback": false,
2078
+ "group_size": 64,
2079
+ "metadata": {},
2080
+ "method": "affine",
2081
+ "module_path": "language_model.model.layers.29.self_attn.k_proj",
2082
+ "note": "MLX-LM affine packing",
2083
+ "success": true
2084
+ },
2085
+ {
2086
+ "bits": 4,
2087
+ "fallback": false,
2088
+ "group_size": 64,
2089
+ "metadata": {},
2090
+ "method": "affine",
2091
+ "module_path": "language_model.model.layers.29.self_attn.o_proj",
2092
+ "note": "MLX-LM affine packing",
2093
+ "success": true
2094
+ },
2095
+ {
2096
+ "bits": 4,
2097
+ "fallback": false,
2098
+ "group_size": 64,
2099
+ "metadata": {},
2100
+ "method": "affine",
2101
+ "module_path": "language_model.model.layers.29.self_attn.q_proj",
2102
+ "note": "MLX-LM affine packing",
2103
+ "success": true
2104
+ },
2105
+ {
2106
+ "bits": 4,
2107
+ "fallback": false,
2108
+ "group_size": 64,
2109
+ "metadata": {},
2110
+ "method": "affine",
2111
+ "module_path": "language_model.model.layers.29.self_attn.v_proj",
2112
+ "note": "MLX-LM affine packing",
2113
+ "success": true
2114
+ },
2115
+ {
2116
+ "bits": 4,
2117
+ "fallback": false,
2118
+ "group_size": 64,
2119
+ "metadata": {},
2120
+ "method": "affine",
2121
+ "module_path": "language_model.model.layers.30.mlp.down_proj",
2122
+ "note": "MLX-LM affine packing",
2123
+ "success": true
2124
+ },
2125
+ {
2126
+ "bits": 4,
2127
+ "fallback": false,
2128
+ "group_size": 64,
2129
+ "metadata": {},
2130
+ "method": "affine",
2131
+ "module_path": "language_model.model.layers.30.mlp.gate_proj",
2132
+ "note": "MLX-LM affine packing",
2133
+ "success": true
2134
+ },
2135
+ {
2136
+ "bits": 4,
2137
+ "fallback": false,
2138
+ "group_size": 64,
2139
+ "metadata": {},
2140
+ "method": "affine",
2141
+ "module_path": "language_model.model.layers.30.mlp.up_proj",
2142
+ "note": "MLX-LM affine packing",
2143
+ "success": true
2144
+ },
2145
+ {
2146
+ "bits": 4,
2147
+ "fallback": false,
2148
+ "group_size": 64,
2149
+ "metadata": {},
2150
+ "method": "affine",
2151
+ "module_path": "language_model.model.layers.30.self_attn.k_proj",
2152
+ "note": "MLX-LM affine packing",
2153
+ "success": true
2154
+ },
2155
+ {
2156
+ "bits": 4,
2157
+ "fallback": false,
2158
+ "group_size": 64,
2159
+ "metadata": {},
2160
+ "method": "affine",
2161
+ "module_path": "language_model.model.layers.30.self_attn.o_proj",
2162
+ "note": "MLX-LM affine packing",
2163
+ "success": true
2164
+ },
2165
+ {
2166
+ "bits": 4,
2167
+ "fallback": false,
2168
+ "group_size": 64,
2169
+ "metadata": {},
2170
+ "method": "affine",
2171
+ "module_path": "language_model.model.layers.30.self_attn.q_proj",
2172
+ "note": "MLX-LM affine packing",
2173
+ "success": true
2174
+ },
2175
+ {
2176
+ "bits": 4,
2177
+ "fallback": false,
2178
+ "group_size": 64,
2179
+ "metadata": {},
2180
+ "method": "affine",
2181
+ "module_path": "language_model.model.layers.30.self_attn.v_proj",
2182
+ "note": "MLX-LM affine packing",
2183
+ "success": true
2184
+ },
2185
+ {
2186
+ "bits": 4,
2187
+ "fallback": false,
2188
+ "group_size": 64,
2189
+ "metadata": {},
2190
+ "method": "affine",
2191
+ "module_path": "language_model.model.layers.31.mlp.down_proj",
2192
+ "note": "MLX-LM affine packing",
2193
+ "success": true
2194
+ },
2195
+ {
2196
+ "bits": 4,
2197
+ "fallback": false,
2198
+ "group_size": 64,
2199
+ "metadata": {},
2200
+ "method": "affine",
2201
+ "module_path": "language_model.model.layers.31.mlp.gate_proj",
2202
+ "note": "MLX-LM affine packing",
2203
+ "success": true
2204
+ },
2205
+ {
2206
+ "bits": 4,
2207
+ "fallback": false,
2208
+ "group_size": 64,
2209
+ "metadata": {},
2210
+ "method": "affine",
2211
+ "module_path": "language_model.model.layers.31.mlp.up_proj",
2212
+ "note": "MLX-LM affine packing",
2213
+ "success": true
2214
+ },
2215
+ {
2216
+ "bits": 4,
2217
+ "fallback": false,
2218
+ "group_size": 64,
2219
+ "metadata": {},
2220
+ "method": "affine",
2221
+ "module_path": "language_model.model.layers.31.self_attn.k_proj",
2222
+ "note": "MLX-LM affine packing",
2223
+ "success": true
2224
+ },
2225
+ {
2226
+ "bits": 4,
2227
+ "fallback": false,
2228
+ "group_size": 64,
2229
+ "metadata": {},
2230
+ "method": "affine",
2231
+ "module_path": "language_model.model.layers.31.self_attn.o_proj",
2232
+ "note": "MLX-LM affine packing",
2233
+ "success": true
2234
+ },
2235
+ {
2236
+ "bits": 4,
2237
+ "fallback": false,
2238
+ "group_size": 64,
2239
+ "metadata": {},
2240
+ "method": "affine",
2241
+ "module_path": "language_model.model.layers.31.self_attn.q_proj",
2242
+ "note": "MLX-LM affine packing",
2243
+ "success": true
2244
+ },
2245
+ {
2246
+ "bits": 4,
2247
+ "fallback": false,
2248
+ "group_size": 64,
2249
+ "metadata": {},
2250
+ "method": "affine",
2251
+ "module_path": "language_model.model.layers.31.self_attn.v_proj",
2252
+ "note": "MLX-LM affine packing",
2253
+ "success": true
2254
+ },
2255
+ {
2256
+ "bits": 4,
2257
+ "fallback": false,
2258
+ "group_size": 64,
2259
+ "metadata": {},
2260
+ "method": "affine",
2261
+ "module_path": "language_model.model.layers.32.mlp.gate_proj",
2262
+ "note": "MLX-LM affine packing",
2263
+ "success": true
2264
+ },
2265
+ {
2266
+ "bits": 4,
2267
+ "fallback": false,
2268
+ "group_size": 64,
2269
+ "metadata": {},
2270
+ "method": "affine",
2271
+ "module_path": "language_model.model.layers.32.self_attn.k_proj",
2272
+ "note": "MLX-LM affine packing",
2273
+ "success": true
2274
+ },
2275
+ {
2276
+ "bits": 4,
2277
+ "fallback": false,
2278
+ "group_size": 64,
2279
+ "metadata": {},
2280
+ "method": "affine",
2281
+ "module_path": "language_model.model.layers.32.self_attn.o_proj",
2282
+ "note": "MLX-LM affine packing",
2283
+ "success": true
2284
+ },
2285
+ {
2286
+ "bits": 4,
2287
+ "fallback": false,
2288
+ "group_size": 64,
2289
+ "metadata": {},
2290
+ "method": "affine",
2291
+ "module_path": "language_model.model.layers.32.self_attn.q_proj",
2292
+ "note": "MLX-LM affine packing",
2293
+ "success": true
2294
+ },
2295
+ {
2296
+ "bits": 4,
2297
+ "fallback": false,
2298
+ "group_size": 64,
2299
+ "metadata": {},
2300
+ "method": "affine",
2301
+ "module_path": "language_model.model.layers.32.self_attn.v_proj",
2302
+ "note": "MLX-LM affine packing",
2303
+ "success": true
2304
+ },
2305
+ {
2306
+ "bits": 4,
2307
+ "fallback": false,
2308
+ "group_size": 64,
2309
+ "metadata": {},
2310
+ "method": "affine",
2311
+ "module_path": "language_model.model.layers.32.mlp.down_proj",
2312
+ "note": "MLX-LM affine packing",
2313
+ "success": true
2314
+ },
2315
+ {
2316
+ "bits": 4,
2317
+ "fallback": false,
2318
+ "group_size": 64,
2319
+ "metadata": {},
2320
+ "method": "affine",
2321
+ "module_path": "language_model.model.layers.32.mlp.up_proj",
2322
+ "note": "MLX-LM affine packing",
2323
+ "success": true
2324
+ },
2325
+ {
2326
+ "bits": 4,
2327
+ "fallback": false,
2328
+ "group_size": 64,
2329
+ "metadata": {},
2330
+ "method": "affine",
2331
+ "module_path": "language_model.model.layers.33.mlp.down_proj",
2332
+ "note": "MLX-LM affine packing",
2333
+ "success": true
2334
+ },
2335
+ {
2336
+ "bits": 4,
2337
+ "fallback": false,
2338
+ "group_size": 64,
2339
+ "metadata": {},
2340
+ "method": "affine",
2341
+ "module_path": "language_model.model.layers.33.mlp.gate_proj",
2342
+ "note": "MLX-LM affine packing",
2343
+ "success": true
2344
+ },
2345
+ {
2346
+ "bits": 4,
2347
+ "fallback": false,
2348
+ "group_size": 64,
2349
+ "metadata": {},
2350
+ "method": "affine",
2351
+ "module_path": "language_model.model.layers.33.mlp.up_proj",
2352
+ "note": "MLX-LM affine packing",
2353
+ "success": true
2354
+ },
2355
+ {
2356
+ "bits": 4,
2357
+ "fallback": false,
2358
+ "group_size": 64,
2359
+ "metadata": {},
2360
+ "method": "affine",
2361
+ "module_path": "language_model.model.layers.33.self_attn.k_proj",
2362
+ "note": "MLX-LM affine packing",
2363
+ "success": true
2364
+ },
2365
+ {
2366
+ "bits": 4,
2367
+ "fallback": false,
2368
+ "group_size": 64,
2369
+ "metadata": {},
2370
+ "method": "affine",
2371
+ "module_path": "language_model.model.layers.33.self_attn.o_proj",
2372
+ "note": "MLX-LM affine packing",
2373
+ "success": true
2374
+ },
2375
+ {
2376
+ "bits": 4,
2377
+ "fallback": false,
2378
+ "group_size": 64,
2379
+ "metadata": {},
2380
+ "method": "affine",
2381
+ "module_path": "language_model.model.layers.33.self_attn.q_proj",
2382
+ "note": "MLX-LM affine packing",
2383
+ "success": true
2384
+ },
2385
+ {
2386
+ "bits": 4,
2387
+ "fallback": false,
2388
+ "group_size": 64,
2389
+ "metadata": {},
2390
+ "method": "affine",
2391
+ "module_path": "language_model.model.layers.33.self_attn.v_proj",
2392
+ "note": "MLX-LM affine packing",
2393
+ "success": true
2394
+ },
2395
+ {
2396
+ "bits": 4,
2397
+ "fallback": false,
2398
+ "group_size": 64,
2399
+ "metadata": {},
2400
+ "method": "affine",
2401
+ "module_path": "language_model.model.layers.34.mlp.down_proj",
2402
+ "note": "MLX-LM affine packing",
2403
+ "success": true
2404
+ },
2405
+ {
2406
+ "bits": 4,
2407
+ "fallback": false,
2408
+ "group_size": 64,
2409
+ "metadata": {},
2410
+ "method": "affine",
2411
+ "module_path": "language_model.model.layers.34.mlp.gate_proj",
2412
+ "note": "MLX-LM affine packing",
2413
+ "success": true
2414
+ },
2415
+ {
2416
+ "bits": 4,
2417
+ "fallback": false,
2418
+ "group_size": 64,
2419
+ "metadata": {},
2420
+ "method": "affine",
2421
+ "module_path": "language_model.model.layers.34.mlp.up_proj",
2422
+ "note": "MLX-LM affine packing",
2423
+ "success": true
2424
+ },
2425
+ {
2426
+ "bits": 4,
2427
+ "fallback": false,
2428
+ "group_size": 32,
2429
+ "metadata": {},
2430
+ "method": "affine",
2431
+ "module_path": "language_model.model.layers.34.self_attn.k_proj",
2432
+ "note": "MLX-LM affine packing",
2433
+ "success": true
2434
+ },
2435
+ {
2436
+ "bits": 4,
2437
+ "fallback": false,
2438
+ "group_size": 64,
2439
+ "metadata": {},
2440
+ "method": "affine",
2441
+ "module_path": "language_model.model.layers.34.self_attn.o_proj",
2442
+ "note": "MLX-LM affine packing",
2443
+ "success": true
2444
+ },
2445
+ {
2446
+ "bits": 4,
2447
+ "fallback": false,
2448
+ "group_size": 64,
2449
+ "metadata": {},
2450
+ "method": "affine",
2451
+ "module_path": "language_model.model.layers.34.self_attn.q_proj",
2452
+ "note": "MLX-LM affine packing",
2453
+ "success": true
2454
+ },
2455
+ {
2456
+ "bits": 4,
2457
+ "fallback": false,
2458
+ "group_size": 32,
2459
+ "metadata": {},
2460
+ "method": "affine",
2461
+ "module_path": "language_model.model.layers.34.self_attn.v_proj",
2462
+ "note": "MLX-LM affine packing",
2463
+ "success": true
2464
+ },
2465
+ {
2466
+ "bits": 4,
2467
+ "fallback": false,
2468
+ "group_size": 64,
2469
+ "metadata": {},
2470
+ "method": "affine",
2471
+ "module_path": "language_model.model.layers.35.mlp.down_proj",
2472
+ "note": "MLX-LM affine packing",
2473
+ "success": true
2474
+ },
2475
+ {
2476
+ "bits": 4,
2477
+ "fallback": false,
2478
+ "group_size": 64,
2479
+ "metadata": {},
2480
+ "method": "affine",
2481
+ "module_path": "language_model.model.layers.35.mlp.gate_proj",
2482
+ "note": "MLX-LM affine packing",
2483
+ "success": true
2484
+ },
2485
+ {
2486
+ "bits": 4,
2487
+ "fallback": false,
2488
+ "group_size": 64,
2489
+ "metadata": {},
2490
+ "method": "affine",
2491
+ "module_path": "language_model.model.layers.35.mlp.up_proj",
2492
+ "note": "MLX-LM affine packing",
2493
+ "success": true
2494
+ },
2495
+ {
2496
+ "bits": 4,
2497
+ "fallback": false,
2498
+ "group_size": 32,
2499
+ "metadata": {},
2500
+ "method": "affine",
2501
+ "module_path": "language_model.model.layers.35.self_attn.k_proj",
2502
+ "note": "MLX-LM affine packing",
2503
+ "success": true
2504
+ },
2505
+ {
2506
+ "bits": 4,
2507
+ "fallback": false,
2508
+ "group_size": 64,
2509
+ "metadata": {},
2510
+ "method": "affine",
2511
+ "module_path": "language_model.model.layers.35.self_attn.o_proj",
2512
+ "note": "MLX-LM affine packing",
2513
+ "success": true
2514
+ },
2515
+ {
2516
+ "bits": 4,
2517
+ "fallback": false,
2518
+ "group_size": 32,
2519
+ "metadata": {},
2520
+ "method": "affine",
2521
+ "module_path": "language_model.model.layers.35.self_attn.q_proj",
2522
+ "note": "MLX-LM affine packing",
2523
+ "success": true
2524
+ },
2525
+ {
2526
+ "bits": 4,
2527
+ "fallback": false,
2528
+ "group_size": 32,
2529
+ "metadata": {},
2530
+ "method": "affine",
2531
+ "module_path": "language_model.model.layers.35.self_attn.v_proj",
2532
+ "note": "MLX-LM affine packing",
2533
+ "success": true
2534
+ },
2535
+ {
2536
+ "bits": 4,
2537
+ "fallback": false,
2538
+ "group_size": 64,
2539
+ "metadata": {},
2540
+ "method": "affine",
2541
+ "module_path": "language_model.model.layers.36.mlp.down_proj",
2542
+ "note": "MLX-LM affine packing",
2543
+ "success": true
2544
+ },
2545
+ {
2546
+ "bits": 4,
2547
+ "fallback": false,
2548
+ "group_size": 64,
2549
+ "metadata": {},
2550
+ "method": "affine",
2551
+ "module_path": "language_model.model.layers.36.mlp.gate_proj",
2552
+ "note": "MLX-LM affine packing",
2553
+ "success": true
2554
+ },
2555
+ {
2556
+ "bits": 4,
2557
+ "fallback": false,
2558
+ "group_size": 64,
2559
+ "metadata": {},
2560
+ "method": "affine",
2561
+ "module_path": "language_model.model.layers.36.mlp.up_proj",
2562
+ "note": "MLX-LM affine packing",
2563
+ "success": true
2564
+ },
2565
+ {
2566
+ "bits": 4,
2567
+ "fallback": false,
2568
+ "group_size": 32,
2569
+ "metadata": {},
2570
+ "method": "affine",
2571
+ "module_path": "language_model.model.layers.36.self_attn.k_proj",
2572
+ "note": "MLX-LM affine packing",
2573
+ "success": true
2574
+ },
2575
+ {
2576
+ "bits": 4,
2577
+ "fallback": false,
2578
+ "group_size": 32,
2579
+ "metadata": {},
2580
+ "method": "affine",
2581
+ "module_path": "language_model.model.layers.36.self_attn.o_proj",
2582
+ "note": "MLX-LM affine packing",
2583
+ "success": true
2584
+ },
2585
+ {
2586
+ "bits": 4,
2587
+ "fallback": false,
2588
+ "group_size": 32,
2589
+ "metadata": {},
2590
+ "method": "affine",
2591
+ "module_path": "language_model.model.layers.36.self_attn.q_proj",
2592
+ "note": "MLX-LM affine packing",
2593
+ "success": true
2594
+ },
2595
+ {
2596
+ "bits": 4,
2597
+ "fallback": false,
2598
+ "group_size": 32,
2599
+ "metadata": {},
2600
+ "method": "affine",
2601
+ "module_path": "language_model.model.layers.36.self_attn.v_proj",
2602
+ "note": "MLX-LM affine packing",
2603
+ "success": true
2604
+ },
2605
+ {
2606
+ "bits": 4,
2607
+ "fallback": false,
2608
+ "group_size": 64,
2609
+ "metadata": {},
2610
+ "method": "affine",
2611
+ "module_path": "language_model.model.layers.37.mlp.down_proj",
2612
+ "note": "MLX-LM affine packing",
2613
+ "success": true
2614
+ },
2615
+ {
2616
+ "bits": 4,
2617
+ "fallback": false,
2618
+ "group_size": 64,
2619
+ "metadata": {},
2620
+ "method": "affine",
2621
+ "module_path": "language_model.model.layers.37.mlp.gate_proj",
2622
+ "note": "MLX-LM affine packing",
2623
+ "success": true
2624
+ },
2625
+ {
2626
+ "bits": 4,
2627
+ "fallback": false,
2628
+ "group_size": 64,
2629
+ "metadata": {},
2630
+ "method": "affine",
2631
+ "module_path": "language_model.model.layers.37.mlp.up_proj",
2632
+ "note": "MLX-LM affine packing",
2633
+ "success": true
2634
+ },
2635
+ {
2636
+ "bits": 4,
2637
+ "fallback": false,
2638
+ "group_size": 32,
2639
+ "metadata": {},
2640
+ "method": "affine",
2641
+ "module_path": "language_model.model.layers.37.self_attn.k_proj",
2642
+ "note": "MLX-LM affine packing",
2643
+ "success": true
2644
+ },
2645
+ {
2646
+ "bits": 4,
2647
+ "fallback": false,
2648
+ "group_size": 32,
2649
+ "metadata": {},
2650
+ "method": "affine",
2651
+ "module_path": "language_model.model.layers.37.self_attn.o_proj",
2652
+ "note": "MLX-LM affine packing",
2653
+ "success": true
2654
+ },
2655
+ {
2656
+ "bits": 4,
2657
+ "fallback": false,
2658
+ "group_size": 32,
2659
+ "metadata": {},
2660
+ "method": "affine",
2661
+ "module_path": "language_model.model.layers.37.self_attn.q_proj",
2662
+ "note": "MLX-LM affine packing",
2663
+ "success": true
2664
+ },
2665
+ {
2666
+ "bits": 4,
2667
+ "fallback": false,
2668
+ "group_size": 32,
2669
+ "metadata": {},
2670
+ "method": "affine",
2671
+ "module_path": "language_model.model.layers.37.self_attn.v_proj",
2672
+ "note": "MLX-LM affine packing",
2673
+ "success": true
2674
+ },
2675
+ {
2676
+ "bits": 4,
2677
+ "fallback": false,
2678
+ "group_size": 64,
2679
+ "metadata": {},
2680
+ "method": "affine",
2681
+ "module_path": "language_model.model.layers.38.mlp.down_proj",
2682
+ "note": "MLX-LM affine packing",
2683
+ "success": true
2684
+ },
2685
+ {
2686
+ "bits": 4,
2687
+ "fallback": false,
2688
+ "group_size": 64,
2689
+ "metadata": {},
2690
+ "method": "affine",
2691
+ "module_path": "language_model.model.layers.38.mlp.gate_proj",
2692
+ "note": "MLX-LM affine packing",
2693
+ "success": true
2694
+ },
2695
+ {
2696
+ "bits": 4,
2697
+ "fallback": false,
2698
+ "group_size": 64,
2699
+ "metadata": {},
2700
+ "method": "affine",
2701
+ "module_path": "language_model.model.layers.38.mlp.up_proj",
2702
+ "note": "MLX-LM affine packing",
2703
+ "success": true
2704
+ },
2705
+ {
2706
+ "bits": 4,
2707
+ "fallback": false,
2708
+ "group_size": 32,
2709
+ "metadata": {},
2710
+ "method": "affine",
2711
+ "module_path": "language_model.model.layers.38.self_attn.k_proj",
2712
+ "note": "MLX-LM affine packing",
2713
+ "success": true
2714
+ },
2715
+ {
2716
+ "bits": 4,
2717
+ "fallback": false,
2718
+ "group_size": 32,
2719
+ "metadata": {},
2720
+ "method": "affine",
2721
+ "module_path": "language_model.model.layers.38.self_attn.o_proj",
2722
+ "note": "MLX-LM affine packing",
2723
+ "success": true
2724
+ },
2725
+ {
2726
+ "bits": 4,
2727
+ "fallback": false,
2728
+ "group_size": 32,
2729
+ "metadata": {},
2730
+ "method": "affine",
2731
+ "module_path": "language_model.model.layers.38.self_attn.q_proj",
2732
+ "note": "MLX-LM affine packing",
2733
+ "success": true
2734
+ },
2735
+ {
2736
+ "bits": 4,
2737
+ "fallback": false,
2738
+ "group_size": 32,
2739
+ "metadata": {},
2740
+ "method": "affine",
2741
+ "module_path": "language_model.model.layers.38.self_attn.v_proj",
2742
+ "note": "MLX-LM affine packing",
2743
+ "success": true
2744
+ },
2745
+ {
2746
+ "bits": 4,
2747
+ "fallback": false,
2748
+ "group_size": 64,
2749
+ "metadata": {},
2750
+ "method": "affine",
2751
+ "module_path": "language_model.model.layers.39.mlp.down_proj",
2752
+ "note": "MLX-LM affine packing",
2753
+ "success": true
2754
+ },
2755
+ {
2756
+ "bits": 4,
2757
+ "fallback": false,
2758
+ "group_size": 64,
2759
+ "metadata": {},
2760
+ "method": "affine",
2761
+ "module_path": "language_model.model.layers.39.mlp.gate_proj",
2762
+ "note": "MLX-LM affine packing",
2763
+ "success": true
2764
+ },
2765
+ {
2766
+ "bits": 4,
2767
+ "fallback": false,
2768
+ "group_size": 64,
2769
+ "metadata": {},
2770
+ "method": "affine",
2771
+ "module_path": "language_model.model.layers.39.mlp.up_proj",
2772
+ "note": "MLX-LM affine packing",
2773
+ "success": true
2774
+ },
2775
+ {
2776
+ "bits": 4,
2777
+ "fallback": false,
2778
+ "group_size": 32,
2779
+ "metadata": {},
2780
+ "method": "affine",
2781
+ "module_path": "language_model.model.layers.39.self_attn.k_proj",
2782
+ "note": "MLX-LM affine packing",
2783
+ "success": true
2784
+ },
2785
+ {
2786
+ "bits": 4,
2787
+ "fallback": false,
2788
+ "group_size": 32,
2789
+ "metadata": {},
2790
+ "method": "affine",
2791
+ "module_path": "language_model.model.layers.39.self_attn.o_proj",
2792
+ "note": "MLX-LM affine packing",
2793
+ "success": true
2794
+ },
2795
+ {
2796
+ "bits": 4,
2797
+ "fallback": false,
2798
+ "group_size": 32,
2799
+ "metadata": {},
2800
+ "method": "affine",
2801
+ "module_path": "language_model.model.layers.39.self_attn.q_proj",
2802
+ "note": "MLX-LM affine packing",
2803
+ "success": true
2804
+ },
2805
+ {
2806
+ "bits": 4,
2807
+ "fallback": false,
2808
+ "group_size": 32,
2809
+ "metadata": {},
2810
+ "method": "affine",
2811
+ "module_path": "language_model.model.layers.39.self_attn.v_proj",
2812
+ "note": "MLX-LM affine packing",
2813
+ "success": true
2814
+ }
2815
+ ],
2816
+ "schema_version": "axquant.quantizer-execution.v1"
2817
+ }
axquant_runtime.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "ax_engine": {
3
+ "decode_kernel": null,
4
+ "fused_mtp": null,
5
+ "kernel_evidence": "unmeasured",
6
+ "model_manifest": "model-manifest.json",
7
+ "preferred_group_size": 32
8
+ },
9
+ "compatible_runtimes": [
10
+ {
11
+ "compatibility_level": "B",
12
+ "manifest": "config.json",
13
+ "mtp_support": "runtime-dependent",
14
+ "name": "mlx-lm",
15
+ "notes": [
16
+ "Standard backbone inference is the compatibility target.",
17
+ "AXQuant MTP metadata may be ignored by MLX-LM."
18
+ ],
19
+ "standard_inference": true,
20
+ "standard_mlx_weights": true,
21
+ "support_level": "standard-inference"
22
+ }
23
+ ],
24
+ "created_at": "2026-08-02T19:52:51.472949Z",
25
+ "kv_cache": null,
26
+ "memory_policy": {
27
+ "kv_cache_precision": "runtime-default",
28
+ "mtp_buffers": "not-required",
29
+ "prefix_cache": "runtime-managed",
30
+ "unified_memory_safety_margin": "benchmark-required"
31
+ },
32
+ "mtp": {
33
+ "acceptance_retention": null,
34
+ "detected": false,
35
+ "draft_tokens": null,
36
+ "enabled_by_default": false,
37
+ "head_precision": null,
38
+ "measured_speedup": null,
39
+ "optimized": false,
40
+ "recommended_temperature_max": null,
41
+ "sidecar_file": null,
42
+ "verification_mode": null
43
+ },
44
+ "optimization_scope": "text-path",
45
+ "primary_runtime": {
46
+ "compatibility_level": "A",
47
+ "manifest": "model-manifest.json",
48
+ "mtp_support": "native",
49
+ "name": "ax-engine",
50
+ "notes": [
51
+ "Runtime claims require a passing AX Engine doctor and benchmark report."
52
+ ],
53
+ "standard_inference": true,
54
+ "standard_mlx_weights": true,
55
+ "support_level": "optimized"
56
+ },
57
+ "schema_version": "axquant.runtime.v1"
58
+ }
chat_template.jinja ADDED
@@ -0,0 +1,216 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {#- Default system message if no system prompt is passed. #}
2
+ {%- set default_system_message = 'You are Ministral-3-14B-Instruct-2512, a Large Language Model (LLM) created by Mistral AI, a French startup headquartered in Paris.
3
+ You power an AI assistant called Le Chat.
4
+ Your knowledge base was last updated on 2023-10-01.
5
+ The current date is {today}.
6
+
7
+ When you\'re not sure about some information or when the user\'s request requires up-to-date or specific data, you must use the available tools to fetch the information. Do not hesitate to use tools whenever they can provide a more accurate or complete response. If no relevant tools are available, then clearly state that you don\'t have the information and avoid making up anything.
8
+ If the user\'s question is not clear, ambiguous, or does not provide enough context for you to accurately answer the question, you do not try to answer it right away and you rather ask the user to clarify their request (e.g. "What are some good restaurants around me?" => "Where are you?" or "When is the next flight to Tokyo" => "Where do you travel from?").
9
+ You are always very attentive to dates, in particular you try to resolve dates (e.g. "yesterday" is {yesterday}) and when asked about information at specific dates, you discard information that is at another date.
10
+ You follow these instructions in all languages, and always respond to the user in the language they use or request.
11
+ Next sections describe the capabilities that you have.
12
+
13
+ # WEB BROWSING INSTRUCTIONS
14
+
15
+ You cannot perform any web search or access internet to open URLs, links etc. If it seems like the user is expecting you to do so, you clarify the situation and ask the user to copy paste the text directly in the chat.
16
+
17
+ # MULTI-MODAL INSTRUCTIONS
18
+
19
+ You have the ability to read images, but you cannot generate images. You also cannot read nor transcribe audio files or videos.
20
+
21
+ # TOOL CALLING INSTRUCTIONS
22
+
23
+ You may have access to tools that you can use to fetch information or perform actions. You must use these tools in the following situations:
24
+
25
+ 1. When the request requires up-to-date information.
26
+ 2. When the request requires specific data that you do not have in your knowledge base.
27
+ 3. When the request involves actions that you cannot perform without tools.
28
+
29
+ Always prioritize using tools to provide the most accurate and helpful response. If tools are not available, inform the user that you cannot perform the requested action at the moment.' %}
30
+
31
+ {#- Begin of sequence token. #}
32
+ {{- bos_token }}
33
+
34
+
35
+ {#- Handle system prompt if it exists. #}
36
+ {%- set loop_messages = messages %}
37
+ {%- if messages[0]['role'] != 'system' and default_system_message != '' %}
38
+ {{- '[SYSTEM_PROMPT]' + default_system_message + '[/SYSTEM_PROMPT]' }}
39
+ {%- endif %}
40
+
41
+
42
+ {#- Tools definition #}
43
+ {%- set available_tools = '' %}
44
+ {%- set has_tools = false %}
45
+ {%- if tools is defined and tools is not none and tools|length > 0 %}
46
+ {%- set has_tools = true %}
47
+ {%- set available_tools = '[AVAILABLE_TOOLS]' + (tools| tojson) + '[/AVAILABLE_TOOLS]' %}
48
+ {%- endif %}
49
+
50
+ {#- Macros #}
51
+ {%- macro render_content(content, context_name, supported_types_desc, support_images) -%}
52
+ {%- if content is string -%}
53
+ {{- content -}}
54
+ {%- elif content -%}
55
+ {%- for block in content -%}
56
+ {%- if block['type'] == 'text' -%}
57
+ {{- block['text'] -}}
58
+ {%- elif support_images and block['type'] in ['image', 'image_url'] -%}
59
+ {{- '[IMG]' -}}
60
+ {%- else -%}
61
+ {{- raise_exception('Only ' + supported_types_desc + ' chunks are supported in ' + context_name + '.') -}}
62
+ {%- endif -%}
63
+ {%- endfor -%}
64
+ {%- else -%}
65
+ {{- raise_exception(context_name + ' must have non-empty content.') -}}
66
+ {%- endif -%}
67
+ {%- endmacro -%}
68
+
69
+ {#- Aggregate consecutive messages with the same role except system and tool. #}
70
+ {#- A sentinel message is appended so the last group gets flushed inside the loop. #}
71
+ {%- set ns_agg = namespace(messages=[], current_group=[], current_role=none) %}
72
+ {%- for message in loop_messages + [{'role': '__sentinel__'}] %}
73
+ {%- if message['role'] != ns_agg.current_role or message['role'] == 'system' or message['role'] == 'tool' %}
74
+ {%- if ns_agg.current_role == 'tool' %}
75
+ {%- set ns_agg.messages = ns_agg.messages + ns_agg.current_group %}
76
+ {%- elif ns_agg.current_role is not none %}
77
+ {%- set ns_c = namespace(text_parts=[], chunks=[], has_non_text=false, tool_calls=[]) %}
78
+ {%- for msg in ns_agg.current_group %}
79
+ {%- if msg['content'] is string %}
80
+ {%- set ns_c.text_parts = ns_c.text_parts + [msg['content']] %}
81
+ {%- elif msg['content'] is not none %}
82
+ {%- for block in msg['content'] %}
83
+ {%- if block['type'] == 'text' %}
84
+ {%- set ns_c.text_parts = ns_c.text_parts + [block['text']] %}
85
+ {%- else %}
86
+ {%- if ns_c.text_parts | length > 0 %}
87
+ {%- set ns_c.chunks = ns_c.chunks + [{'type': 'text', 'text': ns_c.text_parts | join('\n\n')}] %}
88
+ {%- set ns_c.text_parts = [] %}
89
+ {%- endif %}
90
+ {%- set ns_c.chunks = ns_c.chunks + [block] %}
91
+ {%- set ns_c.has_non_text = true %}
92
+ {%- endif %}
93
+ {%- endfor %}
94
+ {%- endif %}
95
+ {%- if msg['tool_calls'] is defined and msg['tool_calls'] is not none %}
96
+ {%- set ns_c.tool_calls = ns_c.tool_calls + msg['tool_calls'] | list %}
97
+ {%- endif %}
98
+ {%- endfor %}
99
+ {%- if ns_c.has_non_text %}
100
+ {%- if ns_c.text_parts | length > 0 %}
101
+ {%- set ns_c.chunks = ns_c.chunks + [{'type': 'text', 'text': ns_c.text_parts | join('\n\n')}] %}
102
+ {%- endif %}
103
+ {%- set merged_content = ns_c.chunks %}
104
+ {%- else %}
105
+ {%- set merged_content = ns_c.text_parts | join('\n\n') %}
106
+ {%- endif %}
107
+ {%- if ns_c.tool_calls | length > 0 %}
108
+ {%- set ns_agg.messages = ns_agg.messages + [{'role': ns_agg.current_role, 'content': merged_content, 'tool_calls': ns_c.tool_calls}] %}
109
+ {%- else %}
110
+ {%- set ns_agg.messages = ns_agg.messages + [{'role': ns_agg.current_role, 'content': merged_content}] %}
111
+ {%- endif %}
112
+ {%- endif %}
113
+ {%- if message['role'] != '__sentinel__' %}
114
+ {%- set ns_agg.current_group = [message] %}
115
+ {%- set ns_agg.current_role = message['role'] %}
116
+ {%- endif %}
117
+ {%- else %}
118
+ {%- set ns_agg.current_group = ns_agg.current_group + [message] %}
119
+ {%- endif %}
120
+ {%- endfor %}
121
+ {%- set loop_messages = ns_agg.messages %}
122
+
123
+ {#- Validates message ordering. #}
124
+ {%- set ns = namespace(available_tools_emitted=false) %}
125
+ {%- if loop_messages | length > 0 and loop_messages[0]['role'] not in ['user', 'system'] %}
126
+ {{- raise_exception('Conversation must start with a user or system message, got ' + loop_messages[0]['role'] + '.') }}
127
+ {%- endif %}
128
+ {%- set ns_order = namespace(previous_role=none) %}
129
+ {%- for message in loop_messages %}
130
+ {%- set current_role = message['role'] %}
131
+ {%- if ns_order.previous_role is not none %}
132
+ {%- if ns_order.previous_role == 'system' %}
133
+ {%- if current_role not in ['user', 'assistant', 'system'] %}
134
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
135
+ {%- endif %}
136
+ {%- elif ns_order.previous_role == 'user' %}
137
+ {%- if current_role not in ['assistant', 'system', 'user'] %}
138
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
139
+ {%- endif %}
140
+ {%- elif ns_order.previous_role == 'assistant' %}
141
+ {%- if current_role not in ['assistant', 'user', 'tool'] %}
142
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
143
+ {%- endif %}
144
+ {%- elif ns_order.previous_role == 'tool' %}
145
+ {%- if current_role not in ['assistant', 'tool', 'user'] %}
146
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
147
+ {%- endif %}
148
+ {%- endif %}
149
+ {%- endif %}
150
+ {%- set ns_order.previous_role = current_role %}
151
+ {%- endfor %}
152
+
153
+ {#- Handle conversation messages. #}
154
+ {%- for message in loop_messages %}
155
+ {#- User messages supports text, image and image_url content. #}
156
+ {%- if message['role'] == 'user' %}
157
+ {%- if not ns.available_tools_emitted %}
158
+ {{- available_tools }}
159
+ {%- set ns.available_tools_emitted = true %}
160
+ {%- endif %}
161
+ {%- if message['content'] is not string and message['content'] %}
162
+ {#- When content has exactly one image and one text block, put image first. #}
163
+ {%- if message['content'] | length == 2 and message['content'][0]['type'] == 'text' and message['content'][1]['type'] in ['image', 'image_url'] %}
164
+ {%- set blocks = [message['content'][1], message['content'][0]] %}
165
+ {%- else %}
166
+ {%- set blocks = message['content'] %}
167
+ {%- endif %}
168
+ {%- set user_content = blocks %}
169
+ {%- else %}
170
+ {%- set user_content = message['content'] %}
171
+ {%- endif %}
172
+ {{- '[INST]' -}}
173
+ {{- render_content(user_content, 'user message content', supported_types_desc='text, image and image_url', support_images=true) -}}
174
+ {{- '[/INST]' }}
175
+
176
+ {#- Assistant messages supports text content. #}
177
+ {%- elif message['role'] == 'assistant' %}
178
+ {%- if (message['content'] is none or message['content'] == '' or message['content']|length == 0) and (message['tool_calls'] is not defined or message['tool_calls'] is none or message['tool_calls']|length == 0) %}
179
+ {{- raise_exception('Assistant message must have a string or a list of chunks in content or a list of tool calls.') }}
180
+ {%- endif %}
181
+
182
+ {%- if message['content'] %}
183
+ {{- render_content(message['content'], 'assistant message contents', supported_types_desc='text', support_images=false) -}}
184
+ {%- endif %}
185
+
186
+ {%- if message['tool_calls'] is defined and message['tool_calls'] is not none and message['tool_calls']|length > 0 %}
187
+ {%- for tool in message['tool_calls'] %}
188
+ {{- '[TOOL_CALLS]' }}
189
+ {%- set name = tool['function']['name'] %}
190
+ {%- set arguments = tool['function']['arguments'] %}
191
+ {%- if arguments is not string %}
192
+ {%- set arguments = arguments|tojson|safe %}
193
+ {%- elif arguments == '' %}
194
+ {%- set arguments = '{}' %}
195
+ {%- endif %}
196
+ {{- name + '[ARGS]' + arguments }}
197
+ {%- endfor %}
198
+ {%- endif %}
199
+
200
+ {{- eos_token }}
201
+
202
+ {#- Tool messages only supports text content. #}
203
+ {%- elif message['role'] == 'tool' %}
204
+ {{- '[TOOL_RESULTS]' + message['content']|string + '[/TOOL_RESULTS]' }}
205
+
206
+ {#- System messages. #}
207
+ {%- elif message['role'] == 'system' %}
208
+ {{- '[SYSTEM_PROMPT]' -}}
209
+ {{- render_content(message['content'], 'system message contents', supported_types_desc='text', support_images=false) -}}
210
+ {{- '[/SYSTEM_PROMPT]' -}}
211
+
212
+ {#- Raise exception for unsupported roles. #}
213
+ {%- else %}
214
+ {{- raise_exception('Only user, assistant, system and tool roles are supported, got ' + message['role'] + '.') }}
215
+ {%- endif %}
216
+ {%- endfor %}
config.json ADDED
@@ -0,0 +1,2865 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Mistral3ForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "eos_token_id": 2,
7
+ "image_token_index": 10,
8
+ "model_type": "mistral3",
9
+ "multimodal_projector_bias": false,
10
+ "projector_hidden_act": "gelu",
11
+ "quantization": {
12
+ "group_size": 32,
13
+ "bits": 4,
14
+ "mode": "affine",
15
+ "language_model.model.embed_tokens": {
16
+ "group_size": 64,
17
+ "bits": 8,
18
+ "mode": "affine"
19
+ },
20
+ "language_model.model.layers.0.self_attn.q_proj": {
21
+ "group_size": 64,
22
+ "bits": 4,
23
+ "mode": "affine"
24
+ },
25
+ "language_model.model.layers.0.self_attn.k_proj": {
26
+ "group_size": 64,
27
+ "bits": 4,
28
+ "mode": "affine"
29
+ },
30
+ "language_model.model.layers.0.self_attn.v_proj": {
31
+ "group_size": 64,
32
+ "bits": 4,
33
+ "mode": "affine"
34
+ },
35
+ "language_model.model.layers.0.self_attn.o_proj": {
36
+ "group_size": 64,
37
+ "bits": 4,
38
+ "mode": "affine"
39
+ },
40
+ "language_model.model.layers.0.mlp.gate_proj": {
41
+ "group_size": 64,
42
+ "bits": 4,
43
+ "mode": "affine"
44
+ },
45
+ "language_model.model.layers.0.mlp.down_proj": {
46
+ "group_size": 64,
47
+ "bits": 4,
48
+ "mode": "affine"
49
+ },
50
+ "language_model.model.layers.0.mlp.up_proj": {
51
+ "group_size": 64,
52
+ "bits": 4,
53
+ "mode": "affine"
54
+ },
55
+ "language_model.model.layers.1.self_attn.q_proj": {
56
+ "group_size": 64,
57
+ "bits": 4,
58
+ "mode": "affine"
59
+ },
60
+ "language_model.model.layers.1.self_attn.k_proj": {
61
+ "group_size": 64,
62
+ "bits": 4,
63
+ "mode": "affine"
64
+ },
65
+ "language_model.model.layers.1.self_attn.v_proj": {
66
+ "group_size": 64,
67
+ "bits": 4,
68
+ "mode": "affine"
69
+ },
70
+ "language_model.model.layers.1.self_attn.o_proj": {
71
+ "group_size": 64,
72
+ "bits": 4,
73
+ "mode": "affine"
74
+ },
75
+ "language_model.model.layers.1.mlp.gate_proj": {
76
+ "group_size": 64,
77
+ "bits": 4,
78
+ "mode": "affine"
79
+ },
80
+ "language_model.model.layers.1.mlp.down_proj": {
81
+ "group_size": 64,
82
+ "bits": 4,
83
+ "mode": "affine"
84
+ },
85
+ "language_model.model.layers.1.mlp.up_proj": {
86
+ "group_size": 64,
87
+ "bits": 4,
88
+ "mode": "affine"
89
+ },
90
+ "language_model.model.layers.2.self_attn.q_proj": {
91
+ "group_size": 64,
92
+ "bits": 4,
93
+ "mode": "affine"
94
+ },
95
+ "language_model.model.layers.2.self_attn.k_proj": {
96
+ "group_size": 64,
97
+ "bits": 4,
98
+ "mode": "affine"
99
+ },
100
+ "language_model.model.layers.2.self_attn.v_proj": {
101
+ "group_size": 64,
102
+ "bits": 4,
103
+ "mode": "affine"
104
+ },
105
+ "language_model.model.layers.2.self_attn.o_proj": {
106
+ "group_size": 64,
107
+ "bits": 4,
108
+ "mode": "affine"
109
+ },
110
+ "language_model.model.layers.2.mlp.gate_proj": {
111
+ "group_size": 64,
112
+ "bits": 4,
113
+ "mode": "affine"
114
+ },
115
+ "language_model.model.layers.2.mlp.down_proj": {
116
+ "group_size": 64,
117
+ "bits": 4,
118
+ "mode": "affine"
119
+ },
120
+ "language_model.model.layers.2.mlp.up_proj": {
121
+ "group_size": 64,
122
+ "bits": 4,
123
+ "mode": "affine"
124
+ },
125
+ "language_model.model.layers.3.self_attn.q_proj": {
126
+ "group_size": 64,
127
+ "bits": 4,
128
+ "mode": "affine"
129
+ },
130
+ "language_model.model.layers.3.self_attn.k_proj": {
131
+ "group_size": 64,
132
+ "bits": 4,
133
+ "mode": "affine"
134
+ },
135
+ "language_model.model.layers.3.self_attn.v_proj": {
136
+ "group_size": 64,
137
+ "bits": 4,
138
+ "mode": "affine"
139
+ },
140
+ "language_model.model.layers.3.self_attn.o_proj": {
141
+ "group_size": 64,
142
+ "bits": 4,
143
+ "mode": "affine"
144
+ },
145
+ "language_model.model.layers.3.mlp.gate_proj": {
146
+ "group_size": 64,
147
+ "bits": 4,
148
+ "mode": "affine"
149
+ },
150
+ "language_model.model.layers.3.mlp.down_proj": {
151
+ "group_size": 64,
152
+ "bits": 4,
153
+ "mode": "affine"
154
+ },
155
+ "language_model.model.layers.3.mlp.up_proj": {
156
+ "group_size": 64,
157
+ "bits": 4,
158
+ "mode": "affine"
159
+ },
160
+ "language_model.model.layers.4.self_attn.q_proj": {
161
+ "group_size": 64,
162
+ "bits": 4,
163
+ "mode": "affine"
164
+ },
165
+ "language_model.model.layers.4.self_attn.k_proj": {
166
+ "group_size": 64,
167
+ "bits": 4,
168
+ "mode": "affine"
169
+ },
170
+ "language_model.model.layers.4.self_attn.v_proj": {
171
+ "group_size": 64,
172
+ "bits": 4,
173
+ "mode": "affine"
174
+ },
175
+ "language_model.model.layers.4.self_attn.o_proj": {
176
+ "group_size": 64,
177
+ "bits": 4,
178
+ "mode": "affine"
179
+ },
180
+ "language_model.model.layers.4.mlp.gate_proj": {
181
+ "group_size": 64,
182
+ "bits": 4,
183
+ "mode": "affine"
184
+ },
185
+ "language_model.model.layers.4.mlp.down_proj": {
186
+ "group_size": 64,
187
+ "bits": 4,
188
+ "mode": "affine"
189
+ },
190
+ "language_model.model.layers.4.mlp.up_proj": {
191
+ "group_size": 64,
192
+ "bits": 4,
193
+ "mode": "affine"
194
+ },
195
+ "language_model.model.layers.5.self_attn.q_proj": {
196
+ "group_size": 64,
197
+ "bits": 4,
198
+ "mode": "affine"
199
+ },
200
+ "language_model.model.layers.5.self_attn.k_proj": {
201
+ "group_size": 64,
202
+ "bits": 4,
203
+ "mode": "affine"
204
+ },
205
+ "language_model.model.layers.5.self_attn.v_proj": {
206
+ "group_size": 64,
207
+ "bits": 4,
208
+ "mode": "affine"
209
+ },
210
+ "language_model.model.layers.5.self_attn.o_proj": {
211
+ "group_size": 64,
212
+ "bits": 4,
213
+ "mode": "affine"
214
+ },
215
+ "language_model.model.layers.5.mlp.gate_proj": {
216
+ "group_size": 64,
217
+ "bits": 4,
218
+ "mode": "affine"
219
+ },
220
+ "language_model.model.layers.5.mlp.down_proj": {
221
+ "group_size": 64,
222
+ "bits": 4,
223
+ "mode": "affine"
224
+ },
225
+ "language_model.model.layers.5.mlp.up_proj": {
226
+ "group_size": 64,
227
+ "bits": 4,
228
+ "mode": "affine"
229
+ },
230
+ "language_model.model.layers.6.self_attn.q_proj": {
231
+ "group_size": 64,
232
+ "bits": 4,
233
+ "mode": "affine"
234
+ },
235
+ "language_model.model.layers.6.self_attn.k_proj": {
236
+ "group_size": 64,
237
+ "bits": 4,
238
+ "mode": "affine"
239
+ },
240
+ "language_model.model.layers.6.self_attn.v_proj": {
241
+ "group_size": 64,
242
+ "bits": 4,
243
+ "mode": "affine"
244
+ },
245
+ "language_model.model.layers.6.self_attn.o_proj": {
246
+ "group_size": 64,
247
+ "bits": 4,
248
+ "mode": "affine"
249
+ },
250
+ "language_model.model.layers.6.mlp.gate_proj": {
251
+ "group_size": 64,
252
+ "bits": 4,
253
+ "mode": "affine"
254
+ },
255
+ "language_model.model.layers.6.mlp.down_proj": {
256
+ "group_size": 64,
257
+ "bits": 4,
258
+ "mode": "affine"
259
+ },
260
+ "language_model.model.layers.6.mlp.up_proj": {
261
+ "group_size": 64,
262
+ "bits": 4,
263
+ "mode": "affine"
264
+ },
265
+ "language_model.model.layers.7.self_attn.q_proj": {
266
+ "group_size": 64,
267
+ "bits": 4,
268
+ "mode": "affine"
269
+ },
270
+ "language_model.model.layers.7.self_attn.k_proj": {
271
+ "group_size": 64,
272
+ "bits": 4,
273
+ "mode": "affine"
274
+ },
275
+ "language_model.model.layers.7.self_attn.v_proj": {
276
+ "group_size": 64,
277
+ "bits": 4,
278
+ "mode": "affine"
279
+ },
280
+ "language_model.model.layers.7.self_attn.o_proj": {
281
+ "group_size": 64,
282
+ "bits": 4,
283
+ "mode": "affine"
284
+ },
285
+ "language_model.model.layers.7.mlp.gate_proj": {
286
+ "group_size": 64,
287
+ "bits": 4,
288
+ "mode": "affine"
289
+ },
290
+ "language_model.model.layers.7.mlp.down_proj": {
291
+ "group_size": 64,
292
+ "bits": 4,
293
+ "mode": "affine"
294
+ },
295
+ "language_model.model.layers.7.mlp.up_proj": {
296
+ "group_size": 64,
297
+ "bits": 4,
298
+ "mode": "affine"
299
+ },
300
+ "language_model.model.layers.8.self_attn.q_proj": {
301
+ "group_size": 64,
302
+ "bits": 4,
303
+ "mode": "affine"
304
+ },
305
+ "language_model.model.layers.8.self_attn.k_proj": {
306
+ "group_size": 64,
307
+ "bits": 4,
308
+ "mode": "affine"
309
+ },
310
+ "language_model.model.layers.8.self_attn.v_proj": {
311
+ "group_size": 64,
312
+ "bits": 4,
313
+ "mode": "affine"
314
+ },
315
+ "language_model.model.layers.8.self_attn.o_proj": {
316
+ "group_size": 64,
317
+ "bits": 4,
318
+ "mode": "affine"
319
+ },
320
+ "language_model.model.layers.8.mlp.gate_proj": {
321
+ "group_size": 64,
322
+ "bits": 4,
323
+ "mode": "affine"
324
+ },
325
+ "language_model.model.layers.8.mlp.down_proj": {
326
+ "group_size": 64,
327
+ "bits": 4,
328
+ "mode": "affine"
329
+ },
330
+ "language_model.model.layers.8.mlp.up_proj": {
331
+ "group_size": 64,
332
+ "bits": 4,
333
+ "mode": "affine"
334
+ },
335
+ "language_model.model.layers.9.self_attn.q_proj": {
336
+ "group_size": 64,
337
+ "bits": 4,
338
+ "mode": "affine"
339
+ },
340
+ "language_model.model.layers.9.self_attn.k_proj": {
341
+ "group_size": 64,
342
+ "bits": 4,
343
+ "mode": "affine"
344
+ },
345
+ "language_model.model.layers.9.self_attn.v_proj": {
346
+ "group_size": 64,
347
+ "bits": 4,
348
+ "mode": "affine"
349
+ },
350
+ "language_model.model.layers.9.self_attn.o_proj": {
351
+ "group_size": 64,
352
+ "bits": 4,
353
+ "mode": "affine"
354
+ },
355
+ "language_model.model.layers.9.mlp.gate_proj": {
356
+ "group_size": 64,
357
+ "bits": 4,
358
+ "mode": "affine"
359
+ },
360
+ "language_model.model.layers.9.mlp.down_proj": {
361
+ "group_size": 64,
362
+ "bits": 4,
363
+ "mode": "affine"
364
+ },
365
+ "language_model.model.layers.9.mlp.up_proj": {
366
+ "group_size": 64,
367
+ "bits": 4,
368
+ "mode": "affine"
369
+ },
370
+ "language_model.model.layers.10.self_attn.q_proj": {
371
+ "group_size": 64,
372
+ "bits": 4,
373
+ "mode": "affine"
374
+ },
375
+ "language_model.model.layers.10.self_attn.k_proj": {
376
+ "group_size": 64,
377
+ "bits": 4,
378
+ "mode": "affine"
379
+ },
380
+ "language_model.model.layers.10.self_attn.v_proj": {
381
+ "group_size": 64,
382
+ "bits": 4,
383
+ "mode": "affine"
384
+ },
385
+ "language_model.model.layers.10.self_attn.o_proj": {
386
+ "group_size": 64,
387
+ "bits": 4,
388
+ "mode": "affine"
389
+ },
390
+ "language_model.model.layers.10.mlp.gate_proj": {
391
+ "group_size": 64,
392
+ "bits": 4,
393
+ "mode": "affine"
394
+ },
395
+ "language_model.model.layers.10.mlp.down_proj": {
396
+ "group_size": 64,
397
+ "bits": 4,
398
+ "mode": "affine"
399
+ },
400
+ "language_model.model.layers.10.mlp.up_proj": {
401
+ "group_size": 64,
402
+ "bits": 4,
403
+ "mode": "affine"
404
+ },
405
+ "language_model.model.layers.11.self_attn.q_proj": {
406
+ "group_size": 64,
407
+ "bits": 4,
408
+ "mode": "affine"
409
+ },
410
+ "language_model.model.layers.11.self_attn.k_proj": {
411
+ "group_size": 64,
412
+ "bits": 4,
413
+ "mode": "affine"
414
+ },
415
+ "language_model.model.layers.11.self_attn.v_proj": {
416
+ "group_size": 64,
417
+ "bits": 4,
418
+ "mode": "affine"
419
+ },
420
+ "language_model.model.layers.11.self_attn.o_proj": {
421
+ "group_size": 64,
422
+ "bits": 4,
423
+ "mode": "affine"
424
+ },
425
+ "language_model.model.layers.11.mlp.gate_proj": {
426
+ "group_size": 64,
427
+ "bits": 4,
428
+ "mode": "affine"
429
+ },
430
+ "language_model.model.layers.11.mlp.down_proj": {
431
+ "group_size": 64,
432
+ "bits": 4,
433
+ "mode": "affine"
434
+ },
435
+ "language_model.model.layers.11.mlp.up_proj": {
436
+ "group_size": 64,
437
+ "bits": 4,
438
+ "mode": "affine"
439
+ },
440
+ "language_model.model.layers.12.self_attn.q_proj": {
441
+ "group_size": 64,
442
+ "bits": 4,
443
+ "mode": "affine"
444
+ },
445
+ "language_model.model.layers.12.self_attn.k_proj": {
446
+ "group_size": 64,
447
+ "bits": 4,
448
+ "mode": "affine"
449
+ },
450
+ "language_model.model.layers.12.self_attn.v_proj": {
451
+ "group_size": 64,
452
+ "bits": 4,
453
+ "mode": "affine"
454
+ },
455
+ "language_model.model.layers.12.self_attn.o_proj": {
456
+ "group_size": 64,
457
+ "bits": 4,
458
+ "mode": "affine"
459
+ },
460
+ "language_model.model.layers.12.mlp.gate_proj": {
461
+ "group_size": 64,
462
+ "bits": 4,
463
+ "mode": "affine"
464
+ },
465
+ "language_model.model.layers.12.mlp.down_proj": {
466
+ "group_size": 64,
467
+ "bits": 4,
468
+ "mode": "affine"
469
+ },
470
+ "language_model.model.layers.12.mlp.up_proj": {
471
+ "group_size": 64,
472
+ "bits": 4,
473
+ "mode": "affine"
474
+ },
475
+ "language_model.model.layers.13.self_attn.q_proj": {
476
+ "group_size": 64,
477
+ "bits": 4,
478
+ "mode": "affine"
479
+ },
480
+ "language_model.model.layers.13.self_attn.k_proj": {
481
+ "group_size": 64,
482
+ "bits": 4,
483
+ "mode": "affine"
484
+ },
485
+ "language_model.model.layers.13.self_attn.v_proj": {
486
+ "group_size": 64,
487
+ "bits": 4,
488
+ "mode": "affine"
489
+ },
490
+ "language_model.model.layers.13.self_attn.o_proj": {
491
+ "group_size": 64,
492
+ "bits": 4,
493
+ "mode": "affine"
494
+ },
495
+ "language_model.model.layers.13.mlp.gate_proj": {
496
+ "group_size": 64,
497
+ "bits": 4,
498
+ "mode": "affine"
499
+ },
500
+ "language_model.model.layers.13.mlp.down_proj": {
501
+ "group_size": 64,
502
+ "bits": 4,
503
+ "mode": "affine"
504
+ },
505
+ "language_model.model.layers.13.mlp.up_proj": {
506
+ "group_size": 64,
507
+ "bits": 4,
508
+ "mode": "affine"
509
+ },
510
+ "language_model.model.layers.14.self_attn.q_proj": {
511
+ "group_size": 64,
512
+ "bits": 4,
513
+ "mode": "affine"
514
+ },
515
+ "language_model.model.layers.14.self_attn.k_proj": {
516
+ "group_size": 64,
517
+ "bits": 4,
518
+ "mode": "affine"
519
+ },
520
+ "language_model.model.layers.14.self_attn.v_proj": {
521
+ "group_size": 64,
522
+ "bits": 4,
523
+ "mode": "affine"
524
+ },
525
+ "language_model.model.layers.14.self_attn.o_proj": {
526
+ "group_size": 64,
527
+ "bits": 4,
528
+ "mode": "affine"
529
+ },
530
+ "language_model.model.layers.14.mlp.gate_proj": {
531
+ "group_size": 64,
532
+ "bits": 4,
533
+ "mode": "affine"
534
+ },
535
+ "language_model.model.layers.14.mlp.down_proj": {
536
+ "group_size": 64,
537
+ "bits": 4,
538
+ "mode": "affine"
539
+ },
540
+ "language_model.model.layers.14.mlp.up_proj": {
541
+ "group_size": 64,
542
+ "bits": 4,
543
+ "mode": "affine"
544
+ },
545
+ "language_model.model.layers.15.self_attn.q_proj": {
546
+ "group_size": 64,
547
+ "bits": 4,
548
+ "mode": "affine"
549
+ },
550
+ "language_model.model.layers.15.self_attn.k_proj": {
551
+ "group_size": 64,
552
+ "bits": 4,
553
+ "mode": "affine"
554
+ },
555
+ "language_model.model.layers.15.self_attn.v_proj": {
556
+ "group_size": 64,
557
+ "bits": 4,
558
+ "mode": "affine"
559
+ },
560
+ "language_model.model.layers.15.self_attn.o_proj": {
561
+ "group_size": 64,
562
+ "bits": 4,
563
+ "mode": "affine"
564
+ },
565
+ "language_model.model.layers.15.mlp.gate_proj": {
566
+ "group_size": 64,
567
+ "bits": 4,
568
+ "mode": "affine"
569
+ },
570
+ "language_model.model.layers.15.mlp.down_proj": {
571
+ "group_size": 64,
572
+ "bits": 4,
573
+ "mode": "affine"
574
+ },
575
+ "language_model.model.layers.15.mlp.up_proj": {
576
+ "group_size": 64,
577
+ "bits": 4,
578
+ "mode": "affine"
579
+ },
580
+ "language_model.model.layers.16.self_attn.q_proj": {
581
+ "group_size": 64,
582
+ "bits": 4,
583
+ "mode": "affine"
584
+ },
585
+ "language_model.model.layers.16.self_attn.k_proj": {
586
+ "group_size": 64,
587
+ "bits": 4,
588
+ "mode": "affine"
589
+ },
590
+ "language_model.model.layers.16.self_attn.v_proj": {
591
+ "group_size": 64,
592
+ "bits": 4,
593
+ "mode": "affine"
594
+ },
595
+ "language_model.model.layers.16.self_attn.o_proj": {
596
+ "group_size": 64,
597
+ "bits": 4,
598
+ "mode": "affine"
599
+ },
600
+ "language_model.model.layers.16.mlp.gate_proj": {
601
+ "group_size": 64,
602
+ "bits": 4,
603
+ "mode": "affine"
604
+ },
605
+ "language_model.model.layers.16.mlp.down_proj": {
606
+ "group_size": 64,
607
+ "bits": 4,
608
+ "mode": "affine"
609
+ },
610
+ "language_model.model.layers.16.mlp.up_proj": {
611
+ "group_size": 64,
612
+ "bits": 4,
613
+ "mode": "affine"
614
+ },
615
+ "language_model.model.layers.17.self_attn.q_proj": {
616
+ "group_size": 64,
617
+ "bits": 4,
618
+ "mode": "affine"
619
+ },
620
+ "language_model.model.layers.17.self_attn.k_proj": {
621
+ "group_size": 64,
622
+ "bits": 4,
623
+ "mode": "affine"
624
+ },
625
+ "language_model.model.layers.17.self_attn.v_proj": {
626
+ "group_size": 64,
627
+ "bits": 4,
628
+ "mode": "affine"
629
+ },
630
+ "language_model.model.layers.17.self_attn.o_proj": {
631
+ "group_size": 64,
632
+ "bits": 4,
633
+ "mode": "affine"
634
+ },
635
+ "language_model.model.layers.17.mlp.gate_proj": {
636
+ "group_size": 64,
637
+ "bits": 4,
638
+ "mode": "affine"
639
+ },
640
+ "language_model.model.layers.17.mlp.down_proj": {
641
+ "group_size": 64,
642
+ "bits": 4,
643
+ "mode": "affine"
644
+ },
645
+ "language_model.model.layers.17.mlp.up_proj": {
646
+ "group_size": 64,
647
+ "bits": 4,
648
+ "mode": "affine"
649
+ },
650
+ "language_model.model.layers.18.self_attn.q_proj": {
651
+ "group_size": 64,
652
+ "bits": 4,
653
+ "mode": "affine"
654
+ },
655
+ "language_model.model.layers.18.self_attn.k_proj": {
656
+ "group_size": 64,
657
+ "bits": 4,
658
+ "mode": "affine"
659
+ },
660
+ "language_model.model.layers.18.self_attn.v_proj": {
661
+ "group_size": 64,
662
+ "bits": 4,
663
+ "mode": "affine"
664
+ },
665
+ "language_model.model.layers.18.self_attn.o_proj": {
666
+ "group_size": 64,
667
+ "bits": 4,
668
+ "mode": "affine"
669
+ },
670
+ "language_model.model.layers.18.mlp.gate_proj": {
671
+ "group_size": 64,
672
+ "bits": 4,
673
+ "mode": "affine"
674
+ },
675
+ "language_model.model.layers.18.mlp.down_proj": {
676
+ "group_size": 64,
677
+ "bits": 4,
678
+ "mode": "affine"
679
+ },
680
+ "language_model.model.layers.18.mlp.up_proj": {
681
+ "group_size": 64,
682
+ "bits": 4,
683
+ "mode": "affine"
684
+ },
685
+ "language_model.model.layers.19.self_attn.q_proj": {
686
+ "group_size": 64,
687
+ "bits": 4,
688
+ "mode": "affine"
689
+ },
690
+ "language_model.model.layers.19.self_attn.k_proj": {
691
+ "group_size": 64,
692
+ "bits": 4,
693
+ "mode": "affine"
694
+ },
695
+ "language_model.model.layers.19.self_attn.v_proj": {
696
+ "group_size": 64,
697
+ "bits": 4,
698
+ "mode": "affine"
699
+ },
700
+ "language_model.model.layers.19.self_attn.o_proj": {
701
+ "group_size": 64,
702
+ "bits": 4,
703
+ "mode": "affine"
704
+ },
705
+ "language_model.model.layers.19.mlp.gate_proj": {
706
+ "group_size": 64,
707
+ "bits": 4,
708
+ "mode": "affine"
709
+ },
710
+ "language_model.model.layers.19.mlp.down_proj": {
711
+ "group_size": 64,
712
+ "bits": 4,
713
+ "mode": "affine"
714
+ },
715
+ "language_model.model.layers.19.mlp.up_proj": {
716
+ "group_size": 64,
717
+ "bits": 4,
718
+ "mode": "affine"
719
+ },
720
+ "language_model.model.layers.20.self_attn.q_proj": {
721
+ "group_size": 64,
722
+ "bits": 4,
723
+ "mode": "affine"
724
+ },
725
+ "language_model.model.layers.20.self_attn.k_proj": {
726
+ "group_size": 64,
727
+ "bits": 4,
728
+ "mode": "affine"
729
+ },
730
+ "language_model.model.layers.20.self_attn.v_proj": {
731
+ "group_size": 64,
732
+ "bits": 4,
733
+ "mode": "affine"
734
+ },
735
+ "language_model.model.layers.20.self_attn.o_proj": {
736
+ "group_size": 64,
737
+ "bits": 4,
738
+ "mode": "affine"
739
+ },
740
+ "language_model.model.layers.20.mlp.gate_proj": {
741
+ "group_size": 64,
742
+ "bits": 4,
743
+ "mode": "affine"
744
+ },
745
+ "language_model.model.layers.20.mlp.down_proj": {
746
+ "group_size": 64,
747
+ "bits": 4,
748
+ "mode": "affine"
749
+ },
750
+ "language_model.model.layers.20.mlp.up_proj": {
751
+ "group_size": 64,
752
+ "bits": 4,
753
+ "mode": "affine"
754
+ },
755
+ "language_model.model.layers.21.self_attn.q_proj": {
756
+ "group_size": 64,
757
+ "bits": 4,
758
+ "mode": "affine"
759
+ },
760
+ "language_model.model.layers.21.self_attn.k_proj": {
761
+ "group_size": 64,
762
+ "bits": 4,
763
+ "mode": "affine"
764
+ },
765
+ "language_model.model.layers.21.self_attn.v_proj": {
766
+ "group_size": 64,
767
+ "bits": 4,
768
+ "mode": "affine"
769
+ },
770
+ "language_model.model.layers.21.self_attn.o_proj": {
771
+ "group_size": 64,
772
+ "bits": 4,
773
+ "mode": "affine"
774
+ },
775
+ "language_model.model.layers.21.mlp.gate_proj": {
776
+ "group_size": 64,
777
+ "bits": 4,
778
+ "mode": "affine"
779
+ },
780
+ "language_model.model.layers.21.mlp.down_proj": {
781
+ "group_size": 64,
782
+ "bits": 4,
783
+ "mode": "affine"
784
+ },
785
+ "language_model.model.layers.21.mlp.up_proj": {
786
+ "group_size": 64,
787
+ "bits": 4,
788
+ "mode": "affine"
789
+ },
790
+ "language_model.model.layers.22.self_attn.q_proj": {
791
+ "group_size": 64,
792
+ "bits": 4,
793
+ "mode": "affine"
794
+ },
795
+ "language_model.model.layers.22.self_attn.k_proj": {
796
+ "group_size": 64,
797
+ "bits": 4,
798
+ "mode": "affine"
799
+ },
800
+ "language_model.model.layers.22.self_attn.v_proj": {
801
+ "group_size": 64,
802
+ "bits": 4,
803
+ "mode": "affine"
804
+ },
805
+ "language_model.model.layers.22.self_attn.o_proj": {
806
+ "group_size": 64,
807
+ "bits": 4,
808
+ "mode": "affine"
809
+ },
810
+ "language_model.model.layers.22.mlp.gate_proj": {
811
+ "group_size": 64,
812
+ "bits": 4,
813
+ "mode": "affine"
814
+ },
815
+ "language_model.model.layers.22.mlp.down_proj": {
816
+ "group_size": 64,
817
+ "bits": 4,
818
+ "mode": "affine"
819
+ },
820
+ "language_model.model.layers.22.mlp.up_proj": {
821
+ "group_size": 64,
822
+ "bits": 4,
823
+ "mode": "affine"
824
+ },
825
+ "language_model.model.layers.23.self_attn.q_proj": {
826
+ "group_size": 64,
827
+ "bits": 4,
828
+ "mode": "affine"
829
+ },
830
+ "language_model.model.layers.23.self_attn.k_proj": {
831
+ "group_size": 64,
832
+ "bits": 4,
833
+ "mode": "affine"
834
+ },
835
+ "language_model.model.layers.23.self_attn.v_proj": {
836
+ "group_size": 64,
837
+ "bits": 4,
838
+ "mode": "affine"
839
+ },
840
+ "language_model.model.layers.23.self_attn.o_proj": {
841
+ "group_size": 64,
842
+ "bits": 4,
843
+ "mode": "affine"
844
+ },
845
+ "language_model.model.layers.23.mlp.gate_proj": {
846
+ "group_size": 64,
847
+ "bits": 4,
848
+ "mode": "affine"
849
+ },
850
+ "language_model.model.layers.23.mlp.down_proj": {
851
+ "group_size": 64,
852
+ "bits": 4,
853
+ "mode": "affine"
854
+ },
855
+ "language_model.model.layers.23.mlp.up_proj": {
856
+ "group_size": 64,
857
+ "bits": 4,
858
+ "mode": "affine"
859
+ },
860
+ "language_model.model.layers.24.self_attn.q_proj": {
861
+ "group_size": 64,
862
+ "bits": 4,
863
+ "mode": "affine"
864
+ },
865
+ "language_model.model.layers.24.self_attn.k_proj": {
866
+ "group_size": 64,
867
+ "bits": 4,
868
+ "mode": "affine"
869
+ },
870
+ "language_model.model.layers.24.self_attn.v_proj": {
871
+ "group_size": 64,
872
+ "bits": 4,
873
+ "mode": "affine"
874
+ },
875
+ "language_model.model.layers.24.self_attn.o_proj": {
876
+ "group_size": 64,
877
+ "bits": 4,
878
+ "mode": "affine"
879
+ },
880
+ "language_model.model.layers.24.mlp.gate_proj": {
881
+ "group_size": 64,
882
+ "bits": 4,
883
+ "mode": "affine"
884
+ },
885
+ "language_model.model.layers.24.mlp.down_proj": {
886
+ "group_size": 64,
887
+ "bits": 4,
888
+ "mode": "affine"
889
+ },
890
+ "language_model.model.layers.24.mlp.up_proj": {
891
+ "group_size": 64,
892
+ "bits": 4,
893
+ "mode": "affine"
894
+ },
895
+ "language_model.model.layers.25.self_attn.q_proj": {
896
+ "group_size": 64,
897
+ "bits": 4,
898
+ "mode": "affine"
899
+ },
900
+ "language_model.model.layers.25.self_attn.k_proj": {
901
+ "group_size": 64,
902
+ "bits": 4,
903
+ "mode": "affine"
904
+ },
905
+ "language_model.model.layers.25.self_attn.v_proj": {
906
+ "group_size": 64,
907
+ "bits": 4,
908
+ "mode": "affine"
909
+ },
910
+ "language_model.model.layers.25.self_attn.o_proj": {
911
+ "group_size": 64,
912
+ "bits": 4,
913
+ "mode": "affine"
914
+ },
915
+ "language_model.model.layers.25.mlp.gate_proj": {
916
+ "group_size": 64,
917
+ "bits": 4,
918
+ "mode": "affine"
919
+ },
920
+ "language_model.model.layers.25.mlp.down_proj": {
921
+ "group_size": 64,
922
+ "bits": 4,
923
+ "mode": "affine"
924
+ },
925
+ "language_model.model.layers.25.mlp.up_proj": {
926
+ "group_size": 64,
927
+ "bits": 4,
928
+ "mode": "affine"
929
+ },
930
+ "language_model.model.layers.26.self_attn.q_proj": {
931
+ "group_size": 64,
932
+ "bits": 4,
933
+ "mode": "affine"
934
+ },
935
+ "language_model.model.layers.26.self_attn.k_proj": {
936
+ "group_size": 64,
937
+ "bits": 4,
938
+ "mode": "affine"
939
+ },
940
+ "language_model.model.layers.26.self_attn.v_proj": {
941
+ "group_size": 64,
942
+ "bits": 4,
943
+ "mode": "affine"
944
+ },
945
+ "language_model.model.layers.26.self_attn.o_proj": {
946
+ "group_size": 64,
947
+ "bits": 4,
948
+ "mode": "affine"
949
+ },
950
+ "language_model.model.layers.26.mlp.gate_proj": {
951
+ "group_size": 64,
952
+ "bits": 4,
953
+ "mode": "affine"
954
+ },
955
+ "language_model.model.layers.26.mlp.down_proj": {
956
+ "group_size": 64,
957
+ "bits": 4,
958
+ "mode": "affine"
959
+ },
960
+ "language_model.model.layers.26.mlp.up_proj": {
961
+ "group_size": 64,
962
+ "bits": 4,
963
+ "mode": "affine"
964
+ },
965
+ "language_model.model.layers.27.self_attn.q_proj": {
966
+ "group_size": 64,
967
+ "bits": 4,
968
+ "mode": "affine"
969
+ },
970
+ "language_model.model.layers.27.self_attn.k_proj": {
971
+ "group_size": 64,
972
+ "bits": 4,
973
+ "mode": "affine"
974
+ },
975
+ "language_model.model.layers.27.self_attn.v_proj": {
976
+ "group_size": 64,
977
+ "bits": 4,
978
+ "mode": "affine"
979
+ },
980
+ "language_model.model.layers.27.self_attn.o_proj": {
981
+ "group_size": 64,
982
+ "bits": 4,
983
+ "mode": "affine"
984
+ },
985
+ "language_model.model.layers.27.mlp.gate_proj": {
986
+ "group_size": 64,
987
+ "bits": 4,
988
+ "mode": "affine"
989
+ },
990
+ "language_model.model.layers.27.mlp.down_proj": {
991
+ "group_size": 64,
992
+ "bits": 4,
993
+ "mode": "affine"
994
+ },
995
+ "language_model.model.layers.27.mlp.up_proj": {
996
+ "group_size": 64,
997
+ "bits": 4,
998
+ "mode": "affine"
999
+ },
1000
+ "language_model.model.layers.28.self_attn.q_proj": {
1001
+ "group_size": 64,
1002
+ "bits": 4,
1003
+ "mode": "affine"
1004
+ },
1005
+ "language_model.model.layers.28.self_attn.k_proj": {
1006
+ "group_size": 64,
1007
+ "bits": 4,
1008
+ "mode": "affine"
1009
+ },
1010
+ "language_model.model.layers.28.self_attn.v_proj": {
1011
+ "group_size": 64,
1012
+ "bits": 4,
1013
+ "mode": "affine"
1014
+ },
1015
+ "language_model.model.layers.28.self_attn.o_proj": {
1016
+ "group_size": 64,
1017
+ "bits": 4,
1018
+ "mode": "affine"
1019
+ },
1020
+ "language_model.model.layers.28.mlp.gate_proj": {
1021
+ "group_size": 64,
1022
+ "bits": 4,
1023
+ "mode": "affine"
1024
+ },
1025
+ "language_model.model.layers.28.mlp.down_proj": {
1026
+ "group_size": 64,
1027
+ "bits": 4,
1028
+ "mode": "affine"
1029
+ },
1030
+ "language_model.model.layers.28.mlp.up_proj": {
1031
+ "group_size": 64,
1032
+ "bits": 4,
1033
+ "mode": "affine"
1034
+ },
1035
+ "language_model.model.layers.29.self_attn.q_proj": {
1036
+ "group_size": 64,
1037
+ "bits": 4,
1038
+ "mode": "affine"
1039
+ },
1040
+ "language_model.model.layers.29.self_attn.k_proj": {
1041
+ "group_size": 64,
1042
+ "bits": 4,
1043
+ "mode": "affine"
1044
+ },
1045
+ "language_model.model.layers.29.self_attn.v_proj": {
1046
+ "group_size": 64,
1047
+ "bits": 4,
1048
+ "mode": "affine"
1049
+ },
1050
+ "language_model.model.layers.29.self_attn.o_proj": {
1051
+ "group_size": 64,
1052
+ "bits": 4,
1053
+ "mode": "affine"
1054
+ },
1055
+ "language_model.model.layers.29.mlp.gate_proj": {
1056
+ "group_size": 64,
1057
+ "bits": 4,
1058
+ "mode": "affine"
1059
+ },
1060
+ "language_model.model.layers.29.mlp.down_proj": {
1061
+ "group_size": 64,
1062
+ "bits": 4,
1063
+ "mode": "affine"
1064
+ },
1065
+ "language_model.model.layers.29.mlp.up_proj": {
1066
+ "group_size": 64,
1067
+ "bits": 4,
1068
+ "mode": "affine"
1069
+ },
1070
+ "language_model.model.layers.30.self_attn.q_proj": {
1071
+ "group_size": 64,
1072
+ "bits": 4,
1073
+ "mode": "affine"
1074
+ },
1075
+ "language_model.model.layers.30.self_attn.k_proj": {
1076
+ "group_size": 64,
1077
+ "bits": 4,
1078
+ "mode": "affine"
1079
+ },
1080
+ "language_model.model.layers.30.self_attn.v_proj": {
1081
+ "group_size": 64,
1082
+ "bits": 4,
1083
+ "mode": "affine"
1084
+ },
1085
+ "language_model.model.layers.30.self_attn.o_proj": {
1086
+ "group_size": 64,
1087
+ "bits": 4,
1088
+ "mode": "affine"
1089
+ },
1090
+ "language_model.model.layers.30.mlp.gate_proj": {
1091
+ "group_size": 64,
1092
+ "bits": 4,
1093
+ "mode": "affine"
1094
+ },
1095
+ "language_model.model.layers.30.mlp.down_proj": {
1096
+ "group_size": 64,
1097
+ "bits": 4,
1098
+ "mode": "affine"
1099
+ },
1100
+ "language_model.model.layers.30.mlp.up_proj": {
1101
+ "group_size": 64,
1102
+ "bits": 4,
1103
+ "mode": "affine"
1104
+ },
1105
+ "language_model.model.layers.31.self_attn.q_proj": {
1106
+ "group_size": 64,
1107
+ "bits": 4,
1108
+ "mode": "affine"
1109
+ },
1110
+ "language_model.model.layers.31.self_attn.k_proj": {
1111
+ "group_size": 64,
1112
+ "bits": 4,
1113
+ "mode": "affine"
1114
+ },
1115
+ "language_model.model.layers.31.self_attn.v_proj": {
1116
+ "group_size": 64,
1117
+ "bits": 4,
1118
+ "mode": "affine"
1119
+ },
1120
+ "language_model.model.layers.31.self_attn.o_proj": {
1121
+ "group_size": 64,
1122
+ "bits": 4,
1123
+ "mode": "affine"
1124
+ },
1125
+ "language_model.model.layers.31.mlp.gate_proj": {
1126
+ "group_size": 64,
1127
+ "bits": 4,
1128
+ "mode": "affine"
1129
+ },
1130
+ "language_model.model.layers.31.mlp.down_proj": {
1131
+ "group_size": 64,
1132
+ "bits": 4,
1133
+ "mode": "affine"
1134
+ },
1135
+ "language_model.model.layers.31.mlp.up_proj": {
1136
+ "group_size": 64,
1137
+ "bits": 4,
1138
+ "mode": "affine"
1139
+ },
1140
+ "language_model.model.layers.32.self_attn.q_proj": {
1141
+ "group_size": 64,
1142
+ "bits": 4,
1143
+ "mode": "affine"
1144
+ },
1145
+ "language_model.model.layers.32.self_attn.k_proj": {
1146
+ "group_size": 64,
1147
+ "bits": 4,
1148
+ "mode": "affine"
1149
+ },
1150
+ "language_model.model.layers.32.self_attn.v_proj": {
1151
+ "group_size": 64,
1152
+ "bits": 4,
1153
+ "mode": "affine"
1154
+ },
1155
+ "language_model.model.layers.32.self_attn.o_proj": {
1156
+ "group_size": 64,
1157
+ "bits": 4,
1158
+ "mode": "affine"
1159
+ },
1160
+ "language_model.model.layers.32.mlp.gate_proj": {
1161
+ "group_size": 64,
1162
+ "bits": 4,
1163
+ "mode": "affine"
1164
+ },
1165
+ "language_model.model.layers.32.mlp.down_proj": {
1166
+ "group_size": 64,
1167
+ "bits": 4,
1168
+ "mode": "affine"
1169
+ },
1170
+ "language_model.model.layers.32.mlp.up_proj": {
1171
+ "group_size": 64,
1172
+ "bits": 4,
1173
+ "mode": "affine"
1174
+ },
1175
+ "language_model.model.layers.33.self_attn.q_proj": {
1176
+ "group_size": 64,
1177
+ "bits": 4,
1178
+ "mode": "affine"
1179
+ },
1180
+ "language_model.model.layers.33.self_attn.k_proj": {
1181
+ "group_size": 64,
1182
+ "bits": 4,
1183
+ "mode": "affine"
1184
+ },
1185
+ "language_model.model.layers.33.self_attn.v_proj": {
1186
+ "group_size": 64,
1187
+ "bits": 4,
1188
+ "mode": "affine"
1189
+ },
1190
+ "language_model.model.layers.33.self_attn.o_proj": {
1191
+ "group_size": 64,
1192
+ "bits": 4,
1193
+ "mode": "affine"
1194
+ },
1195
+ "language_model.model.layers.33.mlp.gate_proj": {
1196
+ "group_size": 64,
1197
+ "bits": 4,
1198
+ "mode": "affine"
1199
+ },
1200
+ "language_model.model.layers.33.mlp.down_proj": {
1201
+ "group_size": 64,
1202
+ "bits": 4,
1203
+ "mode": "affine"
1204
+ },
1205
+ "language_model.model.layers.33.mlp.up_proj": {
1206
+ "group_size": 64,
1207
+ "bits": 4,
1208
+ "mode": "affine"
1209
+ },
1210
+ "language_model.model.layers.34.self_attn.q_proj": {
1211
+ "group_size": 64,
1212
+ "bits": 4,
1213
+ "mode": "affine"
1214
+ },
1215
+ "language_model.model.layers.34.self_attn.k_proj": {
1216
+ "group_size": 32,
1217
+ "bits": 4,
1218
+ "mode": "affine"
1219
+ },
1220
+ "language_model.model.layers.34.self_attn.v_proj": {
1221
+ "group_size": 32,
1222
+ "bits": 4,
1223
+ "mode": "affine"
1224
+ },
1225
+ "language_model.model.layers.34.self_attn.o_proj": {
1226
+ "group_size": 64,
1227
+ "bits": 4,
1228
+ "mode": "affine"
1229
+ },
1230
+ "language_model.model.layers.34.mlp.gate_proj": {
1231
+ "group_size": 64,
1232
+ "bits": 4,
1233
+ "mode": "affine"
1234
+ },
1235
+ "language_model.model.layers.34.mlp.down_proj": {
1236
+ "group_size": 64,
1237
+ "bits": 4,
1238
+ "mode": "affine"
1239
+ },
1240
+ "language_model.model.layers.34.mlp.up_proj": {
1241
+ "group_size": 64,
1242
+ "bits": 4,
1243
+ "mode": "affine"
1244
+ },
1245
+ "language_model.model.layers.35.self_attn.q_proj": {
1246
+ "group_size": 32,
1247
+ "bits": 4,
1248
+ "mode": "affine"
1249
+ },
1250
+ "language_model.model.layers.35.self_attn.k_proj": {
1251
+ "group_size": 32,
1252
+ "bits": 4,
1253
+ "mode": "affine"
1254
+ },
1255
+ "language_model.model.layers.35.self_attn.v_proj": {
1256
+ "group_size": 32,
1257
+ "bits": 4,
1258
+ "mode": "affine"
1259
+ },
1260
+ "language_model.model.layers.35.self_attn.o_proj": {
1261
+ "group_size": 64,
1262
+ "bits": 4,
1263
+ "mode": "affine"
1264
+ },
1265
+ "language_model.model.layers.35.mlp.gate_proj": {
1266
+ "group_size": 64,
1267
+ "bits": 4,
1268
+ "mode": "affine"
1269
+ },
1270
+ "language_model.model.layers.35.mlp.down_proj": {
1271
+ "group_size": 64,
1272
+ "bits": 4,
1273
+ "mode": "affine"
1274
+ },
1275
+ "language_model.model.layers.35.mlp.up_proj": {
1276
+ "group_size": 64,
1277
+ "bits": 4,
1278
+ "mode": "affine"
1279
+ },
1280
+ "language_model.model.layers.36.self_attn.q_proj": {
1281
+ "group_size": 32,
1282
+ "bits": 4,
1283
+ "mode": "affine"
1284
+ },
1285
+ "language_model.model.layers.36.self_attn.k_proj": {
1286
+ "group_size": 32,
1287
+ "bits": 4,
1288
+ "mode": "affine"
1289
+ },
1290
+ "language_model.model.layers.36.self_attn.v_proj": {
1291
+ "group_size": 32,
1292
+ "bits": 4,
1293
+ "mode": "affine"
1294
+ },
1295
+ "language_model.model.layers.36.self_attn.o_proj": {
1296
+ "group_size": 32,
1297
+ "bits": 4,
1298
+ "mode": "affine"
1299
+ },
1300
+ "language_model.model.layers.36.mlp.gate_proj": {
1301
+ "group_size": 64,
1302
+ "bits": 4,
1303
+ "mode": "affine"
1304
+ },
1305
+ "language_model.model.layers.36.mlp.down_proj": {
1306
+ "group_size": 64,
1307
+ "bits": 4,
1308
+ "mode": "affine"
1309
+ },
1310
+ "language_model.model.layers.36.mlp.up_proj": {
1311
+ "group_size": 64,
1312
+ "bits": 4,
1313
+ "mode": "affine"
1314
+ },
1315
+ "language_model.model.layers.37.self_attn.q_proj": {
1316
+ "group_size": 32,
1317
+ "bits": 4,
1318
+ "mode": "affine"
1319
+ },
1320
+ "language_model.model.layers.37.self_attn.k_proj": {
1321
+ "group_size": 32,
1322
+ "bits": 4,
1323
+ "mode": "affine"
1324
+ },
1325
+ "language_model.model.layers.37.self_attn.v_proj": {
1326
+ "group_size": 32,
1327
+ "bits": 4,
1328
+ "mode": "affine"
1329
+ },
1330
+ "language_model.model.layers.37.self_attn.o_proj": {
1331
+ "group_size": 32,
1332
+ "bits": 4,
1333
+ "mode": "affine"
1334
+ },
1335
+ "language_model.model.layers.37.mlp.gate_proj": {
1336
+ "group_size": 64,
1337
+ "bits": 4,
1338
+ "mode": "affine"
1339
+ },
1340
+ "language_model.model.layers.37.mlp.down_proj": {
1341
+ "group_size": 64,
1342
+ "bits": 4,
1343
+ "mode": "affine"
1344
+ },
1345
+ "language_model.model.layers.37.mlp.up_proj": {
1346
+ "group_size": 64,
1347
+ "bits": 4,
1348
+ "mode": "affine"
1349
+ },
1350
+ "language_model.model.layers.38.self_attn.q_proj": {
1351
+ "group_size": 32,
1352
+ "bits": 4,
1353
+ "mode": "affine"
1354
+ },
1355
+ "language_model.model.layers.38.self_attn.k_proj": {
1356
+ "group_size": 32,
1357
+ "bits": 4,
1358
+ "mode": "affine"
1359
+ },
1360
+ "language_model.model.layers.38.self_attn.v_proj": {
1361
+ "group_size": 32,
1362
+ "bits": 4,
1363
+ "mode": "affine"
1364
+ },
1365
+ "language_model.model.layers.38.self_attn.o_proj": {
1366
+ "group_size": 32,
1367
+ "bits": 4,
1368
+ "mode": "affine"
1369
+ },
1370
+ "language_model.model.layers.38.mlp.gate_proj": {
1371
+ "group_size": 64,
1372
+ "bits": 4,
1373
+ "mode": "affine"
1374
+ },
1375
+ "language_model.model.layers.38.mlp.down_proj": {
1376
+ "group_size": 64,
1377
+ "bits": 4,
1378
+ "mode": "affine"
1379
+ },
1380
+ "language_model.model.layers.38.mlp.up_proj": {
1381
+ "group_size": 64,
1382
+ "bits": 4,
1383
+ "mode": "affine"
1384
+ },
1385
+ "language_model.model.layers.39.self_attn.q_proj": {
1386
+ "group_size": 32,
1387
+ "bits": 4,
1388
+ "mode": "affine"
1389
+ },
1390
+ "language_model.model.layers.39.self_attn.k_proj": {
1391
+ "group_size": 32,
1392
+ "bits": 4,
1393
+ "mode": "affine"
1394
+ },
1395
+ "language_model.model.layers.39.self_attn.v_proj": {
1396
+ "group_size": 32,
1397
+ "bits": 4,
1398
+ "mode": "affine"
1399
+ },
1400
+ "language_model.model.layers.39.self_attn.o_proj": {
1401
+ "group_size": 32,
1402
+ "bits": 4,
1403
+ "mode": "affine"
1404
+ },
1405
+ "language_model.model.layers.39.mlp.gate_proj": {
1406
+ "group_size": 64,
1407
+ "bits": 4,
1408
+ "mode": "affine"
1409
+ },
1410
+ "language_model.model.layers.39.mlp.down_proj": {
1411
+ "group_size": 64,
1412
+ "bits": 4,
1413
+ "mode": "affine"
1414
+ },
1415
+ "language_model.model.layers.39.mlp.up_proj": {
1416
+ "group_size": 64,
1417
+ "bits": 4,
1418
+ "mode": "affine"
1419
+ }
1420
+ },
1421
+ "quantization_config": {
1422
+ "group_size": 32,
1423
+ "bits": 4,
1424
+ "mode": "affine",
1425
+ "language_model.model.embed_tokens": {
1426
+ "group_size": 64,
1427
+ "bits": 8,
1428
+ "mode": "affine"
1429
+ },
1430
+ "language_model.model.layers.0.self_attn.q_proj": {
1431
+ "group_size": 64,
1432
+ "bits": 4,
1433
+ "mode": "affine"
1434
+ },
1435
+ "language_model.model.layers.0.self_attn.k_proj": {
1436
+ "group_size": 64,
1437
+ "bits": 4,
1438
+ "mode": "affine"
1439
+ },
1440
+ "language_model.model.layers.0.self_attn.v_proj": {
1441
+ "group_size": 64,
1442
+ "bits": 4,
1443
+ "mode": "affine"
1444
+ },
1445
+ "language_model.model.layers.0.self_attn.o_proj": {
1446
+ "group_size": 64,
1447
+ "bits": 4,
1448
+ "mode": "affine"
1449
+ },
1450
+ "language_model.model.layers.0.mlp.gate_proj": {
1451
+ "group_size": 64,
1452
+ "bits": 4,
1453
+ "mode": "affine"
1454
+ },
1455
+ "language_model.model.layers.0.mlp.down_proj": {
1456
+ "group_size": 64,
1457
+ "bits": 4,
1458
+ "mode": "affine"
1459
+ },
1460
+ "language_model.model.layers.0.mlp.up_proj": {
1461
+ "group_size": 64,
1462
+ "bits": 4,
1463
+ "mode": "affine"
1464
+ },
1465
+ "language_model.model.layers.1.self_attn.q_proj": {
1466
+ "group_size": 64,
1467
+ "bits": 4,
1468
+ "mode": "affine"
1469
+ },
1470
+ "language_model.model.layers.1.self_attn.k_proj": {
1471
+ "group_size": 64,
1472
+ "bits": 4,
1473
+ "mode": "affine"
1474
+ },
1475
+ "language_model.model.layers.1.self_attn.v_proj": {
1476
+ "group_size": 64,
1477
+ "bits": 4,
1478
+ "mode": "affine"
1479
+ },
1480
+ "language_model.model.layers.1.self_attn.o_proj": {
1481
+ "group_size": 64,
1482
+ "bits": 4,
1483
+ "mode": "affine"
1484
+ },
1485
+ "language_model.model.layers.1.mlp.gate_proj": {
1486
+ "group_size": 64,
1487
+ "bits": 4,
1488
+ "mode": "affine"
1489
+ },
1490
+ "language_model.model.layers.1.mlp.down_proj": {
1491
+ "group_size": 64,
1492
+ "bits": 4,
1493
+ "mode": "affine"
1494
+ },
1495
+ "language_model.model.layers.1.mlp.up_proj": {
1496
+ "group_size": 64,
1497
+ "bits": 4,
1498
+ "mode": "affine"
1499
+ },
1500
+ "language_model.model.layers.2.self_attn.q_proj": {
1501
+ "group_size": 64,
1502
+ "bits": 4,
1503
+ "mode": "affine"
1504
+ },
1505
+ "language_model.model.layers.2.self_attn.k_proj": {
1506
+ "group_size": 64,
1507
+ "bits": 4,
1508
+ "mode": "affine"
1509
+ },
1510
+ "language_model.model.layers.2.self_attn.v_proj": {
1511
+ "group_size": 64,
1512
+ "bits": 4,
1513
+ "mode": "affine"
1514
+ },
1515
+ "language_model.model.layers.2.self_attn.o_proj": {
1516
+ "group_size": 64,
1517
+ "bits": 4,
1518
+ "mode": "affine"
1519
+ },
1520
+ "language_model.model.layers.2.mlp.gate_proj": {
1521
+ "group_size": 64,
1522
+ "bits": 4,
1523
+ "mode": "affine"
1524
+ },
1525
+ "language_model.model.layers.2.mlp.down_proj": {
1526
+ "group_size": 64,
1527
+ "bits": 4,
1528
+ "mode": "affine"
1529
+ },
1530
+ "language_model.model.layers.2.mlp.up_proj": {
1531
+ "group_size": 64,
1532
+ "bits": 4,
1533
+ "mode": "affine"
1534
+ },
1535
+ "language_model.model.layers.3.self_attn.q_proj": {
1536
+ "group_size": 64,
1537
+ "bits": 4,
1538
+ "mode": "affine"
1539
+ },
1540
+ "language_model.model.layers.3.self_attn.k_proj": {
1541
+ "group_size": 64,
1542
+ "bits": 4,
1543
+ "mode": "affine"
1544
+ },
1545
+ "language_model.model.layers.3.self_attn.v_proj": {
1546
+ "group_size": 64,
1547
+ "bits": 4,
1548
+ "mode": "affine"
1549
+ },
1550
+ "language_model.model.layers.3.self_attn.o_proj": {
1551
+ "group_size": 64,
1552
+ "bits": 4,
1553
+ "mode": "affine"
1554
+ },
1555
+ "language_model.model.layers.3.mlp.gate_proj": {
1556
+ "group_size": 64,
1557
+ "bits": 4,
1558
+ "mode": "affine"
1559
+ },
1560
+ "language_model.model.layers.3.mlp.down_proj": {
1561
+ "group_size": 64,
1562
+ "bits": 4,
1563
+ "mode": "affine"
1564
+ },
1565
+ "language_model.model.layers.3.mlp.up_proj": {
1566
+ "group_size": 64,
1567
+ "bits": 4,
1568
+ "mode": "affine"
1569
+ },
1570
+ "language_model.model.layers.4.self_attn.q_proj": {
1571
+ "group_size": 64,
1572
+ "bits": 4,
1573
+ "mode": "affine"
1574
+ },
1575
+ "language_model.model.layers.4.self_attn.k_proj": {
1576
+ "group_size": 64,
1577
+ "bits": 4,
1578
+ "mode": "affine"
1579
+ },
1580
+ "language_model.model.layers.4.self_attn.v_proj": {
1581
+ "group_size": 64,
1582
+ "bits": 4,
1583
+ "mode": "affine"
1584
+ },
1585
+ "language_model.model.layers.4.self_attn.o_proj": {
1586
+ "group_size": 64,
1587
+ "bits": 4,
1588
+ "mode": "affine"
1589
+ },
1590
+ "language_model.model.layers.4.mlp.gate_proj": {
1591
+ "group_size": 64,
1592
+ "bits": 4,
1593
+ "mode": "affine"
1594
+ },
1595
+ "language_model.model.layers.4.mlp.down_proj": {
1596
+ "group_size": 64,
1597
+ "bits": 4,
1598
+ "mode": "affine"
1599
+ },
1600
+ "language_model.model.layers.4.mlp.up_proj": {
1601
+ "group_size": 64,
1602
+ "bits": 4,
1603
+ "mode": "affine"
1604
+ },
1605
+ "language_model.model.layers.5.self_attn.q_proj": {
1606
+ "group_size": 64,
1607
+ "bits": 4,
1608
+ "mode": "affine"
1609
+ },
1610
+ "language_model.model.layers.5.self_attn.k_proj": {
1611
+ "group_size": 64,
1612
+ "bits": 4,
1613
+ "mode": "affine"
1614
+ },
1615
+ "language_model.model.layers.5.self_attn.v_proj": {
1616
+ "group_size": 64,
1617
+ "bits": 4,
1618
+ "mode": "affine"
1619
+ },
1620
+ "language_model.model.layers.5.self_attn.o_proj": {
1621
+ "group_size": 64,
1622
+ "bits": 4,
1623
+ "mode": "affine"
1624
+ },
1625
+ "language_model.model.layers.5.mlp.gate_proj": {
1626
+ "group_size": 64,
1627
+ "bits": 4,
1628
+ "mode": "affine"
1629
+ },
1630
+ "language_model.model.layers.5.mlp.down_proj": {
1631
+ "group_size": 64,
1632
+ "bits": 4,
1633
+ "mode": "affine"
1634
+ },
1635
+ "language_model.model.layers.5.mlp.up_proj": {
1636
+ "group_size": 64,
1637
+ "bits": 4,
1638
+ "mode": "affine"
1639
+ },
1640
+ "language_model.model.layers.6.self_attn.q_proj": {
1641
+ "group_size": 64,
1642
+ "bits": 4,
1643
+ "mode": "affine"
1644
+ },
1645
+ "language_model.model.layers.6.self_attn.k_proj": {
1646
+ "group_size": 64,
1647
+ "bits": 4,
1648
+ "mode": "affine"
1649
+ },
1650
+ "language_model.model.layers.6.self_attn.v_proj": {
1651
+ "group_size": 64,
1652
+ "bits": 4,
1653
+ "mode": "affine"
1654
+ },
1655
+ "language_model.model.layers.6.self_attn.o_proj": {
1656
+ "group_size": 64,
1657
+ "bits": 4,
1658
+ "mode": "affine"
1659
+ },
1660
+ "language_model.model.layers.6.mlp.gate_proj": {
1661
+ "group_size": 64,
1662
+ "bits": 4,
1663
+ "mode": "affine"
1664
+ },
1665
+ "language_model.model.layers.6.mlp.down_proj": {
1666
+ "group_size": 64,
1667
+ "bits": 4,
1668
+ "mode": "affine"
1669
+ },
1670
+ "language_model.model.layers.6.mlp.up_proj": {
1671
+ "group_size": 64,
1672
+ "bits": 4,
1673
+ "mode": "affine"
1674
+ },
1675
+ "language_model.model.layers.7.self_attn.q_proj": {
1676
+ "group_size": 64,
1677
+ "bits": 4,
1678
+ "mode": "affine"
1679
+ },
1680
+ "language_model.model.layers.7.self_attn.k_proj": {
1681
+ "group_size": 64,
1682
+ "bits": 4,
1683
+ "mode": "affine"
1684
+ },
1685
+ "language_model.model.layers.7.self_attn.v_proj": {
1686
+ "group_size": 64,
1687
+ "bits": 4,
1688
+ "mode": "affine"
1689
+ },
1690
+ "language_model.model.layers.7.self_attn.o_proj": {
1691
+ "group_size": 64,
1692
+ "bits": 4,
1693
+ "mode": "affine"
1694
+ },
1695
+ "language_model.model.layers.7.mlp.gate_proj": {
1696
+ "group_size": 64,
1697
+ "bits": 4,
1698
+ "mode": "affine"
1699
+ },
1700
+ "language_model.model.layers.7.mlp.down_proj": {
1701
+ "group_size": 64,
1702
+ "bits": 4,
1703
+ "mode": "affine"
1704
+ },
1705
+ "language_model.model.layers.7.mlp.up_proj": {
1706
+ "group_size": 64,
1707
+ "bits": 4,
1708
+ "mode": "affine"
1709
+ },
1710
+ "language_model.model.layers.8.self_attn.q_proj": {
1711
+ "group_size": 64,
1712
+ "bits": 4,
1713
+ "mode": "affine"
1714
+ },
1715
+ "language_model.model.layers.8.self_attn.k_proj": {
1716
+ "group_size": 64,
1717
+ "bits": 4,
1718
+ "mode": "affine"
1719
+ },
1720
+ "language_model.model.layers.8.self_attn.v_proj": {
1721
+ "group_size": 64,
1722
+ "bits": 4,
1723
+ "mode": "affine"
1724
+ },
1725
+ "language_model.model.layers.8.self_attn.o_proj": {
1726
+ "group_size": 64,
1727
+ "bits": 4,
1728
+ "mode": "affine"
1729
+ },
1730
+ "language_model.model.layers.8.mlp.gate_proj": {
1731
+ "group_size": 64,
1732
+ "bits": 4,
1733
+ "mode": "affine"
1734
+ },
1735
+ "language_model.model.layers.8.mlp.down_proj": {
1736
+ "group_size": 64,
1737
+ "bits": 4,
1738
+ "mode": "affine"
1739
+ },
1740
+ "language_model.model.layers.8.mlp.up_proj": {
1741
+ "group_size": 64,
1742
+ "bits": 4,
1743
+ "mode": "affine"
1744
+ },
1745
+ "language_model.model.layers.9.self_attn.q_proj": {
1746
+ "group_size": 64,
1747
+ "bits": 4,
1748
+ "mode": "affine"
1749
+ },
1750
+ "language_model.model.layers.9.self_attn.k_proj": {
1751
+ "group_size": 64,
1752
+ "bits": 4,
1753
+ "mode": "affine"
1754
+ },
1755
+ "language_model.model.layers.9.self_attn.v_proj": {
1756
+ "group_size": 64,
1757
+ "bits": 4,
1758
+ "mode": "affine"
1759
+ },
1760
+ "language_model.model.layers.9.self_attn.o_proj": {
1761
+ "group_size": 64,
1762
+ "bits": 4,
1763
+ "mode": "affine"
1764
+ },
1765
+ "language_model.model.layers.9.mlp.gate_proj": {
1766
+ "group_size": 64,
1767
+ "bits": 4,
1768
+ "mode": "affine"
1769
+ },
1770
+ "language_model.model.layers.9.mlp.down_proj": {
1771
+ "group_size": 64,
1772
+ "bits": 4,
1773
+ "mode": "affine"
1774
+ },
1775
+ "language_model.model.layers.9.mlp.up_proj": {
1776
+ "group_size": 64,
1777
+ "bits": 4,
1778
+ "mode": "affine"
1779
+ },
1780
+ "language_model.model.layers.10.self_attn.q_proj": {
1781
+ "group_size": 64,
1782
+ "bits": 4,
1783
+ "mode": "affine"
1784
+ },
1785
+ "language_model.model.layers.10.self_attn.k_proj": {
1786
+ "group_size": 64,
1787
+ "bits": 4,
1788
+ "mode": "affine"
1789
+ },
1790
+ "language_model.model.layers.10.self_attn.v_proj": {
1791
+ "group_size": 64,
1792
+ "bits": 4,
1793
+ "mode": "affine"
1794
+ },
1795
+ "language_model.model.layers.10.self_attn.o_proj": {
1796
+ "group_size": 64,
1797
+ "bits": 4,
1798
+ "mode": "affine"
1799
+ },
1800
+ "language_model.model.layers.10.mlp.gate_proj": {
1801
+ "group_size": 64,
1802
+ "bits": 4,
1803
+ "mode": "affine"
1804
+ },
1805
+ "language_model.model.layers.10.mlp.down_proj": {
1806
+ "group_size": 64,
1807
+ "bits": 4,
1808
+ "mode": "affine"
1809
+ },
1810
+ "language_model.model.layers.10.mlp.up_proj": {
1811
+ "group_size": 64,
1812
+ "bits": 4,
1813
+ "mode": "affine"
1814
+ },
1815
+ "language_model.model.layers.11.self_attn.q_proj": {
1816
+ "group_size": 64,
1817
+ "bits": 4,
1818
+ "mode": "affine"
1819
+ },
1820
+ "language_model.model.layers.11.self_attn.k_proj": {
1821
+ "group_size": 64,
1822
+ "bits": 4,
1823
+ "mode": "affine"
1824
+ },
1825
+ "language_model.model.layers.11.self_attn.v_proj": {
1826
+ "group_size": 64,
1827
+ "bits": 4,
1828
+ "mode": "affine"
1829
+ },
1830
+ "language_model.model.layers.11.self_attn.o_proj": {
1831
+ "group_size": 64,
1832
+ "bits": 4,
1833
+ "mode": "affine"
1834
+ },
1835
+ "language_model.model.layers.11.mlp.gate_proj": {
1836
+ "group_size": 64,
1837
+ "bits": 4,
1838
+ "mode": "affine"
1839
+ },
1840
+ "language_model.model.layers.11.mlp.down_proj": {
1841
+ "group_size": 64,
1842
+ "bits": 4,
1843
+ "mode": "affine"
1844
+ },
1845
+ "language_model.model.layers.11.mlp.up_proj": {
1846
+ "group_size": 64,
1847
+ "bits": 4,
1848
+ "mode": "affine"
1849
+ },
1850
+ "language_model.model.layers.12.self_attn.q_proj": {
1851
+ "group_size": 64,
1852
+ "bits": 4,
1853
+ "mode": "affine"
1854
+ },
1855
+ "language_model.model.layers.12.self_attn.k_proj": {
1856
+ "group_size": 64,
1857
+ "bits": 4,
1858
+ "mode": "affine"
1859
+ },
1860
+ "language_model.model.layers.12.self_attn.v_proj": {
1861
+ "group_size": 64,
1862
+ "bits": 4,
1863
+ "mode": "affine"
1864
+ },
1865
+ "language_model.model.layers.12.self_attn.o_proj": {
1866
+ "group_size": 64,
1867
+ "bits": 4,
1868
+ "mode": "affine"
1869
+ },
1870
+ "language_model.model.layers.12.mlp.gate_proj": {
1871
+ "group_size": 64,
1872
+ "bits": 4,
1873
+ "mode": "affine"
1874
+ },
1875
+ "language_model.model.layers.12.mlp.down_proj": {
1876
+ "group_size": 64,
1877
+ "bits": 4,
1878
+ "mode": "affine"
1879
+ },
1880
+ "language_model.model.layers.12.mlp.up_proj": {
1881
+ "group_size": 64,
1882
+ "bits": 4,
1883
+ "mode": "affine"
1884
+ },
1885
+ "language_model.model.layers.13.self_attn.q_proj": {
1886
+ "group_size": 64,
1887
+ "bits": 4,
1888
+ "mode": "affine"
1889
+ },
1890
+ "language_model.model.layers.13.self_attn.k_proj": {
1891
+ "group_size": 64,
1892
+ "bits": 4,
1893
+ "mode": "affine"
1894
+ },
1895
+ "language_model.model.layers.13.self_attn.v_proj": {
1896
+ "group_size": 64,
1897
+ "bits": 4,
1898
+ "mode": "affine"
1899
+ },
1900
+ "language_model.model.layers.13.self_attn.o_proj": {
1901
+ "group_size": 64,
1902
+ "bits": 4,
1903
+ "mode": "affine"
1904
+ },
1905
+ "language_model.model.layers.13.mlp.gate_proj": {
1906
+ "group_size": 64,
1907
+ "bits": 4,
1908
+ "mode": "affine"
1909
+ },
1910
+ "language_model.model.layers.13.mlp.down_proj": {
1911
+ "group_size": 64,
1912
+ "bits": 4,
1913
+ "mode": "affine"
1914
+ },
1915
+ "language_model.model.layers.13.mlp.up_proj": {
1916
+ "group_size": 64,
1917
+ "bits": 4,
1918
+ "mode": "affine"
1919
+ },
1920
+ "language_model.model.layers.14.self_attn.q_proj": {
1921
+ "group_size": 64,
1922
+ "bits": 4,
1923
+ "mode": "affine"
1924
+ },
1925
+ "language_model.model.layers.14.self_attn.k_proj": {
1926
+ "group_size": 64,
1927
+ "bits": 4,
1928
+ "mode": "affine"
1929
+ },
1930
+ "language_model.model.layers.14.self_attn.v_proj": {
1931
+ "group_size": 64,
1932
+ "bits": 4,
1933
+ "mode": "affine"
1934
+ },
1935
+ "language_model.model.layers.14.self_attn.o_proj": {
1936
+ "group_size": 64,
1937
+ "bits": 4,
1938
+ "mode": "affine"
1939
+ },
1940
+ "language_model.model.layers.14.mlp.gate_proj": {
1941
+ "group_size": 64,
1942
+ "bits": 4,
1943
+ "mode": "affine"
1944
+ },
1945
+ "language_model.model.layers.14.mlp.down_proj": {
1946
+ "group_size": 64,
1947
+ "bits": 4,
1948
+ "mode": "affine"
1949
+ },
1950
+ "language_model.model.layers.14.mlp.up_proj": {
1951
+ "group_size": 64,
1952
+ "bits": 4,
1953
+ "mode": "affine"
1954
+ },
1955
+ "language_model.model.layers.15.self_attn.q_proj": {
1956
+ "group_size": 64,
1957
+ "bits": 4,
1958
+ "mode": "affine"
1959
+ },
1960
+ "language_model.model.layers.15.self_attn.k_proj": {
1961
+ "group_size": 64,
1962
+ "bits": 4,
1963
+ "mode": "affine"
1964
+ },
1965
+ "language_model.model.layers.15.self_attn.v_proj": {
1966
+ "group_size": 64,
1967
+ "bits": 4,
1968
+ "mode": "affine"
1969
+ },
1970
+ "language_model.model.layers.15.self_attn.o_proj": {
1971
+ "group_size": 64,
1972
+ "bits": 4,
1973
+ "mode": "affine"
1974
+ },
1975
+ "language_model.model.layers.15.mlp.gate_proj": {
1976
+ "group_size": 64,
1977
+ "bits": 4,
1978
+ "mode": "affine"
1979
+ },
1980
+ "language_model.model.layers.15.mlp.down_proj": {
1981
+ "group_size": 64,
1982
+ "bits": 4,
1983
+ "mode": "affine"
1984
+ },
1985
+ "language_model.model.layers.15.mlp.up_proj": {
1986
+ "group_size": 64,
1987
+ "bits": 4,
1988
+ "mode": "affine"
1989
+ },
1990
+ "language_model.model.layers.16.self_attn.q_proj": {
1991
+ "group_size": 64,
1992
+ "bits": 4,
1993
+ "mode": "affine"
1994
+ },
1995
+ "language_model.model.layers.16.self_attn.k_proj": {
1996
+ "group_size": 64,
1997
+ "bits": 4,
1998
+ "mode": "affine"
1999
+ },
2000
+ "language_model.model.layers.16.self_attn.v_proj": {
2001
+ "group_size": 64,
2002
+ "bits": 4,
2003
+ "mode": "affine"
2004
+ },
2005
+ "language_model.model.layers.16.self_attn.o_proj": {
2006
+ "group_size": 64,
2007
+ "bits": 4,
2008
+ "mode": "affine"
2009
+ },
2010
+ "language_model.model.layers.16.mlp.gate_proj": {
2011
+ "group_size": 64,
2012
+ "bits": 4,
2013
+ "mode": "affine"
2014
+ },
2015
+ "language_model.model.layers.16.mlp.down_proj": {
2016
+ "group_size": 64,
2017
+ "bits": 4,
2018
+ "mode": "affine"
2019
+ },
2020
+ "language_model.model.layers.16.mlp.up_proj": {
2021
+ "group_size": 64,
2022
+ "bits": 4,
2023
+ "mode": "affine"
2024
+ },
2025
+ "language_model.model.layers.17.self_attn.q_proj": {
2026
+ "group_size": 64,
2027
+ "bits": 4,
2028
+ "mode": "affine"
2029
+ },
2030
+ "language_model.model.layers.17.self_attn.k_proj": {
2031
+ "group_size": 64,
2032
+ "bits": 4,
2033
+ "mode": "affine"
2034
+ },
2035
+ "language_model.model.layers.17.self_attn.v_proj": {
2036
+ "group_size": 64,
2037
+ "bits": 4,
2038
+ "mode": "affine"
2039
+ },
2040
+ "language_model.model.layers.17.self_attn.o_proj": {
2041
+ "group_size": 64,
2042
+ "bits": 4,
2043
+ "mode": "affine"
2044
+ },
2045
+ "language_model.model.layers.17.mlp.gate_proj": {
2046
+ "group_size": 64,
2047
+ "bits": 4,
2048
+ "mode": "affine"
2049
+ },
2050
+ "language_model.model.layers.17.mlp.down_proj": {
2051
+ "group_size": 64,
2052
+ "bits": 4,
2053
+ "mode": "affine"
2054
+ },
2055
+ "language_model.model.layers.17.mlp.up_proj": {
2056
+ "group_size": 64,
2057
+ "bits": 4,
2058
+ "mode": "affine"
2059
+ },
2060
+ "language_model.model.layers.18.self_attn.q_proj": {
2061
+ "group_size": 64,
2062
+ "bits": 4,
2063
+ "mode": "affine"
2064
+ },
2065
+ "language_model.model.layers.18.self_attn.k_proj": {
2066
+ "group_size": 64,
2067
+ "bits": 4,
2068
+ "mode": "affine"
2069
+ },
2070
+ "language_model.model.layers.18.self_attn.v_proj": {
2071
+ "group_size": 64,
2072
+ "bits": 4,
2073
+ "mode": "affine"
2074
+ },
2075
+ "language_model.model.layers.18.self_attn.o_proj": {
2076
+ "group_size": 64,
2077
+ "bits": 4,
2078
+ "mode": "affine"
2079
+ },
2080
+ "language_model.model.layers.18.mlp.gate_proj": {
2081
+ "group_size": 64,
2082
+ "bits": 4,
2083
+ "mode": "affine"
2084
+ },
2085
+ "language_model.model.layers.18.mlp.down_proj": {
2086
+ "group_size": 64,
2087
+ "bits": 4,
2088
+ "mode": "affine"
2089
+ },
2090
+ "language_model.model.layers.18.mlp.up_proj": {
2091
+ "group_size": 64,
2092
+ "bits": 4,
2093
+ "mode": "affine"
2094
+ },
2095
+ "language_model.model.layers.19.self_attn.q_proj": {
2096
+ "group_size": 64,
2097
+ "bits": 4,
2098
+ "mode": "affine"
2099
+ },
2100
+ "language_model.model.layers.19.self_attn.k_proj": {
2101
+ "group_size": 64,
2102
+ "bits": 4,
2103
+ "mode": "affine"
2104
+ },
2105
+ "language_model.model.layers.19.self_attn.v_proj": {
2106
+ "group_size": 64,
2107
+ "bits": 4,
2108
+ "mode": "affine"
2109
+ },
2110
+ "language_model.model.layers.19.self_attn.o_proj": {
2111
+ "group_size": 64,
2112
+ "bits": 4,
2113
+ "mode": "affine"
2114
+ },
2115
+ "language_model.model.layers.19.mlp.gate_proj": {
2116
+ "group_size": 64,
2117
+ "bits": 4,
2118
+ "mode": "affine"
2119
+ },
2120
+ "language_model.model.layers.19.mlp.down_proj": {
2121
+ "group_size": 64,
2122
+ "bits": 4,
2123
+ "mode": "affine"
2124
+ },
2125
+ "language_model.model.layers.19.mlp.up_proj": {
2126
+ "group_size": 64,
2127
+ "bits": 4,
2128
+ "mode": "affine"
2129
+ },
2130
+ "language_model.model.layers.20.self_attn.q_proj": {
2131
+ "group_size": 64,
2132
+ "bits": 4,
2133
+ "mode": "affine"
2134
+ },
2135
+ "language_model.model.layers.20.self_attn.k_proj": {
2136
+ "group_size": 64,
2137
+ "bits": 4,
2138
+ "mode": "affine"
2139
+ },
2140
+ "language_model.model.layers.20.self_attn.v_proj": {
2141
+ "group_size": 64,
2142
+ "bits": 4,
2143
+ "mode": "affine"
2144
+ },
2145
+ "language_model.model.layers.20.self_attn.o_proj": {
2146
+ "group_size": 64,
2147
+ "bits": 4,
2148
+ "mode": "affine"
2149
+ },
2150
+ "language_model.model.layers.20.mlp.gate_proj": {
2151
+ "group_size": 64,
2152
+ "bits": 4,
2153
+ "mode": "affine"
2154
+ },
2155
+ "language_model.model.layers.20.mlp.down_proj": {
2156
+ "group_size": 64,
2157
+ "bits": 4,
2158
+ "mode": "affine"
2159
+ },
2160
+ "language_model.model.layers.20.mlp.up_proj": {
2161
+ "group_size": 64,
2162
+ "bits": 4,
2163
+ "mode": "affine"
2164
+ },
2165
+ "language_model.model.layers.21.self_attn.q_proj": {
2166
+ "group_size": 64,
2167
+ "bits": 4,
2168
+ "mode": "affine"
2169
+ },
2170
+ "language_model.model.layers.21.self_attn.k_proj": {
2171
+ "group_size": 64,
2172
+ "bits": 4,
2173
+ "mode": "affine"
2174
+ },
2175
+ "language_model.model.layers.21.self_attn.v_proj": {
2176
+ "group_size": 64,
2177
+ "bits": 4,
2178
+ "mode": "affine"
2179
+ },
2180
+ "language_model.model.layers.21.self_attn.o_proj": {
2181
+ "group_size": 64,
2182
+ "bits": 4,
2183
+ "mode": "affine"
2184
+ },
2185
+ "language_model.model.layers.21.mlp.gate_proj": {
2186
+ "group_size": 64,
2187
+ "bits": 4,
2188
+ "mode": "affine"
2189
+ },
2190
+ "language_model.model.layers.21.mlp.down_proj": {
2191
+ "group_size": 64,
2192
+ "bits": 4,
2193
+ "mode": "affine"
2194
+ },
2195
+ "language_model.model.layers.21.mlp.up_proj": {
2196
+ "group_size": 64,
2197
+ "bits": 4,
2198
+ "mode": "affine"
2199
+ },
2200
+ "language_model.model.layers.22.self_attn.q_proj": {
2201
+ "group_size": 64,
2202
+ "bits": 4,
2203
+ "mode": "affine"
2204
+ },
2205
+ "language_model.model.layers.22.self_attn.k_proj": {
2206
+ "group_size": 64,
2207
+ "bits": 4,
2208
+ "mode": "affine"
2209
+ },
2210
+ "language_model.model.layers.22.self_attn.v_proj": {
2211
+ "group_size": 64,
2212
+ "bits": 4,
2213
+ "mode": "affine"
2214
+ },
2215
+ "language_model.model.layers.22.self_attn.o_proj": {
2216
+ "group_size": 64,
2217
+ "bits": 4,
2218
+ "mode": "affine"
2219
+ },
2220
+ "language_model.model.layers.22.mlp.gate_proj": {
2221
+ "group_size": 64,
2222
+ "bits": 4,
2223
+ "mode": "affine"
2224
+ },
2225
+ "language_model.model.layers.22.mlp.down_proj": {
2226
+ "group_size": 64,
2227
+ "bits": 4,
2228
+ "mode": "affine"
2229
+ },
2230
+ "language_model.model.layers.22.mlp.up_proj": {
2231
+ "group_size": 64,
2232
+ "bits": 4,
2233
+ "mode": "affine"
2234
+ },
2235
+ "language_model.model.layers.23.self_attn.q_proj": {
2236
+ "group_size": 64,
2237
+ "bits": 4,
2238
+ "mode": "affine"
2239
+ },
2240
+ "language_model.model.layers.23.self_attn.k_proj": {
2241
+ "group_size": 64,
2242
+ "bits": 4,
2243
+ "mode": "affine"
2244
+ },
2245
+ "language_model.model.layers.23.self_attn.v_proj": {
2246
+ "group_size": 64,
2247
+ "bits": 4,
2248
+ "mode": "affine"
2249
+ },
2250
+ "language_model.model.layers.23.self_attn.o_proj": {
2251
+ "group_size": 64,
2252
+ "bits": 4,
2253
+ "mode": "affine"
2254
+ },
2255
+ "language_model.model.layers.23.mlp.gate_proj": {
2256
+ "group_size": 64,
2257
+ "bits": 4,
2258
+ "mode": "affine"
2259
+ },
2260
+ "language_model.model.layers.23.mlp.down_proj": {
2261
+ "group_size": 64,
2262
+ "bits": 4,
2263
+ "mode": "affine"
2264
+ },
2265
+ "language_model.model.layers.23.mlp.up_proj": {
2266
+ "group_size": 64,
2267
+ "bits": 4,
2268
+ "mode": "affine"
2269
+ },
2270
+ "language_model.model.layers.24.self_attn.q_proj": {
2271
+ "group_size": 64,
2272
+ "bits": 4,
2273
+ "mode": "affine"
2274
+ },
2275
+ "language_model.model.layers.24.self_attn.k_proj": {
2276
+ "group_size": 64,
2277
+ "bits": 4,
2278
+ "mode": "affine"
2279
+ },
2280
+ "language_model.model.layers.24.self_attn.v_proj": {
2281
+ "group_size": 64,
2282
+ "bits": 4,
2283
+ "mode": "affine"
2284
+ },
2285
+ "language_model.model.layers.24.self_attn.o_proj": {
2286
+ "group_size": 64,
2287
+ "bits": 4,
2288
+ "mode": "affine"
2289
+ },
2290
+ "language_model.model.layers.24.mlp.gate_proj": {
2291
+ "group_size": 64,
2292
+ "bits": 4,
2293
+ "mode": "affine"
2294
+ },
2295
+ "language_model.model.layers.24.mlp.down_proj": {
2296
+ "group_size": 64,
2297
+ "bits": 4,
2298
+ "mode": "affine"
2299
+ },
2300
+ "language_model.model.layers.24.mlp.up_proj": {
2301
+ "group_size": 64,
2302
+ "bits": 4,
2303
+ "mode": "affine"
2304
+ },
2305
+ "language_model.model.layers.25.self_attn.q_proj": {
2306
+ "group_size": 64,
2307
+ "bits": 4,
2308
+ "mode": "affine"
2309
+ },
2310
+ "language_model.model.layers.25.self_attn.k_proj": {
2311
+ "group_size": 64,
2312
+ "bits": 4,
2313
+ "mode": "affine"
2314
+ },
2315
+ "language_model.model.layers.25.self_attn.v_proj": {
2316
+ "group_size": 64,
2317
+ "bits": 4,
2318
+ "mode": "affine"
2319
+ },
2320
+ "language_model.model.layers.25.self_attn.o_proj": {
2321
+ "group_size": 64,
2322
+ "bits": 4,
2323
+ "mode": "affine"
2324
+ },
2325
+ "language_model.model.layers.25.mlp.gate_proj": {
2326
+ "group_size": 64,
2327
+ "bits": 4,
2328
+ "mode": "affine"
2329
+ },
2330
+ "language_model.model.layers.25.mlp.down_proj": {
2331
+ "group_size": 64,
2332
+ "bits": 4,
2333
+ "mode": "affine"
2334
+ },
2335
+ "language_model.model.layers.25.mlp.up_proj": {
2336
+ "group_size": 64,
2337
+ "bits": 4,
2338
+ "mode": "affine"
2339
+ },
2340
+ "language_model.model.layers.26.self_attn.q_proj": {
2341
+ "group_size": 64,
2342
+ "bits": 4,
2343
+ "mode": "affine"
2344
+ },
2345
+ "language_model.model.layers.26.self_attn.k_proj": {
2346
+ "group_size": 64,
2347
+ "bits": 4,
2348
+ "mode": "affine"
2349
+ },
2350
+ "language_model.model.layers.26.self_attn.v_proj": {
2351
+ "group_size": 64,
2352
+ "bits": 4,
2353
+ "mode": "affine"
2354
+ },
2355
+ "language_model.model.layers.26.self_attn.o_proj": {
2356
+ "group_size": 64,
2357
+ "bits": 4,
2358
+ "mode": "affine"
2359
+ },
2360
+ "language_model.model.layers.26.mlp.gate_proj": {
2361
+ "group_size": 64,
2362
+ "bits": 4,
2363
+ "mode": "affine"
2364
+ },
2365
+ "language_model.model.layers.26.mlp.down_proj": {
2366
+ "group_size": 64,
2367
+ "bits": 4,
2368
+ "mode": "affine"
2369
+ },
2370
+ "language_model.model.layers.26.mlp.up_proj": {
2371
+ "group_size": 64,
2372
+ "bits": 4,
2373
+ "mode": "affine"
2374
+ },
2375
+ "language_model.model.layers.27.self_attn.q_proj": {
2376
+ "group_size": 64,
2377
+ "bits": 4,
2378
+ "mode": "affine"
2379
+ },
2380
+ "language_model.model.layers.27.self_attn.k_proj": {
2381
+ "group_size": 64,
2382
+ "bits": 4,
2383
+ "mode": "affine"
2384
+ },
2385
+ "language_model.model.layers.27.self_attn.v_proj": {
2386
+ "group_size": 64,
2387
+ "bits": 4,
2388
+ "mode": "affine"
2389
+ },
2390
+ "language_model.model.layers.27.self_attn.o_proj": {
2391
+ "group_size": 64,
2392
+ "bits": 4,
2393
+ "mode": "affine"
2394
+ },
2395
+ "language_model.model.layers.27.mlp.gate_proj": {
2396
+ "group_size": 64,
2397
+ "bits": 4,
2398
+ "mode": "affine"
2399
+ },
2400
+ "language_model.model.layers.27.mlp.down_proj": {
2401
+ "group_size": 64,
2402
+ "bits": 4,
2403
+ "mode": "affine"
2404
+ },
2405
+ "language_model.model.layers.27.mlp.up_proj": {
2406
+ "group_size": 64,
2407
+ "bits": 4,
2408
+ "mode": "affine"
2409
+ },
2410
+ "language_model.model.layers.28.self_attn.q_proj": {
2411
+ "group_size": 64,
2412
+ "bits": 4,
2413
+ "mode": "affine"
2414
+ },
2415
+ "language_model.model.layers.28.self_attn.k_proj": {
2416
+ "group_size": 64,
2417
+ "bits": 4,
2418
+ "mode": "affine"
2419
+ },
2420
+ "language_model.model.layers.28.self_attn.v_proj": {
2421
+ "group_size": 64,
2422
+ "bits": 4,
2423
+ "mode": "affine"
2424
+ },
2425
+ "language_model.model.layers.28.self_attn.o_proj": {
2426
+ "group_size": 64,
2427
+ "bits": 4,
2428
+ "mode": "affine"
2429
+ },
2430
+ "language_model.model.layers.28.mlp.gate_proj": {
2431
+ "group_size": 64,
2432
+ "bits": 4,
2433
+ "mode": "affine"
2434
+ },
2435
+ "language_model.model.layers.28.mlp.down_proj": {
2436
+ "group_size": 64,
2437
+ "bits": 4,
2438
+ "mode": "affine"
2439
+ },
2440
+ "language_model.model.layers.28.mlp.up_proj": {
2441
+ "group_size": 64,
2442
+ "bits": 4,
2443
+ "mode": "affine"
2444
+ },
2445
+ "language_model.model.layers.29.self_attn.q_proj": {
2446
+ "group_size": 64,
2447
+ "bits": 4,
2448
+ "mode": "affine"
2449
+ },
2450
+ "language_model.model.layers.29.self_attn.k_proj": {
2451
+ "group_size": 64,
2452
+ "bits": 4,
2453
+ "mode": "affine"
2454
+ },
2455
+ "language_model.model.layers.29.self_attn.v_proj": {
2456
+ "group_size": 64,
2457
+ "bits": 4,
2458
+ "mode": "affine"
2459
+ },
2460
+ "language_model.model.layers.29.self_attn.o_proj": {
2461
+ "group_size": 64,
2462
+ "bits": 4,
2463
+ "mode": "affine"
2464
+ },
2465
+ "language_model.model.layers.29.mlp.gate_proj": {
2466
+ "group_size": 64,
2467
+ "bits": 4,
2468
+ "mode": "affine"
2469
+ },
2470
+ "language_model.model.layers.29.mlp.down_proj": {
2471
+ "group_size": 64,
2472
+ "bits": 4,
2473
+ "mode": "affine"
2474
+ },
2475
+ "language_model.model.layers.29.mlp.up_proj": {
2476
+ "group_size": 64,
2477
+ "bits": 4,
2478
+ "mode": "affine"
2479
+ },
2480
+ "language_model.model.layers.30.self_attn.q_proj": {
2481
+ "group_size": 64,
2482
+ "bits": 4,
2483
+ "mode": "affine"
2484
+ },
2485
+ "language_model.model.layers.30.self_attn.k_proj": {
2486
+ "group_size": 64,
2487
+ "bits": 4,
2488
+ "mode": "affine"
2489
+ },
2490
+ "language_model.model.layers.30.self_attn.v_proj": {
2491
+ "group_size": 64,
2492
+ "bits": 4,
2493
+ "mode": "affine"
2494
+ },
2495
+ "language_model.model.layers.30.self_attn.o_proj": {
2496
+ "group_size": 64,
2497
+ "bits": 4,
2498
+ "mode": "affine"
2499
+ },
2500
+ "language_model.model.layers.30.mlp.gate_proj": {
2501
+ "group_size": 64,
2502
+ "bits": 4,
2503
+ "mode": "affine"
2504
+ },
2505
+ "language_model.model.layers.30.mlp.down_proj": {
2506
+ "group_size": 64,
2507
+ "bits": 4,
2508
+ "mode": "affine"
2509
+ },
2510
+ "language_model.model.layers.30.mlp.up_proj": {
2511
+ "group_size": 64,
2512
+ "bits": 4,
2513
+ "mode": "affine"
2514
+ },
2515
+ "language_model.model.layers.31.self_attn.q_proj": {
2516
+ "group_size": 64,
2517
+ "bits": 4,
2518
+ "mode": "affine"
2519
+ },
2520
+ "language_model.model.layers.31.self_attn.k_proj": {
2521
+ "group_size": 64,
2522
+ "bits": 4,
2523
+ "mode": "affine"
2524
+ },
2525
+ "language_model.model.layers.31.self_attn.v_proj": {
2526
+ "group_size": 64,
2527
+ "bits": 4,
2528
+ "mode": "affine"
2529
+ },
2530
+ "language_model.model.layers.31.self_attn.o_proj": {
2531
+ "group_size": 64,
2532
+ "bits": 4,
2533
+ "mode": "affine"
2534
+ },
2535
+ "language_model.model.layers.31.mlp.gate_proj": {
2536
+ "group_size": 64,
2537
+ "bits": 4,
2538
+ "mode": "affine"
2539
+ },
2540
+ "language_model.model.layers.31.mlp.down_proj": {
2541
+ "group_size": 64,
2542
+ "bits": 4,
2543
+ "mode": "affine"
2544
+ },
2545
+ "language_model.model.layers.31.mlp.up_proj": {
2546
+ "group_size": 64,
2547
+ "bits": 4,
2548
+ "mode": "affine"
2549
+ },
2550
+ "language_model.model.layers.32.self_attn.q_proj": {
2551
+ "group_size": 64,
2552
+ "bits": 4,
2553
+ "mode": "affine"
2554
+ },
2555
+ "language_model.model.layers.32.self_attn.k_proj": {
2556
+ "group_size": 64,
2557
+ "bits": 4,
2558
+ "mode": "affine"
2559
+ },
2560
+ "language_model.model.layers.32.self_attn.v_proj": {
2561
+ "group_size": 64,
2562
+ "bits": 4,
2563
+ "mode": "affine"
2564
+ },
2565
+ "language_model.model.layers.32.self_attn.o_proj": {
2566
+ "group_size": 64,
2567
+ "bits": 4,
2568
+ "mode": "affine"
2569
+ },
2570
+ "language_model.model.layers.32.mlp.gate_proj": {
2571
+ "group_size": 64,
2572
+ "bits": 4,
2573
+ "mode": "affine"
2574
+ },
2575
+ "language_model.model.layers.32.mlp.down_proj": {
2576
+ "group_size": 64,
2577
+ "bits": 4,
2578
+ "mode": "affine"
2579
+ },
2580
+ "language_model.model.layers.32.mlp.up_proj": {
2581
+ "group_size": 64,
2582
+ "bits": 4,
2583
+ "mode": "affine"
2584
+ },
2585
+ "language_model.model.layers.33.self_attn.q_proj": {
2586
+ "group_size": 64,
2587
+ "bits": 4,
2588
+ "mode": "affine"
2589
+ },
2590
+ "language_model.model.layers.33.self_attn.k_proj": {
2591
+ "group_size": 64,
2592
+ "bits": 4,
2593
+ "mode": "affine"
2594
+ },
2595
+ "language_model.model.layers.33.self_attn.v_proj": {
2596
+ "group_size": 64,
2597
+ "bits": 4,
2598
+ "mode": "affine"
2599
+ },
2600
+ "language_model.model.layers.33.self_attn.o_proj": {
2601
+ "group_size": 64,
2602
+ "bits": 4,
2603
+ "mode": "affine"
2604
+ },
2605
+ "language_model.model.layers.33.mlp.gate_proj": {
2606
+ "group_size": 64,
2607
+ "bits": 4,
2608
+ "mode": "affine"
2609
+ },
2610
+ "language_model.model.layers.33.mlp.down_proj": {
2611
+ "group_size": 64,
2612
+ "bits": 4,
2613
+ "mode": "affine"
2614
+ },
2615
+ "language_model.model.layers.33.mlp.up_proj": {
2616
+ "group_size": 64,
2617
+ "bits": 4,
2618
+ "mode": "affine"
2619
+ },
2620
+ "language_model.model.layers.34.self_attn.q_proj": {
2621
+ "group_size": 64,
2622
+ "bits": 4,
2623
+ "mode": "affine"
2624
+ },
2625
+ "language_model.model.layers.34.self_attn.k_proj": {
2626
+ "group_size": 32,
2627
+ "bits": 4,
2628
+ "mode": "affine"
2629
+ },
2630
+ "language_model.model.layers.34.self_attn.v_proj": {
2631
+ "group_size": 32,
2632
+ "bits": 4,
2633
+ "mode": "affine"
2634
+ },
2635
+ "language_model.model.layers.34.self_attn.o_proj": {
2636
+ "group_size": 64,
2637
+ "bits": 4,
2638
+ "mode": "affine"
2639
+ },
2640
+ "language_model.model.layers.34.mlp.gate_proj": {
2641
+ "group_size": 64,
2642
+ "bits": 4,
2643
+ "mode": "affine"
2644
+ },
2645
+ "language_model.model.layers.34.mlp.down_proj": {
2646
+ "group_size": 64,
2647
+ "bits": 4,
2648
+ "mode": "affine"
2649
+ },
2650
+ "language_model.model.layers.34.mlp.up_proj": {
2651
+ "group_size": 64,
2652
+ "bits": 4,
2653
+ "mode": "affine"
2654
+ },
2655
+ "language_model.model.layers.35.self_attn.q_proj": {
2656
+ "group_size": 32,
2657
+ "bits": 4,
2658
+ "mode": "affine"
2659
+ },
2660
+ "language_model.model.layers.35.self_attn.k_proj": {
2661
+ "group_size": 32,
2662
+ "bits": 4,
2663
+ "mode": "affine"
2664
+ },
2665
+ "language_model.model.layers.35.self_attn.v_proj": {
2666
+ "group_size": 32,
2667
+ "bits": 4,
2668
+ "mode": "affine"
2669
+ },
2670
+ "language_model.model.layers.35.self_attn.o_proj": {
2671
+ "group_size": 64,
2672
+ "bits": 4,
2673
+ "mode": "affine"
2674
+ },
2675
+ "language_model.model.layers.35.mlp.gate_proj": {
2676
+ "group_size": 64,
2677
+ "bits": 4,
2678
+ "mode": "affine"
2679
+ },
2680
+ "language_model.model.layers.35.mlp.down_proj": {
2681
+ "group_size": 64,
2682
+ "bits": 4,
2683
+ "mode": "affine"
2684
+ },
2685
+ "language_model.model.layers.35.mlp.up_proj": {
2686
+ "group_size": 64,
2687
+ "bits": 4,
2688
+ "mode": "affine"
2689
+ },
2690
+ "language_model.model.layers.36.self_attn.q_proj": {
2691
+ "group_size": 32,
2692
+ "bits": 4,
2693
+ "mode": "affine"
2694
+ },
2695
+ "language_model.model.layers.36.self_attn.k_proj": {
2696
+ "group_size": 32,
2697
+ "bits": 4,
2698
+ "mode": "affine"
2699
+ },
2700
+ "language_model.model.layers.36.self_attn.v_proj": {
2701
+ "group_size": 32,
2702
+ "bits": 4,
2703
+ "mode": "affine"
2704
+ },
2705
+ "language_model.model.layers.36.self_attn.o_proj": {
2706
+ "group_size": 32,
2707
+ "bits": 4,
2708
+ "mode": "affine"
2709
+ },
2710
+ "language_model.model.layers.36.mlp.gate_proj": {
2711
+ "group_size": 64,
2712
+ "bits": 4,
2713
+ "mode": "affine"
2714
+ },
2715
+ "language_model.model.layers.36.mlp.down_proj": {
2716
+ "group_size": 64,
2717
+ "bits": 4,
2718
+ "mode": "affine"
2719
+ },
2720
+ "language_model.model.layers.36.mlp.up_proj": {
2721
+ "group_size": 64,
2722
+ "bits": 4,
2723
+ "mode": "affine"
2724
+ },
2725
+ "language_model.model.layers.37.self_attn.q_proj": {
2726
+ "group_size": 32,
2727
+ "bits": 4,
2728
+ "mode": "affine"
2729
+ },
2730
+ "language_model.model.layers.37.self_attn.k_proj": {
2731
+ "group_size": 32,
2732
+ "bits": 4,
2733
+ "mode": "affine"
2734
+ },
2735
+ "language_model.model.layers.37.self_attn.v_proj": {
2736
+ "group_size": 32,
2737
+ "bits": 4,
2738
+ "mode": "affine"
2739
+ },
2740
+ "language_model.model.layers.37.self_attn.o_proj": {
2741
+ "group_size": 32,
2742
+ "bits": 4,
2743
+ "mode": "affine"
2744
+ },
2745
+ "language_model.model.layers.37.mlp.gate_proj": {
2746
+ "group_size": 64,
2747
+ "bits": 4,
2748
+ "mode": "affine"
2749
+ },
2750
+ "language_model.model.layers.37.mlp.down_proj": {
2751
+ "group_size": 64,
2752
+ "bits": 4,
2753
+ "mode": "affine"
2754
+ },
2755
+ "language_model.model.layers.37.mlp.up_proj": {
2756
+ "group_size": 64,
2757
+ "bits": 4,
2758
+ "mode": "affine"
2759
+ },
2760
+ "language_model.model.layers.38.self_attn.q_proj": {
2761
+ "group_size": 32,
2762
+ "bits": 4,
2763
+ "mode": "affine"
2764
+ },
2765
+ "language_model.model.layers.38.self_attn.k_proj": {
2766
+ "group_size": 32,
2767
+ "bits": 4,
2768
+ "mode": "affine"
2769
+ },
2770
+ "language_model.model.layers.38.self_attn.v_proj": {
2771
+ "group_size": 32,
2772
+ "bits": 4,
2773
+ "mode": "affine"
2774
+ },
2775
+ "language_model.model.layers.38.self_attn.o_proj": {
2776
+ "group_size": 32,
2777
+ "bits": 4,
2778
+ "mode": "affine"
2779
+ },
2780
+ "language_model.model.layers.38.mlp.gate_proj": {
2781
+ "group_size": 64,
2782
+ "bits": 4,
2783
+ "mode": "affine"
2784
+ },
2785
+ "language_model.model.layers.38.mlp.down_proj": {
2786
+ "group_size": 64,
2787
+ "bits": 4,
2788
+ "mode": "affine"
2789
+ },
2790
+ "language_model.model.layers.38.mlp.up_proj": {
2791
+ "group_size": 64,
2792
+ "bits": 4,
2793
+ "mode": "affine"
2794
+ },
2795
+ "language_model.model.layers.39.self_attn.q_proj": {
2796
+ "group_size": 32,
2797
+ "bits": 4,
2798
+ "mode": "affine"
2799
+ },
2800
+ "language_model.model.layers.39.self_attn.k_proj": {
2801
+ "group_size": 32,
2802
+ "bits": 4,
2803
+ "mode": "affine"
2804
+ },
2805
+ "language_model.model.layers.39.self_attn.v_proj": {
2806
+ "group_size": 32,
2807
+ "bits": 4,
2808
+ "mode": "affine"
2809
+ },
2810
+ "language_model.model.layers.39.self_attn.o_proj": {
2811
+ "group_size": 32,
2812
+ "bits": 4,
2813
+ "mode": "affine"
2814
+ },
2815
+ "language_model.model.layers.39.mlp.gate_proj": {
2816
+ "group_size": 64,
2817
+ "bits": 4,
2818
+ "mode": "affine"
2819
+ },
2820
+ "language_model.model.layers.39.mlp.down_proj": {
2821
+ "group_size": 64,
2822
+ "bits": 4,
2823
+ "mode": "affine"
2824
+ },
2825
+ "language_model.model.layers.39.mlp.up_proj": {
2826
+ "group_size": 64,
2827
+ "bits": 4,
2828
+ "mode": "affine"
2829
+ }
2830
+ },
2831
+ "spatial_merge_size": 2,
2832
+ "text_config": {
2833
+ "attention_dropout": 0.0,
2834
+ "head_dim": 128,
2835
+ "hidden_act": "silu",
2836
+ "hidden_size": 5120,
2837
+ "initializer_range": 0.02,
2838
+ "intermediate_size": 16384,
2839
+ "max_position_embeddings": 262144,
2840
+ "model_type": "ministral3",
2841
+ "num_attention_heads": 32,
2842
+ "num_hidden_layers": 40,
2843
+ "num_key_value_heads": 8,
2844
+ "rms_norm_eps": 1e-05,
2845
+ "rope_parameters": {
2846
+ "beta_fast": 32.0,
2847
+ "beta_slow": 1.0,
2848
+ "factor": 16.0,
2849
+ "llama_4_scaling_beta": 0.1,
2850
+ "mscale": 1.0,
2851
+ "mscale_all_dim": 1.0,
2852
+ "original_max_position_embeddings": 16384,
2853
+ "rope_theta": 1000000000.0,
2854
+ "rope_type": "yarn",
2855
+ "type": "yarn"
2856
+ },
2857
+ "sliding_window": null,
2858
+ "use_cache": true,
2859
+ "vocab_size": 131072,
2860
+ "tie_word_embeddings": false
2861
+ },
2862
+ "tie_word_embeddings": false,
2863
+ "transformers_version": "5.0.0.dev0",
2864
+ "vision_feature_layer": -1
2865
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "eos_token_id": 2,
4
+ "max_length": 262144,
5
+ "pad_token_id": 11,
6
+ "transformers_version": "5.0.0.dev0"
7
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2581a3f8a5beb43f361c0dd2886cdff436bdbad90a2b46ca3e13864e3b1e1f41
3
+ size 5361479864
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e6eba9b62bd4e3ed0035efc5254d961768e21915feb6c9adc84c52c6d91561d
3
+ size 3552365268
model.safetensors.index.json ADDED
@@ -0,0 +1,933 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 8913725440,
4
+ "total_parameters": 13506073600
5
+ },
6
+ "weight_map": {
7
+ "language_model.lm_head.weight": "model-00002-of-00002.safetensors",
8
+ "language_model.model.embed_tokens.biases": "model-00001-of-00002.safetensors",
9
+ "language_model.model.embed_tokens.scales": "model-00001-of-00002.safetensors",
10
+ "language_model.model.embed_tokens.weight": "model-00001-of-00002.safetensors",
11
+ "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
12
+ "language_model.model.layers.0.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
13
+ "language_model.model.layers.0.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
14
+ "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
15
+ "language_model.model.layers.0.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
16
+ "language_model.model.layers.0.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
17
+ "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
18
+ "language_model.model.layers.0.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
19
+ "language_model.model.layers.0.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
20
+ "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
21
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
22
+ "language_model.model.layers.0.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
23
+ "language_model.model.layers.0.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
24
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
25
+ "language_model.model.layers.0.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
26
+ "language_model.model.layers.0.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
27
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
28
+ "language_model.model.layers.0.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
29
+ "language_model.model.layers.0.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
30
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
31
+ "language_model.model.layers.0.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
32
+ "language_model.model.layers.0.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
33
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
34
+ "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
35
+ "language_model.model.layers.1.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
36
+ "language_model.model.layers.1.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
37
+ "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
38
+ "language_model.model.layers.1.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
39
+ "language_model.model.layers.1.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
40
+ "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
41
+ "language_model.model.layers.1.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
42
+ "language_model.model.layers.1.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
43
+ "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
44
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
45
+ "language_model.model.layers.1.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
46
+ "language_model.model.layers.1.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
47
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
48
+ "language_model.model.layers.1.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
49
+ "language_model.model.layers.1.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
50
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
51
+ "language_model.model.layers.1.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
52
+ "language_model.model.layers.1.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
53
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
54
+ "language_model.model.layers.1.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
55
+ "language_model.model.layers.1.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
56
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
57
+ "language_model.model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
58
+ "language_model.model.layers.10.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
59
+ "language_model.model.layers.10.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
60
+ "language_model.model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
61
+ "language_model.model.layers.10.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
62
+ "language_model.model.layers.10.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
63
+ "language_model.model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
64
+ "language_model.model.layers.10.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
65
+ "language_model.model.layers.10.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
66
+ "language_model.model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
67
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
68
+ "language_model.model.layers.10.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
69
+ "language_model.model.layers.10.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
70
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
71
+ "language_model.model.layers.10.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
72
+ "language_model.model.layers.10.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
73
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
74
+ "language_model.model.layers.10.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
75
+ "language_model.model.layers.10.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
76
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
77
+ "language_model.model.layers.10.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
78
+ "language_model.model.layers.10.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
79
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
80
+ "language_model.model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
81
+ "language_model.model.layers.11.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
82
+ "language_model.model.layers.11.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
83
+ "language_model.model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
84
+ "language_model.model.layers.11.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
85
+ "language_model.model.layers.11.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
86
+ "language_model.model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
87
+ "language_model.model.layers.11.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
88
+ "language_model.model.layers.11.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
89
+ "language_model.model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
90
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
91
+ "language_model.model.layers.11.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
92
+ "language_model.model.layers.11.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
93
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
94
+ "language_model.model.layers.11.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
95
+ "language_model.model.layers.11.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
96
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
97
+ "language_model.model.layers.11.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
98
+ "language_model.model.layers.11.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
99
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
100
+ "language_model.model.layers.11.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
101
+ "language_model.model.layers.11.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
102
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
103
+ "language_model.model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
104
+ "language_model.model.layers.12.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
105
+ "language_model.model.layers.12.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
106
+ "language_model.model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
107
+ "language_model.model.layers.12.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
108
+ "language_model.model.layers.12.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
109
+ "language_model.model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
110
+ "language_model.model.layers.12.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
111
+ "language_model.model.layers.12.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
112
+ "language_model.model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
113
+ "language_model.model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
114
+ "language_model.model.layers.12.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
115
+ "language_model.model.layers.12.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
116
+ "language_model.model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
117
+ "language_model.model.layers.12.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
118
+ "language_model.model.layers.12.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
119
+ "language_model.model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
120
+ "language_model.model.layers.12.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
121
+ "language_model.model.layers.12.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
122
+ "language_model.model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
123
+ "language_model.model.layers.12.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
124
+ "language_model.model.layers.12.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
125
+ "language_model.model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
126
+ "language_model.model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
127
+ "language_model.model.layers.13.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
128
+ "language_model.model.layers.13.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
129
+ "language_model.model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
130
+ "language_model.model.layers.13.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
131
+ "language_model.model.layers.13.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
132
+ "language_model.model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
133
+ "language_model.model.layers.13.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
134
+ "language_model.model.layers.13.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
135
+ "language_model.model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
136
+ "language_model.model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
137
+ "language_model.model.layers.13.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
138
+ "language_model.model.layers.13.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
139
+ "language_model.model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
140
+ "language_model.model.layers.13.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
141
+ "language_model.model.layers.13.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
142
+ "language_model.model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
143
+ "language_model.model.layers.13.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
144
+ "language_model.model.layers.13.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
145
+ "language_model.model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
146
+ "language_model.model.layers.13.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
147
+ "language_model.model.layers.13.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
148
+ "language_model.model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
149
+ "language_model.model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
150
+ "language_model.model.layers.14.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
151
+ "language_model.model.layers.14.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
152
+ "language_model.model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
153
+ "language_model.model.layers.14.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
154
+ "language_model.model.layers.14.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
155
+ "language_model.model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
156
+ "language_model.model.layers.14.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
157
+ "language_model.model.layers.14.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
158
+ "language_model.model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
159
+ "language_model.model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
160
+ "language_model.model.layers.14.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
161
+ "language_model.model.layers.14.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
162
+ "language_model.model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
163
+ "language_model.model.layers.14.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
164
+ "language_model.model.layers.14.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
165
+ "language_model.model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
166
+ "language_model.model.layers.14.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
167
+ "language_model.model.layers.14.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
168
+ "language_model.model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
169
+ "language_model.model.layers.14.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
170
+ "language_model.model.layers.14.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
171
+ "language_model.model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
172
+ "language_model.model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
173
+ "language_model.model.layers.15.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
174
+ "language_model.model.layers.15.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
175
+ "language_model.model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
176
+ "language_model.model.layers.15.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
177
+ "language_model.model.layers.15.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
178
+ "language_model.model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
179
+ "language_model.model.layers.15.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
180
+ "language_model.model.layers.15.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
181
+ "language_model.model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
182
+ "language_model.model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
183
+ "language_model.model.layers.15.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
184
+ "language_model.model.layers.15.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
185
+ "language_model.model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
186
+ "language_model.model.layers.15.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
187
+ "language_model.model.layers.15.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
188
+ "language_model.model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
189
+ "language_model.model.layers.15.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
190
+ "language_model.model.layers.15.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
191
+ "language_model.model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
192
+ "language_model.model.layers.15.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
193
+ "language_model.model.layers.15.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
194
+ "language_model.model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
195
+ "language_model.model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
196
+ "language_model.model.layers.16.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
197
+ "language_model.model.layers.16.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
198
+ "language_model.model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
199
+ "language_model.model.layers.16.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
200
+ "language_model.model.layers.16.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
201
+ "language_model.model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
202
+ "language_model.model.layers.16.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
203
+ "language_model.model.layers.16.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
204
+ "language_model.model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
205
+ "language_model.model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
206
+ "language_model.model.layers.16.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
207
+ "language_model.model.layers.16.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
208
+ "language_model.model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
209
+ "language_model.model.layers.16.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
210
+ "language_model.model.layers.16.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
211
+ "language_model.model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
212
+ "language_model.model.layers.16.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
213
+ "language_model.model.layers.16.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
214
+ "language_model.model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
215
+ "language_model.model.layers.16.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
216
+ "language_model.model.layers.16.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
217
+ "language_model.model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
218
+ "language_model.model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
219
+ "language_model.model.layers.17.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
220
+ "language_model.model.layers.17.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
221
+ "language_model.model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
222
+ "language_model.model.layers.17.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
223
+ "language_model.model.layers.17.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
224
+ "language_model.model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
225
+ "language_model.model.layers.17.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
226
+ "language_model.model.layers.17.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
227
+ "language_model.model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
228
+ "language_model.model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
229
+ "language_model.model.layers.17.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
230
+ "language_model.model.layers.17.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
231
+ "language_model.model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
232
+ "language_model.model.layers.17.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
233
+ "language_model.model.layers.17.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
234
+ "language_model.model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
235
+ "language_model.model.layers.17.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
236
+ "language_model.model.layers.17.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
237
+ "language_model.model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
238
+ "language_model.model.layers.17.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
239
+ "language_model.model.layers.17.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
240
+ "language_model.model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
241
+ "language_model.model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
242
+ "language_model.model.layers.18.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
243
+ "language_model.model.layers.18.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
244
+ "language_model.model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
245
+ "language_model.model.layers.18.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
246
+ "language_model.model.layers.18.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
247
+ "language_model.model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
248
+ "language_model.model.layers.18.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
249
+ "language_model.model.layers.18.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
250
+ "language_model.model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
251
+ "language_model.model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
252
+ "language_model.model.layers.18.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
253
+ "language_model.model.layers.18.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
254
+ "language_model.model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
255
+ "language_model.model.layers.18.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
256
+ "language_model.model.layers.18.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
257
+ "language_model.model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
258
+ "language_model.model.layers.18.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
259
+ "language_model.model.layers.18.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
260
+ "language_model.model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
261
+ "language_model.model.layers.18.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
262
+ "language_model.model.layers.18.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
263
+ "language_model.model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
264
+ "language_model.model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
265
+ "language_model.model.layers.19.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
266
+ "language_model.model.layers.19.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
267
+ "language_model.model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
268
+ "language_model.model.layers.19.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
269
+ "language_model.model.layers.19.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
270
+ "language_model.model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
271
+ "language_model.model.layers.19.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
272
+ "language_model.model.layers.19.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
273
+ "language_model.model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
274
+ "language_model.model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
275
+ "language_model.model.layers.19.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
276
+ "language_model.model.layers.19.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
277
+ "language_model.model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
278
+ "language_model.model.layers.19.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
279
+ "language_model.model.layers.19.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
280
+ "language_model.model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
281
+ "language_model.model.layers.19.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
282
+ "language_model.model.layers.19.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
283
+ "language_model.model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
284
+ "language_model.model.layers.19.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
285
+ "language_model.model.layers.19.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
286
+ "language_model.model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
287
+ "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
288
+ "language_model.model.layers.2.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
289
+ "language_model.model.layers.2.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
290
+ "language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
291
+ "language_model.model.layers.2.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
292
+ "language_model.model.layers.2.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
293
+ "language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
294
+ "language_model.model.layers.2.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
295
+ "language_model.model.layers.2.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
296
+ "language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
297
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
298
+ "language_model.model.layers.2.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
299
+ "language_model.model.layers.2.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
300
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
301
+ "language_model.model.layers.2.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
302
+ "language_model.model.layers.2.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
303
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
304
+ "language_model.model.layers.2.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
305
+ "language_model.model.layers.2.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
306
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
307
+ "language_model.model.layers.2.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
308
+ "language_model.model.layers.2.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
309
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
310
+ "language_model.model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
311
+ "language_model.model.layers.20.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
312
+ "language_model.model.layers.20.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
313
+ "language_model.model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
314
+ "language_model.model.layers.20.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
315
+ "language_model.model.layers.20.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
316
+ "language_model.model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
317
+ "language_model.model.layers.20.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
318
+ "language_model.model.layers.20.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
319
+ "language_model.model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
320
+ "language_model.model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
321
+ "language_model.model.layers.20.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
322
+ "language_model.model.layers.20.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
323
+ "language_model.model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
324
+ "language_model.model.layers.20.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
325
+ "language_model.model.layers.20.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
326
+ "language_model.model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
327
+ "language_model.model.layers.20.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
328
+ "language_model.model.layers.20.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
329
+ "language_model.model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
330
+ "language_model.model.layers.20.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
331
+ "language_model.model.layers.20.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
332
+ "language_model.model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
333
+ "language_model.model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
334
+ "language_model.model.layers.21.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
335
+ "language_model.model.layers.21.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
336
+ "language_model.model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
337
+ "language_model.model.layers.21.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
338
+ "language_model.model.layers.21.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
339
+ "language_model.model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
340
+ "language_model.model.layers.21.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
341
+ "language_model.model.layers.21.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
342
+ "language_model.model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
343
+ "language_model.model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
344
+ "language_model.model.layers.21.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
345
+ "language_model.model.layers.21.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
346
+ "language_model.model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
347
+ "language_model.model.layers.21.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
348
+ "language_model.model.layers.21.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
349
+ "language_model.model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
350
+ "language_model.model.layers.21.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
351
+ "language_model.model.layers.21.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
352
+ "language_model.model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
353
+ "language_model.model.layers.21.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
354
+ "language_model.model.layers.21.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
355
+ "language_model.model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
356
+ "language_model.model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
357
+ "language_model.model.layers.22.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
358
+ "language_model.model.layers.22.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
359
+ "language_model.model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
360
+ "language_model.model.layers.22.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
361
+ "language_model.model.layers.22.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
362
+ "language_model.model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
363
+ "language_model.model.layers.22.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
364
+ "language_model.model.layers.22.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
365
+ "language_model.model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
366
+ "language_model.model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
367
+ "language_model.model.layers.22.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
368
+ "language_model.model.layers.22.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
369
+ "language_model.model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
370
+ "language_model.model.layers.22.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
371
+ "language_model.model.layers.22.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
372
+ "language_model.model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
373
+ "language_model.model.layers.22.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
374
+ "language_model.model.layers.22.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
375
+ "language_model.model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
376
+ "language_model.model.layers.22.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
377
+ "language_model.model.layers.22.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
378
+ "language_model.model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
379
+ "language_model.model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
380
+ "language_model.model.layers.23.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
381
+ "language_model.model.layers.23.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
382
+ "language_model.model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
383
+ "language_model.model.layers.23.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
384
+ "language_model.model.layers.23.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
385
+ "language_model.model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
386
+ "language_model.model.layers.23.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
387
+ "language_model.model.layers.23.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
388
+ "language_model.model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
389
+ "language_model.model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
390
+ "language_model.model.layers.23.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
391
+ "language_model.model.layers.23.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
392
+ "language_model.model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
393
+ "language_model.model.layers.23.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
394
+ "language_model.model.layers.23.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
395
+ "language_model.model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
396
+ "language_model.model.layers.23.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
397
+ "language_model.model.layers.23.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
398
+ "language_model.model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
399
+ "language_model.model.layers.23.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
400
+ "language_model.model.layers.23.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
401
+ "language_model.model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
402
+ "language_model.model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
403
+ "language_model.model.layers.24.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
404
+ "language_model.model.layers.24.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
405
+ "language_model.model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
406
+ "language_model.model.layers.24.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
407
+ "language_model.model.layers.24.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
408
+ "language_model.model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
409
+ "language_model.model.layers.24.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
410
+ "language_model.model.layers.24.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
411
+ "language_model.model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
412
+ "language_model.model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
413
+ "language_model.model.layers.24.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
414
+ "language_model.model.layers.24.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
415
+ "language_model.model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
416
+ "language_model.model.layers.24.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
417
+ "language_model.model.layers.24.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
418
+ "language_model.model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
419
+ "language_model.model.layers.24.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
420
+ "language_model.model.layers.24.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
421
+ "language_model.model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
422
+ "language_model.model.layers.24.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
423
+ "language_model.model.layers.24.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
424
+ "language_model.model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
425
+ "language_model.model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
426
+ "language_model.model.layers.25.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
427
+ "language_model.model.layers.25.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
428
+ "language_model.model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
429
+ "language_model.model.layers.25.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
430
+ "language_model.model.layers.25.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
431
+ "language_model.model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
432
+ "language_model.model.layers.25.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
433
+ "language_model.model.layers.25.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
434
+ "language_model.model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
435
+ "language_model.model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
436
+ "language_model.model.layers.25.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
437
+ "language_model.model.layers.25.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
438
+ "language_model.model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
439
+ "language_model.model.layers.25.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
440
+ "language_model.model.layers.25.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
441
+ "language_model.model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
442
+ "language_model.model.layers.25.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
443
+ "language_model.model.layers.25.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
444
+ "language_model.model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
445
+ "language_model.model.layers.25.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
446
+ "language_model.model.layers.25.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
447
+ "language_model.model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
448
+ "language_model.model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
449
+ "language_model.model.layers.26.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
450
+ "language_model.model.layers.26.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
451
+ "language_model.model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
452
+ "language_model.model.layers.26.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
453
+ "language_model.model.layers.26.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
454
+ "language_model.model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
455
+ "language_model.model.layers.26.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
456
+ "language_model.model.layers.26.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
457
+ "language_model.model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
458
+ "language_model.model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
459
+ "language_model.model.layers.26.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
460
+ "language_model.model.layers.26.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
461
+ "language_model.model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
462
+ "language_model.model.layers.26.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
463
+ "language_model.model.layers.26.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
464
+ "language_model.model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
465
+ "language_model.model.layers.26.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
466
+ "language_model.model.layers.26.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
467
+ "language_model.model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
468
+ "language_model.model.layers.26.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
469
+ "language_model.model.layers.26.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
470
+ "language_model.model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
471
+ "language_model.model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
472
+ "language_model.model.layers.27.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
473
+ "language_model.model.layers.27.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
474
+ "language_model.model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
475
+ "language_model.model.layers.27.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
476
+ "language_model.model.layers.27.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
477
+ "language_model.model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
478
+ "language_model.model.layers.27.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
479
+ "language_model.model.layers.27.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
480
+ "language_model.model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
481
+ "language_model.model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
482
+ "language_model.model.layers.27.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
483
+ "language_model.model.layers.27.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
484
+ "language_model.model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
485
+ "language_model.model.layers.27.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
486
+ "language_model.model.layers.27.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
487
+ "language_model.model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
488
+ "language_model.model.layers.27.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
489
+ "language_model.model.layers.27.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
490
+ "language_model.model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
491
+ "language_model.model.layers.27.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
492
+ "language_model.model.layers.27.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
493
+ "language_model.model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
494
+ "language_model.model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
495
+ "language_model.model.layers.28.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
496
+ "language_model.model.layers.28.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
497
+ "language_model.model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
498
+ "language_model.model.layers.28.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
499
+ "language_model.model.layers.28.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
500
+ "language_model.model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
501
+ "language_model.model.layers.28.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
502
+ "language_model.model.layers.28.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
503
+ "language_model.model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
504
+ "language_model.model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
505
+ "language_model.model.layers.28.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
506
+ "language_model.model.layers.28.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
507
+ "language_model.model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
508
+ "language_model.model.layers.28.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
509
+ "language_model.model.layers.28.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
510
+ "language_model.model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
511
+ "language_model.model.layers.28.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
512
+ "language_model.model.layers.28.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
513
+ "language_model.model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
514
+ "language_model.model.layers.28.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
515
+ "language_model.model.layers.28.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
516
+ "language_model.model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
517
+ "language_model.model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
518
+ "language_model.model.layers.29.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
519
+ "language_model.model.layers.29.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
520
+ "language_model.model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
521
+ "language_model.model.layers.29.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
522
+ "language_model.model.layers.29.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
523
+ "language_model.model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
524
+ "language_model.model.layers.29.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
525
+ "language_model.model.layers.29.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
526
+ "language_model.model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
527
+ "language_model.model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
528
+ "language_model.model.layers.29.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
529
+ "language_model.model.layers.29.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
530
+ "language_model.model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
531
+ "language_model.model.layers.29.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
532
+ "language_model.model.layers.29.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
533
+ "language_model.model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
534
+ "language_model.model.layers.29.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
535
+ "language_model.model.layers.29.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
536
+ "language_model.model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
537
+ "language_model.model.layers.29.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
538
+ "language_model.model.layers.29.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
539
+ "language_model.model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
540
+ "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
541
+ "language_model.model.layers.3.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
542
+ "language_model.model.layers.3.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
543
+ "language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
544
+ "language_model.model.layers.3.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
545
+ "language_model.model.layers.3.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
546
+ "language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
547
+ "language_model.model.layers.3.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
548
+ "language_model.model.layers.3.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
549
+ "language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
550
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
551
+ "language_model.model.layers.3.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
552
+ "language_model.model.layers.3.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
553
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
554
+ "language_model.model.layers.3.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
555
+ "language_model.model.layers.3.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
556
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
557
+ "language_model.model.layers.3.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
558
+ "language_model.model.layers.3.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
559
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
560
+ "language_model.model.layers.3.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
561
+ "language_model.model.layers.3.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
562
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
563
+ "language_model.model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
564
+ "language_model.model.layers.30.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
565
+ "language_model.model.layers.30.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
566
+ "language_model.model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
567
+ "language_model.model.layers.30.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
568
+ "language_model.model.layers.30.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
569
+ "language_model.model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
570
+ "language_model.model.layers.30.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
571
+ "language_model.model.layers.30.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
572
+ "language_model.model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
573
+ "language_model.model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
574
+ "language_model.model.layers.30.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
575
+ "language_model.model.layers.30.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
576
+ "language_model.model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
577
+ "language_model.model.layers.30.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
578
+ "language_model.model.layers.30.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
579
+ "language_model.model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
580
+ "language_model.model.layers.30.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
581
+ "language_model.model.layers.30.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
582
+ "language_model.model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
583
+ "language_model.model.layers.30.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
584
+ "language_model.model.layers.30.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
585
+ "language_model.model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
586
+ "language_model.model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
587
+ "language_model.model.layers.31.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
588
+ "language_model.model.layers.31.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
589
+ "language_model.model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
590
+ "language_model.model.layers.31.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
591
+ "language_model.model.layers.31.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
592
+ "language_model.model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
593
+ "language_model.model.layers.31.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
594
+ "language_model.model.layers.31.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
595
+ "language_model.model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
596
+ "language_model.model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
597
+ "language_model.model.layers.31.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
598
+ "language_model.model.layers.31.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
599
+ "language_model.model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
600
+ "language_model.model.layers.31.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
601
+ "language_model.model.layers.31.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
602
+ "language_model.model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
603
+ "language_model.model.layers.31.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
604
+ "language_model.model.layers.31.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
605
+ "language_model.model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
606
+ "language_model.model.layers.31.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
607
+ "language_model.model.layers.31.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
608
+ "language_model.model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
609
+ "language_model.model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
610
+ "language_model.model.layers.32.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
611
+ "language_model.model.layers.32.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
612
+ "language_model.model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
613
+ "language_model.model.layers.32.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
614
+ "language_model.model.layers.32.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
615
+ "language_model.model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
616
+ "language_model.model.layers.32.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
617
+ "language_model.model.layers.32.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
618
+ "language_model.model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
619
+ "language_model.model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
620
+ "language_model.model.layers.32.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
621
+ "language_model.model.layers.32.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
622
+ "language_model.model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
623
+ "language_model.model.layers.32.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
624
+ "language_model.model.layers.32.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
625
+ "language_model.model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
626
+ "language_model.model.layers.32.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
627
+ "language_model.model.layers.32.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
628
+ "language_model.model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
629
+ "language_model.model.layers.32.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
630
+ "language_model.model.layers.32.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
631
+ "language_model.model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
632
+ "language_model.model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
633
+ "language_model.model.layers.33.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
634
+ "language_model.model.layers.33.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
635
+ "language_model.model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
636
+ "language_model.model.layers.33.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
637
+ "language_model.model.layers.33.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
638
+ "language_model.model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
639
+ "language_model.model.layers.33.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
640
+ "language_model.model.layers.33.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
641
+ "language_model.model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
642
+ "language_model.model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
643
+ "language_model.model.layers.33.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
644
+ "language_model.model.layers.33.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
645
+ "language_model.model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
646
+ "language_model.model.layers.33.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
647
+ "language_model.model.layers.33.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
648
+ "language_model.model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
649
+ "language_model.model.layers.33.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
650
+ "language_model.model.layers.33.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
651
+ "language_model.model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
652
+ "language_model.model.layers.33.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
653
+ "language_model.model.layers.33.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
654
+ "language_model.model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
655
+ "language_model.model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
656
+ "language_model.model.layers.34.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
657
+ "language_model.model.layers.34.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
658
+ "language_model.model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
659
+ "language_model.model.layers.34.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
660
+ "language_model.model.layers.34.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
661
+ "language_model.model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
662
+ "language_model.model.layers.34.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
663
+ "language_model.model.layers.34.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
664
+ "language_model.model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
665
+ "language_model.model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
666
+ "language_model.model.layers.34.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
667
+ "language_model.model.layers.34.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
668
+ "language_model.model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
669
+ "language_model.model.layers.34.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
670
+ "language_model.model.layers.34.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
671
+ "language_model.model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
672
+ "language_model.model.layers.34.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
673
+ "language_model.model.layers.34.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
674
+ "language_model.model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
675
+ "language_model.model.layers.34.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
676
+ "language_model.model.layers.34.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
677
+ "language_model.model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
678
+ "language_model.model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
679
+ "language_model.model.layers.35.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
680
+ "language_model.model.layers.35.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
681
+ "language_model.model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
682
+ "language_model.model.layers.35.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
683
+ "language_model.model.layers.35.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
684
+ "language_model.model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
685
+ "language_model.model.layers.35.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
686
+ "language_model.model.layers.35.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
687
+ "language_model.model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
688
+ "language_model.model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
689
+ "language_model.model.layers.35.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
690
+ "language_model.model.layers.35.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
691
+ "language_model.model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
692
+ "language_model.model.layers.35.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
693
+ "language_model.model.layers.35.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
694
+ "language_model.model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
695
+ "language_model.model.layers.35.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
696
+ "language_model.model.layers.35.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
697
+ "language_model.model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
698
+ "language_model.model.layers.35.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
699
+ "language_model.model.layers.35.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
700
+ "language_model.model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
701
+ "language_model.model.layers.36.input_layernorm.weight": "model-00002-of-00002.safetensors",
702
+ "language_model.model.layers.36.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
703
+ "language_model.model.layers.36.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
704
+ "language_model.model.layers.36.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
705
+ "language_model.model.layers.36.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
706
+ "language_model.model.layers.36.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
707
+ "language_model.model.layers.36.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
708
+ "language_model.model.layers.36.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
709
+ "language_model.model.layers.36.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
710
+ "language_model.model.layers.36.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
711
+ "language_model.model.layers.36.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
712
+ "language_model.model.layers.36.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
713
+ "language_model.model.layers.36.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
714
+ "language_model.model.layers.36.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
715
+ "language_model.model.layers.36.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
716
+ "language_model.model.layers.36.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
717
+ "language_model.model.layers.36.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
718
+ "language_model.model.layers.36.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
719
+ "language_model.model.layers.36.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
720
+ "language_model.model.layers.36.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
721
+ "language_model.model.layers.36.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
722
+ "language_model.model.layers.36.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
723
+ "language_model.model.layers.36.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
724
+ "language_model.model.layers.37.input_layernorm.weight": "model-00002-of-00002.safetensors",
725
+ "language_model.model.layers.37.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
726
+ "language_model.model.layers.37.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
727
+ "language_model.model.layers.37.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
728
+ "language_model.model.layers.37.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
729
+ "language_model.model.layers.37.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
730
+ "language_model.model.layers.37.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
731
+ "language_model.model.layers.37.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
732
+ "language_model.model.layers.37.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
733
+ "language_model.model.layers.37.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
734
+ "language_model.model.layers.37.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
735
+ "language_model.model.layers.37.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
736
+ "language_model.model.layers.37.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
737
+ "language_model.model.layers.37.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
738
+ "language_model.model.layers.37.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
739
+ "language_model.model.layers.37.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
740
+ "language_model.model.layers.37.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
741
+ "language_model.model.layers.37.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
742
+ "language_model.model.layers.37.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
743
+ "language_model.model.layers.37.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
744
+ "language_model.model.layers.37.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
745
+ "language_model.model.layers.37.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
746
+ "language_model.model.layers.37.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
747
+ "language_model.model.layers.38.input_layernorm.weight": "model-00002-of-00002.safetensors",
748
+ "language_model.model.layers.38.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
749
+ "language_model.model.layers.38.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
750
+ "language_model.model.layers.38.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
751
+ "language_model.model.layers.38.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
752
+ "language_model.model.layers.38.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
753
+ "language_model.model.layers.38.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
754
+ "language_model.model.layers.38.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
755
+ "language_model.model.layers.38.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
756
+ "language_model.model.layers.38.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
757
+ "language_model.model.layers.38.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
758
+ "language_model.model.layers.38.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
759
+ "language_model.model.layers.38.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
760
+ "language_model.model.layers.38.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
761
+ "language_model.model.layers.38.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
762
+ "language_model.model.layers.38.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
763
+ "language_model.model.layers.38.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
764
+ "language_model.model.layers.38.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
765
+ "language_model.model.layers.38.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
766
+ "language_model.model.layers.38.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
767
+ "language_model.model.layers.38.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
768
+ "language_model.model.layers.38.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
769
+ "language_model.model.layers.38.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
770
+ "language_model.model.layers.39.input_layernorm.weight": "model-00002-of-00002.safetensors",
771
+ "language_model.model.layers.39.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
772
+ "language_model.model.layers.39.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
773
+ "language_model.model.layers.39.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
774
+ "language_model.model.layers.39.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
775
+ "language_model.model.layers.39.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
776
+ "language_model.model.layers.39.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
777
+ "language_model.model.layers.39.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
778
+ "language_model.model.layers.39.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
779
+ "language_model.model.layers.39.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
780
+ "language_model.model.layers.39.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
781
+ "language_model.model.layers.39.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
782
+ "language_model.model.layers.39.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
783
+ "language_model.model.layers.39.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
784
+ "language_model.model.layers.39.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
785
+ "language_model.model.layers.39.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
786
+ "language_model.model.layers.39.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
787
+ "language_model.model.layers.39.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
788
+ "language_model.model.layers.39.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
789
+ "language_model.model.layers.39.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
790
+ "language_model.model.layers.39.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
791
+ "language_model.model.layers.39.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
792
+ "language_model.model.layers.39.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
793
+ "language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
794
+ "language_model.model.layers.4.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
795
+ "language_model.model.layers.4.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
796
+ "language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
797
+ "language_model.model.layers.4.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
798
+ "language_model.model.layers.4.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
799
+ "language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
800
+ "language_model.model.layers.4.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
801
+ "language_model.model.layers.4.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
802
+ "language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
803
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
804
+ "language_model.model.layers.4.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
805
+ "language_model.model.layers.4.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
806
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
807
+ "language_model.model.layers.4.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
808
+ "language_model.model.layers.4.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
809
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
810
+ "language_model.model.layers.4.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
811
+ "language_model.model.layers.4.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
812
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
813
+ "language_model.model.layers.4.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
814
+ "language_model.model.layers.4.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
815
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
816
+ "language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
817
+ "language_model.model.layers.5.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
818
+ "language_model.model.layers.5.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
819
+ "language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
820
+ "language_model.model.layers.5.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
821
+ "language_model.model.layers.5.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
822
+ "language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
823
+ "language_model.model.layers.5.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
824
+ "language_model.model.layers.5.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
825
+ "language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
826
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
827
+ "language_model.model.layers.5.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
828
+ "language_model.model.layers.5.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
829
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
830
+ "language_model.model.layers.5.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
831
+ "language_model.model.layers.5.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
832
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
833
+ "language_model.model.layers.5.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
834
+ "language_model.model.layers.5.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
835
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
836
+ "language_model.model.layers.5.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
837
+ "language_model.model.layers.5.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
838
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
839
+ "language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
840
+ "language_model.model.layers.6.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
841
+ "language_model.model.layers.6.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
842
+ "language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
843
+ "language_model.model.layers.6.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
844
+ "language_model.model.layers.6.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
845
+ "language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
846
+ "language_model.model.layers.6.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
847
+ "language_model.model.layers.6.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
848
+ "language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
849
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
850
+ "language_model.model.layers.6.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
851
+ "language_model.model.layers.6.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
852
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
853
+ "language_model.model.layers.6.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
854
+ "language_model.model.layers.6.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
855
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
856
+ "language_model.model.layers.6.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
857
+ "language_model.model.layers.6.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
858
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
859
+ "language_model.model.layers.6.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
860
+ "language_model.model.layers.6.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
861
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
862
+ "language_model.model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
863
+ "language_model.model.layers.7.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
864
+ "language_model.model.layers.7.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
865
+ "language_model.model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
866
+ "language_model.model.layers.7.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
867
+ "language_model.model.layers.7.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
868
+ "language_model.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
869
+ "language_model.model.layers.7.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
870
+ "language_model.model.layers.7.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
871
+ "language_model.model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
872
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
873
+ "language_model.model.layers.7.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
874
+ "language_model.model.layers.7.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
875
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
876
+ "language_model.model.layers.7.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
877
+ "language_model.model.layers.7.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
878
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
879
+ "language_model.model.layers.7.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
880
+ "language_model.model.layers.7.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
881
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
882
+ "language_model.model.layers.7.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
883
+ "language_model.model.layers.7.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
884
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
885
+ "language_model.model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
886
+ "language_model.model.layers.8.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
887
+ "language_model.model.layers.8.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
888
+ "language_model.model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
889
+ "language_model.model.layers.8.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
890
+ "language_model.model.layers.8.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
891
+ "language_model.model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
892
+ "language_model.model.layers.8.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
893
+ "language_model.model.layers.8.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
894
+ "language_model.model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
895
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
896
+ "language_model.model.layers.8.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
897
+ "language_model.model.layers.8.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
898
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
899
+ "language_model.model.layers.8.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
900
+ "language_model.model.layers.8.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
901
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
902
+ "language_model.model.layers.8.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
903
+ "language_model.model.layers.8.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
904
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
905
+ "language_model.model.layers.8.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
906
+ "language_model.model.layers.8.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
907
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
908
+ "language_model.model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
909
+ "language_model.model.layers.9.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
910
+ "language_model.model.layers.9.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
911
+ "language_model.model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
912
+ "language_model.model.layers.9.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
913
+ "language_model.model.layers.9.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
914
+ "language_model.model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
915
+ "language_model.model.layers.9.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
916
+ "language_model.model.layers.9.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
917
+ "language_model.model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
918
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
919
+ "language_model.model.layers.9.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
920
+ "language_model.model.layers.9.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
921
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
922
+ "language_model.model.layers.9.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
923
+ "language_model.model.layers.9.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
924
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
925
+ "language_model.model.layers.9.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
926
+ "language_model.model.layers.9.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
927
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
928
+ "language_model.model.layers.9.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
929
+ "language_model.model.layers.9.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
930
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
931
+ "language_model.model.norm.weight": "model-00002-of-00002.safetensors"
932
+ }
933
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99cf274236c60277fcfad861a5a1007518687ad06ba8938760f50b55ffa0b1ef
3
+ size 17077420
tokenizer_config.json ADDED
@@ -0,0 +1,1015 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "extra_special_tokens": [
6
+ "<unk>",
7
+ "<s>",
8
+ "</s>",
9
+ "[INST]",
10
+ "[/INST]",
11
+ "[AVAILABLE_TOOLS]",
12
+ "[/AVAILABLE_TOOLS]",
13
+ "[TOOL_RESULTS]",
14
+ "[/TOOL_RESULTS]",
15
+ "[TOOL_CALLS]",
16
+ "[IMG]",
17
+ "<pad>",
18
+ "[IMG_BREAK]",
19
+ "[IMG_END]",
20
+ "[PREFIX]",
21
+ "[MIDDLE]",
22
+ "[SUFFIX]",
23
+ "[SYSTEM_PROMPT]",
24
+ "[/SYSTEM_PROMPT]",
25
+ "[TOOL_CONTENT]",
26
+ "<SPECIAL_20>",
27
+ "<SPECIAL_21>",
28
+ "<SPECIAL_22>",
29
+ "<SPECIAL_23>",
30
+ "[AUDIO]",
31
+ "[BEGIN_AUDIO]",
32
+ "<SPECIAL_26>",
33
+ "<SPECIAL_27>",
34
+ "<SPECIAL_28>",
35
+ "<SPECIAL_29>",
36
+ "<SPECIAL_30>",
37
+ "<SPECIAL_31>",
38
+ "[ARGS]",
39
+ "[CALL_ID]",
40
+ "<SPECIAL_34>",
41
+ "<SPECIAL_35>",
42
+ "<SPECIAL_36>",
43
+ "<SPECIAL_37>",
44
+ "<SPECIAL_38>",
45
+ "<SPECIAL_39>",
46
+ "<SPECIAL_40>",
47
+ "<SPECIAL_41>",
48
+ "<SPECIAL_42>",
49
+ "<SPECIAL_43>",
50
+ "<SPECIAL_44>",
51
+ "<SPECIAL_45>",
52
+ "<SPECIAL_46>",
53
+ "<SPECIAL_47>",
54
+ "<SPECIAL_48>",
55
+ "<SPECIAL_49>",
56
+ "<SPECIAL_50>",
57
+ "<SPECIAL_51>",
58
+ "<SPECIAL_52>",
59
+ "<SPECIAL_53>",
60
+ "<SPECIAL_54>",
61
+ "<SPECIAL_55>",
62
+ "<SPECIAL_56>",
63
+ "<SPECIAL_57>",
64
+ "<SPECIAL_58>",
65
+ "<SPECIAL_59>",
66
+ "<SPECIAL_60>",
67
+ "<SPECIAL_61>",
68
+ "<SPECIAL_62>",
69
+ "<SPECIAL_63>",
70
+ "<SPECIAL_64>",
71
+ "<SPECIAL_65>",
72
+ "<SPECIAL_66>",
73
+ "<SPECIAL_67>",
74
+ "<SPECIAL_68>",
75
+ "<SPECIAL_69>",
76
+ "<SPECIAL_70>",
77
+ "<SPECIAL_71>",
78
+ "<SPECIAL_72>",
79
+ "<SPECIAL_73>",
80
+ "<SPECIAL_74>",
81
+ "<SPECIAL_75>",
82
+ "<SPECIAL_76>",
83
+ "<SPECIAL_77>",
84
+ "<SPECIAL_78>",
85
+ "<SPECIAL_79>",
86
+ "<SPECIAL_80>",
87
+ "<SPECIAL_81>",
88
+ "<SPECIAL_82>",
89
+ "<SPECIAL_83>",
90
+ "<SPECIAL_84>",
91
+ "<SPECIAL_85>",
92
+ "<SPECIAL_86>",
93
+ "<SPECIAL_87>",
94
+ "<SPECIAL_88>",
95
+ "<SPECIAL_89>",
96
+ "<SPECIAL_90>",
97
+ "<SPECIAL_91>",
98
+ "<SPECIAL_92>",
99
+ "<SPECIAL_93>",
100
+ "<SPECIAL_94>",
101
+ "<SPECIAL_95>",
102
+ "<SPECIAL_96>",
103
+ "<SPECIAL_97>",
104
+ "<SPECIAL_98>",
105
+ "<SPECIAL_99>",
106
+ "<SPECIAL_100>",
107
+ "<SPECIAL_101>",
108
+ "<SPECIAL_102>",
109
+ "<SPECIAL_103>",
110
+ "<SPECIAL_104>",
111
+ "<SPECIAL_105>",
112
+ "<SPECIAL_106>",
113
+ "<SPECIAL_107>",
114
+ "<SPECIAL_108>",
115
+ "<SPECIAL_109>",
116
+ "<SPECIAL_110>",
117
+ "<SPECIAL_111>",
118
+ "<SPECIAL_112>",
119
+ "<SPECIAL_113>",
120
+ "<SPECIAL_114>",
121
+ "<SPECIAL_115>",
122
+ "<SPECIAL_116>",
123
+ "<SPECIAL_117>",
124
+ "<SPECIAL_118>",
125
+ "<SPECIAL_119>",
126
+ "<SPECIAL_120>",
127
+ "<SPECIAL_121>",
128
+ "<SPECIAL_122>",
129
+ "<SPECIAL_123>",
130
+ "<SPECIAL_124>",
131
+ "<SPECIAL_125>",
132
+ "<SPECIAL_126>",
133
+ "<SPECIAL_127>",
134
+ "<SPECIAL_128>",
135
+ "<SPECIAL_129>",
136
+ "<SPECIAL_130>",
137
+ "<SPECIAL_131>",
138
+ "<SPECIAL_132>",
139
+ "<SPECIAL_133>",
140
+ "<SPECIAL_134>",
141
+ "<SPECIAL_135>",
142
+ "<SPECIAL_136>",
143
+ "<SPECIAL_137>",
144
+ "<SPECIAL_138>",
145
+ "<SPECIAL_139>",
146
+ "<SPECIAL_140>",
147
+ "<SPECIAL_141>",
148
+ "<SPECIAL_142>",
149
+ "<SPECIAL_143>",
150
+ "<SPECIAL_144>",
151
+ "<SPECIAL_145>",
152
+ "<SPECIAL_146>",
153
+ "<SPECIAL_147>",
154
+ "<SPECIAL_148>",
155
+ "<SPECIAL_149>",
156
+ "<SPECIAL_150>",
157
+ "<SPECIAL_151>",
158
+ "<SPECIAL_152>",
159
+ "<SPECIAL_153>",
160
+ "<SPECIAL_154>",
161
+ "<SPECIAL_155>",
162
+ "<SPECIAL_156>",
163
+ "<SPECIAL_157>",
164
+ "<SPECIAL_158>",
165
+ "<SPECIAL_159>",
166
+ "<SPECIAL_160>",
167
+ "<SPECIAL_161>",
168
+ "<SPECIAL_162>",
169
+ "<SPECIAL_163>",
170
+ "<SPECIAL_164>",
171
+ "<SPECIAL_165>",
172
+ "<SPECIAL_166>",
173
+ "<SPECIAL_167>",
174
+ "<SPECIAL_168>",
175
+ "<SPECIAL_169>",
176
+ "<SPECIAL_170>",
177
+ "<SPECIAL_171>",
178
+ "<SPECIAL_172>",
179
+ "<SPECIAL_173>",
180
+ "<SPECIAL_174>",
181
+ "<SPECIAL_175>",
182
+ "<SPECIAL_176>",
183
+ "<SPECIAL_177>",
184
+ "<SPECIAL_178>",
185
+ "<SPECIAL_179>",
186
+ "<SPECIAL_180>",
187
+ "<SPECIAL_181>",
188
+ "<SPECIAL_182>",
189
+ "<SPECIAL_183>",
190
+ "<SPECIAL_184>",
191
+ "<SPECIAL_185>",
192
+ "<SPECIAL_186>",
193
+ "<SPECIAL_187>",
194
+ "<SPECIAL_188>",
195
+ "<SPECIAL_189>",
196
+ "<SPECIAL_190>",
197
+ "<SPECIAL_191>",
198
+ "<SPECIAL_192>",
199
+ "<SPECIAL_193>",
200
+ "<SPECIAL_194>",
201
+ "<SPECIAL_195>",
202
+ "<SPECIAL_196>",
203
+ "<SPECIAL_197>",
204
+ "<SPECIAL_198>",
205
+ "<SPECIAL_199>",
206
+ "<SPECIAL_200>",
207
+ "<SPECIAL_201>",
208
+ "<SPECIAL_202>",
209
+ "<SPECIAL_203>",
210
+ "<SPECIAL_204>",
211
+ "<SPECIAL_205>",
212
+ "<SPECIAL_206>",
213
+ "<SPECIAL_207>",
214
+ "<SPECIAL_208>",
215
+ "<SPECIAL_209>",
216
+ "<SPECIAL_210>",
217
+ "<SPECIAL_211>",
218
+ "<SPECIAL_212>",
219
+ "<SPECIAL_213>",
220
+ "<SPECIAL_214>",
221
+ "<SPECIAL_215>",
222
+ "<SPECIAL_216>",
223
+ "<SPECIAL_217>",
224
+ "<SPECIAL_218>",
225
+ "<SPECIAL_219>",
226
+ "<SPECIAL_220>",
227
+ "<SPECIAL_221>",
228
+ "<SPECIAL_222>",
229
+ "<SPECIAL_223>",
230
+ "<SPECIAL_224>",
231
+ "<SPECIAL_225>",
232
+ "<SPECIAL_226>",
233
+ "<SPECIAL_227>",
234
+ "<SPECIAL_228>",
235
+ "<SPECIAL_229>",
236
+ "<SPECIAL_230>",
237
+ "<SPECIAL_231>",
238
+ "<SPECIAL_232>",
239
+ "<SPECIAL_233>",
240
+ "<SPECIAL_234>",
241
+ "<SPECIAL_235>",
242
+ "<SPECIAL_236>",
243
+ "<SPECIAL_237>",
244
+ "<SPECIAL_238>",
245
+ "<SPECIAL_239>",
246
+ "<SPECIAL_240>",
247
+ "<SPECIAL_241>",
248
+ "<SPECIAL_242>",
249
+ "<SPECIAL_243>",
250
+ "<SPECIAL_244>",
251
+ "<SPECIAL_245>",
252
+ "<SPECIAL_246>",
253
+ "<SPECIAL_247>",
254
+ "<SPECIAL_248>",
255
+ "<SPECIAL_249>",
256
+ "<SPECIAL_250>",
257
+ "<SPECIAL_251>",
258
+ "<SPECIAL_252>",
259
+ "<SPECIAL_253>",
260
+ "<SPECIAL_254>",
261
+ "<SPECIAL_255>",
262
+ "<SPECIAL_256>",
263
+ "<SPECIAL_257>",
264
+ "<SPECIAL_258>",
265
+ "<SPECIAL_259>",
266
+ "<SPECIAL_260>",
267
+ "<SPECIAL_261>",
268
+ "<SPECIAL_262>",
269
+ "<SPECIAL_263>",
270
+ "<SPECIAL_264>",
271
+ "<SPECIAL_265>",
272
+ "<SPECIAL_266>",
273
+ "<SPECIAL_267>",
274
+ "<SPECIAL_268>",
275
+ "<SPECIAL_269>",
276
+ "<SPECIAL_270>",
277
+ "<SPECIAL_271>",
278
+ "<SPECIAL_272>",
279
+ "<SPECIAL_273>",
280
+ "<SPECIAL_274>",
281
+ "<SPECIAL_275>",
282
+ "<SPECIAL_276>",
283
+ "<SPECIAL_277>",
284
+ "<SPECIAL_278>",
285
+ "<SPECIAL_279>",
286
+ "<SPECIAL_280>",
287
+ "<SPECIAL_281>",
288
+ "<SPECIAL_282>",
289
+ "<SPECIAL_283>",
290
+ "<SPECIAL_284>",
291
+ "<SPECIAL_285>",
292
+ "<SPECIAL_286>",
293
+ "<SPECIAL_287>",
294
+ "<SPECIAL_288>",
295
+ "<SPECIAL_289>",
296
+ "<SPECIAL_290>",
297
+ "<SPECIAL_291>",
298
+ "<SPECIAL_292>",
299
+ "<SPECIAL_293>",
300
+ "<SPECIAL_294>",
301
+ "<SPECIAL_295>",
302
+ "<SPECIAL_296>",
303
+ "<SPECIAL_297>",
304
+ "<SPECIAL_298>",
305
+ "<SPECIAL_299>",
306
+ "<SPECIAL_300>",
307
+ "<SPECIAL_301>",
308
+ "<SPECIAL_302>",
309
+ "<SPECIAL_303>",
310
+ "<SPECIAL_304>",
311
+ "<SPECIAL_305>",
312
+ "<SPECIAL_306>",
313
+ "<SPECIAL_307>",
314
+ "<SPECIAL_308>",
315
+ "<SPECIAL_309>",
316
+ "<SPECIAL_310>",
317
+ "<SPECIAL_311>",
318
+ "<SPECIAL_312>",
319
+ "<SPECIAL_313>",
320
+ "<SPECIAL_314>",
321
+ "<SPECIAL_315>",
322
+ "<SPECIAL_316>",
323
+ "<SPECIAL_317>",
324
+ "<SPECIAL_318>",
325
+ "<SPECIAL_319>",
326
+ "<SPECIAL_320>",
327
+ "<SPECIAL_321>",
328
+ "<SPECIAL_322>",
329
+ "<SPECIAL_323>",
330
+ "<SPECIAL_324>",
331
+ "<SPECIAL_325>",
332
+ "<SPECIAL_326>",
333
+ "<SPECIAL_327>",
334
+ "<SPECIAL_328>",
335
+ "<SPECIAL_329>",
336
+ "<SPECIAL_330>",
337
+ "<SPECIAL_331>",
338
+ "<SPECIAL_332>",
339
+ "<SPECIAL_333>",
340
+ "<SPECIAL_334>",
341
+ "<SPECIAL_335>",
342
+ "<SPECIAL_336>",
343
+ "<SPECIAL_337>",
344
+ "<SPECIAL_338>",
345
+ "<SPECIAL_339>",
346
+ "<SPECIAL_340>",
347
+ "<SPECIAL_341>",
348
+ "<SPECIAL_342>",
349
+ "<SPECIAL_343>",
350
+ "<SPECIAL_344>",
351
+ "<SPECIAL_345>",
352
+ "<SPECIAL_346>",
353
+ "<SPECIAL_347>",
354
+ "<SPECIAL_348>",
355
+ "<SPECIAL_349>",
356
+ "<SPECIAL_350>",
357
+ "<SPECIAL_351>",
358
+ "<SPECIAL_352>",
359
+ "<SPECIAL_353>",
360
+ "<SPECIAL_354>",
361
+ "<SPECIAL_355>",
362
+ "<SPECIAL_356>",
363
+ "<SPECIAL_357>",
364
+ "<SPECIAL_358>",
365
+ "<SPECIAL_359>",
366
+ "<SPECIAL_360>",
367
+ "<SPECIAL_361>",
368
+ "<SPECIAL_362>",
369
+ "<SPECIAL_363>",
370
+ "<SPECIAL_364>",
371
+ "<SPECIAL_365>",
372
+ "<SPECIAL_366>",
373
+ "<SPECIAL_367>",
374
+ "<SPECIAL_368>",
375
+ "<SPECIAL_369>",
376
+ "<SPECIAL_370>",
377
+ "<SPECIAL_371>",
378
+ "<SPECIAL_372>",
379
+ "<SPECIAL_373>",
380
+ "<SPECIAL_374>",
381
+ "<SPECIAL_375>",
382
+ "<SPECIAL_376>",
383
+ "<SPECIAL_377>",
384
+ "<SPECIAL_378>",
385
+ "<SPECIAL_379>",
386
+ "<SPECIAL_380>",
387
+ "<SPECIAL_381>",
388
+ "<SPECIAL_382>",
389
+ "<SPECIAL_383>",
390
+ "<SPECIAL_384>",
391
+ "<SPECIAL_385>",
392
+ "<SPECIAL_386>",
393
+ "<SPECIAL_387>",
394
+ "<SPECIAL_388>",
395
+ "<SPECIAL_389>",
396
+ "<SPECIAL_390>",
397
+ "<SPECIAL_391>",
398
+ "<SPECIAL_392>",
399
+ "<SPECIAL_393>",
400
+ "<SPECIAL_394>",
401
+ "<SPECIAL_395>",
402
+ "<SPECIAL_396>",
403
+ "<SPECIAL_397>",
404
+ "<SPECIAL_398>",
405
+ "<SPECIAL_399>",
406
+ "<SPECIAL_400>",
407
+ "<SPECIAL_401>",
408
+ "<SPECIAL_402>",
409
+ "<SPECIAL_403>",
410
+ "<SPECIAL_404>",
411
+ "<SPECIAL_405>",
412
+ "<SPECIAL_406>",
413
+ "<SPECIAL_407>",
414
+ "<SPECIAL_408>",
415
+ "<SPECIAL_409>",
416
+ "<SPECIAL_410>",
417
+ "<SPECIAL_411>",
418
+ "<SPECIAL_412>",
419
+ "<SPECIAL_413>",
420
+ "<SPECIAL_414>",
421
+ "<SPECIAL_415>",
422
+ "<SPECIAL_416>",
423
+ "<SPECIAL_417>",
424
+ "<SPECIAL_418>",
425
+ "<SPECIAL_419>",
426
+ "<SPECIAL_420>",
427
+ "<SPECIAL_421>",
428
+ "<SPECIAL_422>",
429
+ "<SPECIAL_423>",
430
+ "<SPECIAL_424>",
431
+ "<SPECIAL_425>",
432
+ "<SPECIAL_426>",
433
+ "<SPECIAL_427>",
434
+ "<SPECIAL_428>",
435
+ "<SPECIAL_429>",
436
+ "<SPECIAL_430>",
437
+ "<SPECIAL_431>",
438
+ "<SPECIAL_432>",
439
+ "<SPECIAL_433>",
440
+ "<SPECIAL_434>",
441
+ "<SPECIAL_435>",
442
+ "<SPECIAL_436>",
443
+ "<SPECIAL_437>",
444
+ "<SPECIAL_438>",
445
+ "<SPECIAL_439>",
446
+ "<SPECIAL_440>",
447
+ "<SPECIAL_441>",
448
+ "<SPECIAL_442>",
449
+ "<SPECIAL_443>",
450
+ "<SPECIAL_444>",
451
+ "<SPECIAL_445>",
452
+ "<SPECIAL_446>",
453
+ "<SPECIAL_447>",
454
+ "<SPECIAL_448>",
455
+ "<SPECIAL_449>",
456
+ "<SPECIAL_450>",
457
+ "<SPECIAL_451>",
458
+ "<SPECIAL_452>",
459
+ "<SPECIAL_453>",
460
+ "<SPECIAL_454>",
461
+ "<SPECIAL_455>",
462
+ "<SPECIAL_456>",
463
+ "<SPECIAL_457>",
464
+ "<SPECIAL_458>",
465
+ "<SPECIAL_459>",
466
+ "<SPECIAL_460>",
467
+ "<SPECIAL_461>",
468
+ "<SPECIAL_462>",
469
+ "<SPECIAL_463>",
470
+ "<SPECIAL_464>",
471
+ "<SPECIAL_465>",
472
+ "<SPECIAL_466>",
473
+ "<SPECIAL_467>",
474
+ "<SPECIAL_468>",
475
+ "<SPECIAL_469>",
476
+ "<SPECIAL_470>",
477
+ "<SPECIAL_471>",
478
+ "<SPECIAL_472>",
479
+ "<SPECIAL_473>",
480
+ "<SPECIAL_474>",
481
+ "<SPECIAL_475>",
482
+ "<SPECIAL_476>",
483
+ "<SPECIAL_477>",
484
+ "<SPECIAL_478>",
485
+ "<SPECIAL_479>",
486
+ "<SPECIAL_480>",
487
+ "<SPECIAL_481>",
488
+ "<SPECIAL_482>",
489
+ "<SPECIAL_483>",
490
+ "<SPECIAL_484>",
491
+ "<SPECIAL_485>",
492
+ "<SPECIAL_486>",
493
+ "<SPECIAL_487>",
494
+ "<SPECIAL_488>",
495
+ "<SPECIAL_489>",
496
+ "<SPECIAL_490>",
497
+ "<SPECIAL_491>",
498
+ "<SPECIAL_492>",
499
+ "<SPECIAL_493>",
500
+ "<SPECIAL_494>",
501
+ "<SPECIAL_495>",
502
+ "<SPECIAL_496>",
503
+ "<SPECIAL_497>",
504
+ "<SPECIAL_498>",
505
+ "<SPECIAL_499>",
506
+ "<SPECIAL_500>",
507
+ "<SPECIAL_501>",
508
+ "<SPECIAL_502>",
509
+ "<SPECIAL_503>",
510
+ "<SPECIAL_504>",
511
+ "<SPECIAL_505>",
512
+ "<SPECIAL_506>",
513
+ "<SPECIAL_507>",
514
+ "<SPECIAL_508>",
515
+ "<SPECIAL_509>",
516
+ "<SPECIAL_510>",
517
+ "<SPECIAL_511>",
518
+ "<SPECIAL_512>",
519
+ "<SPECIAL_513>",
520
+ "<SPECIAL_514>",
521
+ "<SPECIAL_515>",
522
+ "<SPECIAL_516>",
523
+ "<SPECIAL_517>",
524
+ "<SPECIAL_518>",
525
+ "<SPECIAL_519>",
526
+ "<SPECIAL_520>",
527
+ "<SPECIAL_521>",
528
+ "<SPECIAL_522>",
529
+ "<SPECIAL_523>",
530
+ "<SPECIAL_524>",
531
+ "<SPECIAL_525>",
532
+ "<SPECIAL_526>",
533
+ "<SPECIAL_527>",
534
+ "<SPECIAL_528>",
535
+ "<SPECIAL_529>",
536
+ "<SPECIAL_530>",
537
+ "<SPECIAL_531>",
538
+ "<SPECIAL_532>",
539
+ "<SPECIAL_533>",
540
+ "<SPECIAL_534>",
541
+ "<SPECIAL_535>",
542
+ "<SPECIAL_536>",
543
+ "<SPECIAL_537>",
544
+ "<SPECIAL_538>",
545
+ "<SPECIAL_539>",
546
+ "<SPECIAL_540>",
547
+ "<SPECIAL_541>",
548
+ "<SPECIAL_542>",
549
+ "<SPECIAL_543>",
550
+ "<SPECIAL_544>",
551
+ "<SPECIAL_545>",
552
+ "<SPECIAL_546>",
553
+ "<SPECIAL_547>",
554
+ "<SPECIAL_548>",
555
+ "<SPECIAL_549>",
556
+ "<SPECIAL_550>",
557
+ "<SPECIAL_551>",
558
+ "<SPECIAL_552>",
559
+ "<SPECIAL_553>",
560
+ "<SPECIAL_554>",
561
+ "<SPECIAL_555>",
562
+ "<SPECIAL_556>",
563
+ "<SPECIAL_557>",
564
+ "<SPECIAL_558>",
565
+ "<SPECIAL_559>",
566
+ "<SPECIAL_560>",
567
+ "<SPECIAL_561>",
568
+ "<SPECIAL_562>",
569
+ "<SPECIAL_563>",
570
+ "<SPECIAL_564>",
571
+ "<SPECIAL_565>",
572
+ "<SPECIAL_566>",
573
+ "<SPECIAL_567>",
574
+ "<SPECIAL_568>",
575
+ "<SPECIAL_569>",
576
+ "<SPECIAL_570>",
577
+ "<SPECIAL_571>",
578
+ "<SPECIAL_572>",
579
+ "<SPECIAL_573>",
580
+ "<SPECIAL_574>",
581
+ "<SPECIAL_575>",
582
+ "<SPECIAL_576>",
583
+ "<SPECIAL_577>",
584
+ "<SPECIAL_578>",
585
+ "<SPECIAL_579>",
586
+ "<SPECIAL_580>",
587
+ "<SPECIAL_581>",
588
+ "<SPECIAL_582>",
589
+ "<SPECIAL_583>",
590
+ "<SPECIAL_584>",
591
+ "<SPECIAL_585>",
592
+ "<SPECIAL_586>",
593
+ "<SPECIAL_587>",
594
+ "<SPECIAL_588>",
595
+ "<SPECIAL_589>",
596
+ "<SPECIAL_590>",
597
+ "<SPECIAL_591>",
598
+ "<SPECIAL_592>",
599
+ "<SPECIAL_593>",
600
+ "<SPECIAL_594>",
601
+ "<SPECIAL_595>",
602
+ "<SPECIAL_596>",
603
+ "<SPECIAL_597>",
604
+ "<SPECIAL_598>",
605
+ "<SPECIAL_599>",
606
+ "<SPECIAL_600>",
607
+ "<SPECIAL_601>",
608
+ "<SPECIAL_602>",
609
+ "<SPECIAL_603>",
610
+ "<SPECIAL_604>",
611
+ "<SPECIAL_605>",
612
+ "<SPECIAL_606>",
613
+ "<SPECIAL_607>",
614
+ "<SPECIAL_608>",
615
+ "<SPECIAL_609>",
616
+ "<SPECIAL_610>",
617
+ "<SPECIAL_611>",
618
+ "<SPECIAL_612>",
619
+ "<SPECIAL_613>",
620
+ "<SPECIAL_614>",
621
+ "<SPECIAL_615>",
622
+ "<SPECIAL_616>",
623
+ "<SPECIAL_617>",
624
+ "<SPECIAL_618>",
625
+ "<SPECIAL_619>",
626
+ "<SPECIAL_620>",
627
+ "<SPECIAL_621>",
628
+ "<SPECIAL_622>",
629
+ "<SPECIAL_623>",
630
+ "<SPECIAL_624>",
631
+ "<SPECIAL_625>",
632
+ "<SPECIAL_626>",
633
+ "<SPECIAL_627>",
634
+ "<SPECIAL_628>",
635
+ "<SPECIAL_629>",
636
+ "<SPECIAL_630>",
637
+ "<SPECIAL_631>",
638
+ "<SPECIAL_632>",
639
+ "<SPECIAL_633>",
640
+ "<SPECIAL_634>",
641
+ "<SPECIAL_635>",
642
+ "<SPECIAL_636>",
643
+ "<SPECIAL_637>",
644
+ "<SPECIAL_638>",
645
+ "<SPECIAL_639>",
646
+ "<SPECIAL_640>",
647
+ "<SPECIAL_641>",
648
+ "<SPECIAL_642>",
649
+ "<SPECIAL_643>",
650
+ "<SPECIAL_644>",
651
+ "<SPECIAL_645>",
652
+ "<SPECIAL_646>",
653
+ "<SPECIAL_647>",
654
+ "<SPECIAL_648>",
655
+ "<SPECIAL_649>",
656
+ "<SPECIAL_650>",
657
+ "<SPECIAL_651>",
658
+ "<SPECIAL_652>",
659
+ "<SPECIAL_653>",
660
+ "<SPECIAL_654>",
661
+ "<SPECIAL_655>",
662
+ "<SPECIAL_656>",
663
+ "<SPECIAL_657>",
664
+ "<SPECIAL_658>",
665
+ "<SPECIAL_659>",
666
+ "<SPECIAL_660>",
667
+ "<SPECIAL_661>",
668
+ "<SPECIAL_662>",
669
+ "<SPECIAL_663>",
670
+ "<SPECIAL_664>",
671
+ "<SPECIAL_665>",
672
+ "<SPECIAL_666>",
673
+ "<SPECIAL_667>",
674
+ "<SPECIAL_668>",
675
+ "<SPECIAL_669>",
676
+ "<SPECIAL_670>",
677
+ "<SPECIAL_671>",
678
+ "<SPECIAL_672>",
679
+ "<SPECIAL_673>",
680
+ "<SPECIAL_674>",
681
+ "<SPECIAL_675>",
682
+ "<SPECIAL_676>",
683
+ "<SPECIAL_677>",
684
+ "<SPECIAL_678>",
685
+ "<SPECIAL_679>",
686
+ "<SPECIAL_680>",
687
+ "<SPECIAL_681>",
688
+ "<SPECIAL_682>",
689
+ "<SPECIAL_683>",
690
+ "<SPECIAL_684>",
691
+ "<SPECIAL_685>",
692
+ "<SPECIAL_686>",
693
+ "<SPECIAL_687>",
694
+ "<SPECIAL_688>",
695
+ "<SPECIAL_689>",
696
+ "<SPECIAL_690>",
697
+ "<SPECIAL_691>",
698
+ "<SPECIAL_692>",
699
+ "<SPECIAL_693>",
700
+ "<SPECIAL_694>",
701
+ "<SPECIAL_695>",
702
+ "<SPECIAL_696>",
703
+ "<SPECIAL_697>",
704
+ "<SPECIAL_698>",
705
+ "<SPECIAL_699>",
706
+ "<SPECIAL_700>",
707
+ "<SPECIAL_701>",
708
+ "<SPECIAL_702>",
709
+ "<SPECIAL_703>",
710
+ "<SPECIAL_704>",
711
+ "<SPECIAL_705>",
712
+ "<SPECIAL_706>",
713
+ "<SPECIAL_707>",
714
+ "<SPECIAL_708>",
715
+ "<SPECIAL_709>",
716
+ "<SPECIAL_710>",
717
+ "<SPECIAL_711>",
718
+ "<SPECIAL_712>",
719
+ "<SPECIAL_713>",
720
+ "<SPECIAL_714>",
721
+ "<SPECIAL_715>",
722
+ "<SPECIAL_716>",
723
+ "<SPECIAL_717>",
724
+ "<SPECIAL_718>",
725
+ "<SPECIAL_719>",
726
+ "<SPECIAL_720>",
727
+ "<SPECIAL_721>",
728
+ "<SPECIAL_722>",
729
+ "<SPECIAL_723>",
730
+ "<SPECIAL_724>",
731
+ "<SPECIAL_725>",
732
+ "<SPECIAL_726>",
733
+ "<SPECIAL_727>",
734
+ "<SPECIAL_728>",
735
+ "<SPECIAL_729>",
736
+ "<SPECIAL_730>",
737
+ "<SPECIAL_731>",
738
+ "<SPECIAL_732>",
739
+ "<SPECIAL_733>",
740
+ "<SPECIAL_734>",
741
+ "<SPECIAL_735>",
742
+ "<SPECIAL_736>",
743
+ "<SPECIAL_737>",
744
+ "<SPECIAL_738>",
745
+ "<SPECIAL_739>",
746
+ "<SPECIAL_740>",
747
+ "<SPECIAL_741>",
748
+ "<SPECIAL_742>",
749
+ "<SPECIAL_743>",
750
+ "<SPECIAL_744>",
751
+ "<SPECIAL_745>",
752
+ "<SPECIAL_746>",
753
+ "<SPECIAL_747>",
754
+ "<SPECIAL_748>",
755
+ "<SPECIAL_749>",
756
+ "<SPECIAL_750>",
757
+ "<SPECIAL_751>",
758
+ "<SPECIAL_752>",
759
+ "<SPECIAL_753>",
760
+ "<SPECIAL_754>",
761
+ "<SPECIAL_755>",
762
+ "<SPECIAL_756>",
763
+ "<SPECIAL_757>",
764
+ "<SPECIAL_758>",
765
+ "<SPECIAL_759>",
766
+ "<SPECIAL_760>",
767
+ "<SPECIAL_761>",
768
+ "<SPECIAL_762>",
769
+ "<SPECIAL_763>",
770
+ "<SPECIAL_764>",
771
+ "<SPECIAL_765>",
772
+ "<SPECIAL_766>",
773
+ "<SPECIAL_767>",
774
+ "<SPECIAL_768>",
775
+ "<SPECIAL_769>",
776
+ "<SPECIAL_770>",
777
+ "<SPECIAL_771>",
778
+ "<SPECIAL_772>",
779
+ "<SPECIAL_773>",
780
+ "<SPECIAL_774>",
781
+ "<SPECIAL_775>",
782
+ "<SPECIAL_776>",
783
+ "<SPECIAL_777>",
784
+ "<SPECIAL_778>",
785
+ "<SPECIAL_779>",
786
+ "<SPECIAL_780>",
787
+ "<SPECIAL_781>",
788
+ "<SPECIAL_782>",
789
+ "<SPECIAL_783>",
790
+ "<SPECIAL_784>",
791
+ "<SPECIAL_785>",
792
+ "<SPECIAL_786>",
793
+ "<SPECIAL_787>",
794
+ "<SPECIAL_788>",
795
+ "<SPECIAL_789>",
796
+ "<SPECIAL_790>",
797
+ "<SPECIAL_791>",
798
+ "<SPECIAL_792>",
799
+ "<SPECIAL_793>",
800
+ "<SPECIAL_794>",
801
+ "<SPECIAL_795>",
802
+ "<SPECIAL_796>",
803
+ "<SPECIAL_797>",
804
+ "<SPECIAL_798>",
805
+ "<SPECIAL_799>",
806
+ "<SPECIAL_800>",
807
+ "<SPECIAL_801>",
808
+ "<SPECIAL_802>",
809
+ "<SPECIAL_803>",
810
+ "<SPECIAL_804>",
811
+ "<SPECIAL_805>",
812
+ "<SPECIAL_806>",
813
+ "<SPECIAL_807>",
814
+ "<SPECIAL_808>",
815
+ "<SPECIAL_809>",
816
+ "<SPECIAL_810>",
817
+ "<SPECIAL_811>",
818
+ "<SPECIAL_812>",
819
+ "<SPECIAL_813>",
820
+ "<SPECIAL_814>",
821
+ "<SPECIAL_815>",
822
+ "<SPECIAL_816>",
823
+ "<SPECIAL_817>",
824
+ "<SPECIAL_818>",
825
+ "<SPECIAL_819>",
826
+ "<SPECIAL_820>",
827
+ "<SPECIAL_821>",
828
+ "<SPECIAL_822>",
829
+ "<SPECIAL_823>",
830
+ "<SPECIAL_824>",
831
+ "<SPECIAL_825>",
832
+ "<SPECIAL_826>",
833
+ "<SPECIAL_827>",
834
+ "<SPECIAL_828>",
835
+ "<SPECIAL_829>",
836
+ "<SPECIAL_830>",
837
+ "<SPECIAL_831>",
838
+ "<SPECIAL_832>",
839
+ "<SPECIAL_833>",
840
+ "<SPECIAL_834>",
841
+ "<SPECIAL_835>",
842
+ "<SPECIAL_836>",
843
+ "<SPECIAL_837>",
844
+ "<SPECIAL_838>",
845
+ "<SPECIAL_839>",
846
+ "<SPECIAL_840>",
847
+ "<SPECIAL_841>",
848
+ "<SPECIAL_842>",
849
+ "<SPECIAL_843>",
850
+ "<SPECIAL_844>",
851
+ "<SPECIAL_845>",
852
+ "<SPECIAL_846>",
853
+ "<SPECIAL_847>",
854
+ "<SPECIAL_848>",
855
+ "<SPECIAL_849>",
856
+ "<SPECIAL_850>",
857
+ "<SPECIAL_851>",
858
+ "<SPECIAL_852>",
859
+ "<SPECIAL_853>",
860
+ "<SPECIAL_854>",
861
+ "<SPECIAL_855>",
862
+ "<SPECIAL_856>",
863
+ "<SPECIAL_857>",
864
+ "<SPECIAL_858>",
865
+ "<SPECIAL_859>",
866
+ "<SPECIAL_860>",
867
+ "<SPECIAL_861>",
868
+ "<SPECIAL_862>",
869
+ "<SPECIAL_863>",
870
+ "<SPECIAL_864>",
871
+ "<SPECIAL_865>",
872
+ "<SPECIAL_866>",
873
+ "<SPECIAL_867>",
874
+ "<SPECIAL_868>",
875
+ "<SPECIAL_869>",
876
+ "<SPECIAL_870>",
877
+ "<SPECIAL_871>",
878
+ "<SPECIAL_872>",
879
+ "<SPECIAL_873>",
880
+ "<SPECIAL_874>",
881
+ "<SPECIAL_875>",
882
+ "<SPECIAL_876>",
883
+ "<SPECIAL_877>",
884
+ "<SPECIAL_878>",
885
+ "<SPECIAL_879>",
886
+ "<SPECIAL_880>",
887
+ "<SPECIAL_881>",
888
+ "<SPECIAL_882>",
889
+ "<SPECIAL_883>",
890
+ "<SPECIAL_884>",
891
+ "<SPECIAL_885>",
892
+ "<SPECIAL_886>",
893
+ "<SPECIAL_887>",
894
+ "<SPECIAL_888>",
895
+ "<SPECIAL_889>",
896
+ "<SPECIAL_890>",
897
+ "<SPECIAL_891>",
898
+ "<SPECIAL_892>",
899
+ "<SPECIAL_893>",
900
+ "<SPECIAL_894>",
901
+ "<SPECIAL_895>",
902
+ "<SPECIAL_896>",
903
+ "<SPECIAL_897>",
904
+ "<SPECIAL_898>",
905
+ "<SPECIAL_899>",
906
+ "<SPECIAL_900>",
907
+ "<SPECIAL_901>",
908
+ "<SPECIAL_902>",
909
+ "<SPECIAL_903>",
910
+ "<SPECIAL_904>",
911
+ "<SPECIAL_905>",
912
+ "<SPECIAL_906>",
913
+ "<SPECIAL_907>",
914
+ "<SPECIAL_908>",
915
+ "<SPECIAL_909>",
916
+ "<SPECIAL_910>",
917
+ "<SPECIAL_911>",
918
+ "<SPECIAL_912>",
919
+ "<SPECIAL_913>",
920
+ "<SPECIAL_914>",
921
+ "<SPECIAL_915>",
922
+ "<SPECIAL_916>",
923
+ "<SPECIAL_917>",
924
+ "<SPECIAL_918>",
925
+ "<SPECIAL_919>",
926
+ "<SPECIAL_920>",
927
+ "<SPECIAL_921>",
928
+ "<SPECIAL_922>",
929
+ "<SPECIAL_923>",
930
+ "<SPECIAL_924>",
931
+ "<SPECIAL_925>",
932
+ "<SPECIAL_926>",
933
+ "<SPECIAL_927>",
934
+ "<SPECIAL_928>",
935
+ "<SPECIAL_929>",
936
+ "<SPECIAL_930>",
937
+ "<SPECIAL_931>",
938
+ "<SPECIAL_932>",
939
+ "<SPECIAL_933>",
940
+ "<SPECIAL_934>",
941
+ "<SPECIAL_935>",
942
+ "<SPECIAL_936>",
943
+ "<SPECIAL_937>",
944
+ "<SPECIAL_938>",
945
+ "<SPECIAL_939>",
946
+ "<SPECIAL_940>",
947
+ "<SPECIAL_941>",
948
+ "<SPECIAL_942>",
949
+ "<SPECIAL_943>",
950
+ "<SPECIAL_944>",
951
+ "<SPECIAL_945>",
952
+ "<SPECIAL_946>",
953
+ "<SPECIAL_947>",
954
+ "<SPECIAL_948>",
955
+ "<SPECIAL_949>",
956
+ "<SPECIAL_950>",
957
+ "<SPECIAL_951>",
958
+ "<SPECIAL_952>",
959
+ "<SPECIAL_953>",
960
+ "<SPECIAL_954>",
961
+ "<SPECIAL_955>",
962
+ "<SPECIAL_956>",
963
+ "<SPECIAL_957>",
964
+ "<SPECIAL_958>",
965
+ "<SPECIAL_959>",
966
+ "<SPECIAL_960>",
967
+ "<SPECIAL_961>",
968
+ "<SPECIAL_962>",
969
+ "<SPECIAL_963>",
970
+ "<SPECIAL_964>",
971
+ "<SPECIAL_965>",
972
+ "<SPECIAL_966>",
973
+ "<SPECIAL_967>",
974
+ "<SPECIAL_968>",
975
+ "<SPECIAL_969>",
976
+ "<SPECIAL_970>",
977
+ "<SPECIAL_971>",
978
+ "<SPECIAL_972>",
979
+ "<SPECIAL_973>",
980
+ "<SPECIAL_974>",
981
+ "<SPECIAL_975>",
982
+ "<SPECIAL_976>",
983
+ "<SPECIAL_977>",
984
+ "<SPECIAL_978>",
985
+ "<SPECIAL_979>",
986
+ "<SPECIAL_980>",
987
+ "<SPECIAL_981>",
988
+ "<SPECIAL_982>",
989
+ "<SPECIAL_983>",
990
+ "<SPECIAL_984>",
991
+ "<SPECIAL_985>",
992
+ "<SPECIAL_986>",
993
+ "<SPECIAL_987>",
994
+ "<SPECIAL_988>",
995
+ "<SPECIAL_989>",
996
+ "<SPECIAL_990>",
997
+ "<SPECIAL_991>",
998
+ "<SPECIAL_992>",
999
+ "<SPECIAL_993>",
1000
+ "<SPECIAL_994>",
1001
+ "<SPECIAL_995>",
1002
+ "<SPECIAL_996>",
1003
+ "<SPECIAL_997>",
1004
+ "<SPECIAL_998>",
1005
+ "<SPECIAL_999>"
1006
+ ],
1007
+ "is_local": true,
1008
+ "local_files_only": false,
1009
+ "model_max_length": 1000000000000000019884624838656,
1010
+ "pad_token": "<pad>",
1011
+ "processor_class": "PixtralProcessor",
1012
+ "tokenizer_class": "TokenizersBackend",
1013
+ "tool_parser_type": "mistral",
1014
+ "unk_token": "<unk>"
1015
+ }