Instructions to use eaddario/gemma-4-E4B-it-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use eaddario/gemma-4-E4B-it-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf eaddario/gemma-4-E4B-it-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf eaddario/gemma-4-E4B-it-GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf eaddario/gemma-4-E4B-it-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf eaddario/gemma-4-E4B-it-GGUF:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf eaddario/gemma-4-E4B-it-GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf eaddario/gemma-4-E4B-it-GGUF:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf eaddario/gemma-4-E4B-it-GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf eaddario/gemma-4-E4B-it-GGUF:F16

Use Docker

docker model run hf.co/eaddario/gemma-4-E4B-it-GGUF:F16

LM Studio
Jan
Ollama
How to use eaddario/gemma-4-E4B-it-GGUF with Ollama:
```
ollama run hf.co/eaddario/gemma-4-E4B-it-GGUF:F16
```
Unsloth Desktop

How to use eaddario/gemma-4-E4B-it-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf eaddario/gemma-4-E4B-it-GGUF:F16

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "eaddario/gemma-4-E4B-it-GGUF:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use eaddario/gemma-4-E4B-it-GGUF with Docker Model Runner:
```
docker model run hf.co/eaddario/gemma-4-E4B-it-GGUF:F16
```

Lemonade

How to use eaddario/gemma-4-E4B-it-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull eaddario/gemma-4-E4B-it-GGUF:F16

Run and chat with the model

lemonade run user.gemma-4-E4B-it-GGUF-F16

List all available models

lemonade list

Hermes Agent

How to use eaddario/gemma-4-E4B-it-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf eaddario/gemma-4-E4B-it-GGUF:F16

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default eaddario/gemma-4-E4B-it-GGUF:F16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use eaddario/gemma-4-E4B-it-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf eaddario/gemma-4-E4B-it-GGUF:F16

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "eaddario/gemma-4-E4B-it-GGUF:F16" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

gemma-4-E4B-it-GGUF / scores /gemma-4-E4B-it-Q2_K.md

eaddario

Add GGUF internal file structure

8399b2d verified 4 months ago

preview code

Raw

History Blame Contribute Delete

207 kB

gemma-4-E4B-it-Q2_K.gguf - GGUF Internal File Dump

Endian: LITTLE endian

Key Value Metadata Store

There are 52 key-value pairs in this file

POS	TYPE	Count	Key	Value
1	UINT32	1	GGUF.version	3
2	UINT64	1	GGUF.tensor_count	720
3	UINT64	1	GGUF.kv_count	49
4	STRING	1	general.architecture	`gemma4`
5	STRING	1	general.type	`model`
6	INT32	1	general.sampling.top_k	64
7	FLOAT32	1	general.sampling.top_p	0.95
8	FLOAT32	1	general.sampling.temp	1.0
9	STRING	1	general.name	`Gemma 4 E4B It`
10	STRING	1	general.size_label	`7.5B`
11	STRING	1	general.license	`apache-2.0`
12	STRING	1	general.license.link	`https://ai.google.dev/gemma/docs/gemma_4_license`
13	[STRING]	1	general.tags	[ `any-to-any` ]
14	UINT32	1	gemma4.block_count	42
15	UINT32	1	gemma4.context_length	131072
16	UINT32	1	gemma4.embedding_length	2560
17	UINT32	1	gemma4.feed_forward_length	10240
18	UINT32	1	gemma4.attention.head_count	8
19	UINT32	1	gemma4.attention.head_count_kv	2
20	FLOAT32	1	gemma4.rope.freq_base	1e+06
21	FLOAT32	1	gemma4.rope.freq_base_swa	10000.0
22	FLOAT32	1	gemma4.attention.layer_norm_rms_epsilon	1e-06
23	UINT32	1	gemma4.attention.key_length	512
24	UINT32	1	gemma4.attention.value_length	512
25	FLOAT32	1	gemma4.final_logit_softcapping	30.0
26	UINT32	1	gemma4.attention.sliding_window	512
27	UINT32	1	gemma4.attention.shared_kv_layers	18
28	UINT32	1	gemma4.embedding_length_per_layer_input	256
29	[BOOL]	42	gemma4.attention.sliding_window_pattern	[ True, True, True, True, True, False, True, ... ]
30	UINT32	1	gemma4.attention.key_length_swa	256
31	UINT32	1	gemma4.attention.value_length_swa	256
32	UINT32	1	gemma4.rope.dimension_count	512
33	UINT32	1	gemma4.rope.dimension_count_swa	256
34	STRING	1	tokenizer.ggml.model	`gemma4`
35	[STRING]	262144	tokenizer.ggml.tokens	[ `<pad>`, `<eos>`, `<bos>`, `<unk>`, `<mask>`, ... ]
36	[FLOAT32]	262144	tokenizer.ggml.scores	[ -1000.0, -1000.0, -1000.0, -1000.0, -1000.0, -1000.0, -1000.0, ... ]
37	[INT32]	262144	tokenizer.ggml.token_type	[ 3, 3, 3, 3, 3, 1, 1, ... ]
38	[STRING]	514906	tokenizer.ggml.merges	[ `...`, `▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁`...`▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁`, `...`, `...`, `▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁`...`▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁`, ... ]
39	UINT32	1	tokenizer.ggml.bos_token_id	2
40	UINT32	1	tokenizer.ggml.eos_token_id	1
41	UINT32	1	tokenizer.ggml.unknown_token_id	3
42	UINT32	1	tokenizer.ggml.padding_token_id	0
43	UINT32	1	tokenizer.ggml.mask_token_id	4
44	STRING	1	tokenizer.chat_template	`{%- macro format_parameters(pr`... `{%- endif -%} {%- endif -%}`
45	BOOL	1	tokenizer.ggml.add_space_prefix	False
46	BOOL	1	tokenizer.ggml.add_bos_token	True
47	UINT32	1	general.quantization_version	2
48	UINT32	1	general.file_type	28
49	STRING	1	quantize.imatrix.file	`gemma-4-E4B-it-WIP/imatrix/ima`...`rix-gemma-4-E4B-it-medium.gguf`
50	STRING	1	quantize.imatrix.dataset	`../datasets/imatrix/combined_eur_medium.txt`
51	UINT32	1	quantize.imatrix.entries_count	342
52	UINT32	1	quantize.imatrix.chunks_count	2471

Tensors Overview ~8B Elements

Total number of elements in all tensors: 7518069290 Elements

/Users/ed/Development/AI/hf/gemma-4-E4B-it-WIP/gemma-4-E4B-it-Q2_K.gguf - GGUF Internal File Dump
- Key Value Metadata Store
- Tensors Overview ~8B Elements

Tensor Data Offset

This table contains the offset and data segment relative to start of file

T_ID	Tensor Layer Name	Data Offset (B)	Data Size (B)
0	output_norm.weight	0xf173e0	0x2800
1	per_layer_model_proj.weight	0xf19be0	0x3480000
2	per_layer_proj_norm.weight	0x4399be0	0x400
3	per_layer_token_embd.weight	0x4399fe0	0x37200000
4	rope_freqs.weight	0x3b599fe0	0x400
5	token_embd.weight	0x3b59a3e0	0xd200000
6	blk.0.attn_k.weight	0x4879a3e0	0x69000
7	blk.0.attn_k_norm.weight	0x488033e0	0x400
8	blk.0.attn_norm.weight	0x488037e0	0x2800
9	blk.0.attn_output.weight	0x48805fe0	0x1a4000
10	blk.0.attn_q.weight	0x489a9fe0	0x1a4000
11	blk.0.attn_q_norm.weight	0x48b4dfe0	0x400
12	blk.0.attn_v.weight	0x48b4e3e0	0x52800
13	blk.0.ffn_down.weight	0x48ba0be0	0x578000
14	blk.0.ffn_gate.weight	0x49118be0	0x578000
15	blk.0.ffn_norm.weight	0x49690be0	0x2800
16	blk.0.ffn_up.weight	0x496933e0	0x672000
17	blk.0.inp_gate.weight	0x49d053e0	0x23000
18	blk.0.layer_output_scale.weight	0x49d283e0	0x4
19	blk.0.post_attention_norm.weight	0x49d28400	0x2800
20	blk.0.post_ffw_norm.weight	0x49d2ac00	0x2800
21	blk.0.post_norm.weight	0x49d2d400	0x2800
22	blk.0.proj.weight	0x49d2fc00	0x5a000
23	blk.1.attn_k.weight	0x49d89c00	0x5c800
24	blk.1.attn_k_norm.weight	0x49de6400	0x400
25	blk.1.attn_norm.weight	0x49de6800	0x2800
26	blk.1.attn_output.weight	0x49de9000	0x226000
27	blk.1.attn_q.weight	0x4a00f000	0x172000
28	blk.1.attn_q_norm.weight	0x4a181000	0x400
29	blk.1.attn_v.weight	0x4a181400	0x69000
30	blk.1.ffn_down.weight	0x4a1ea400	0x73a000
31	blk.1.ffn_gate.weight	0x4a924400	0x578000
32	blk.1.ffn_norm.weight	0x4ae9c400	0x2800
33	blk.1.ffn_up.weight	0x4ae9ec00	0x834000
34	blk.1.inp_gate.weight	0x4b6d2c00	0x34800
35	blk.1.layer_output_scale.weight	0x4b707400	0x4
36	blk.1.post_attention_norm.weight	0x4b707420	0x2800
37	blk.1.post_ffw_norm.weight	0x4b709c20	0x2800
38	blk.1.post_norm.weight	0x4b70c420	0x2800
39	blk.1.proj.weight	0x4b70ec20	0x34800
40	blk.2.attn_k.weight	0x4b743420	0x5c800
41	blk.2.attn_k_norm.weight	0x4b79fc20	0x400
42	blk.2.attn_norm.weight	0x4b7a0020	0x2800
43	blk.2.attn_output.weight	0x4b7a2820	0x226000
44	blk.2.attn_q.weight	0x4b9c8820	0x1ea000
45	blk.2.attn_q_norm.weight	0x4bbb2820	0x400
46	blk.2.attn_v.weight	0x4bbb2c20	0x69000
47	blk.2.ffn_down.weight	0x4bc1bc20	0x73a000
48	blk.2.ffn_gate.weight	0x4c355c20	0x578000
49	blk.2.ffn_norm.weight	0x4c8cdc20	0x2800
50	blk.2.ffn_up.weight	0x4c8d0420	0x672000
51	blk.2.inp_gate.weight	0x4cf42420	0x34800
52	blk.2.layer_output_scale.weight	0x4cf76c20	0x4
53	blk.2.post_attention_norm.weight	0x4cf76c40	0x2800
54	blk.2.post_ffw_norm.weight	0x4cf79440	0x2800
55	blk.2.post_norm.weight	0x4cf7bc40	0x2800
56	blk.2.proj.weight	0x4cf7e440	0x34800
57	blk.3.attn_k.weight	0x4cfb2c40	0x5c800
58	blk.3.attn_k_norm.weight	0x4d00f440	0x400
59	blk.3.attn_norm.weight	0x4d00f840	0x2800
60	blk.3.attn_output.weight	0x4d012040	0x226000
61	blk.3.attn_q.weight	0x4d238040	0x172000
62	blk.3.attn_q_norm.weight	0x4d3aa040	0x400
63	blk.3.attn_v.weight	0x4d3aa440	0x69000
64	blk.3.ffn_down.weight	0x4d413440	0x672000
65	blk.3.ffn_gate.weight	0x4da85440	0x578000
66	blk.3.ffn_norm.weight	0x4dffd440	0x2800
67	blk.3.ffn_up.weight	0x4dfffc40	0x834000
68	blk.3.inp_gate.weight	0x4e833c40	0x23000
69	blk.3.layer_output_scale.weight	0x4e856c40	0x4
70	blk.3.post_attention_norm.weight	0x4e856c60	0x2800
71	blk.3.post_ffw_norm.weight	0x4e859460	0x2800
72	blk.3.post_norm.weight	0x4e85bc60	0x2800
73	blk.3.proj.weight	0x4e85e460	0x34800
74	blk.4.attn_k.weight	0x4e892c60	0x5c800
75	blk.4.attn_k_norm.weight	0x4e8ef460	0x400
76	blk.4.attn_norm.weight	0x4e8ef860	0x2800
77	blk.4.attn_output.weight	0x4e8f2060	0x226000
78	blk.4.attn_q.weight	0x4eb18060	0x172000
79	blk.4.attn_q_norm.weight	0x4ec8a060	0x400
80	blk.4.attn_v.weight	0x4ec8a460	0x52800
81	blk.4.ffn_down.weight	0x4ecdcc60	0x672000
82	blk.4.ffn_gate.weight	0x4f34ec60	0x578000
83	blk.4.ffn_norm.weight	0x4f8c6c60	0x2800
84	blk.4.ffn_up.weight	0x4f8c9460	0x672000
85	blk.4.inp_gate.weight	0x4ff3b460	0x23000
86	blk.4.layer_output_scale.weight	0x4ff5e460	0x4
87	blk.4.post_attention_norm.weight	0x4ff5e480	0x2800
88	blk.4.post_ffw_norm.weight	0x4ff60c80	0x2800
89	blk.4.post_norm.weight	0x4ff63480	0x2800
90	blk.4.proj.weight	0x4ff65c80	0x34800
91	blk.5.attn_k.weight	0x4ff9a480	0xb9000
92	blk.5.attn_k_norm.weight	0x50053480	0x800
93	blk.5.attn_norm.weight	0x50053c80	0x2800
94	blk.5.attn_output.weight	0x50056480	0x348000
95	blk.5.attn_q.weight	0x5039e480	0x2e4000
96	blk.5.attn_q_norm.weight	0x50682480	0x800
97	blk.5.attn_v.weight	0x50682c80	0xa5000
98	blk.5.ffn_down.weight	0x50727c80	0x73a000
99	blk.5.ffn_gate.weight	0x50e61c80	0x578000
100	blk.5.ffn_norm.weight	0x513d9c80	0x2800
101	blk.5.ffn_up.weight	0x513dc480	0x834000
102	blk.5.inp_gate.weight	0x51c10480	0x34800
103	blk.5.layer_output_scale.weight	0x51c44c80	0x4
104	blk.5.post_attention_norm.weight	0x51c44ca0	0x2800
105	blk.5.post_ffw_norm.weight	0x51c474a0	0x2800
106	blk.5.post_norm.weight	0x51c49ca0	0x2800
107	blk.5.proj.weight	0x51c4c4a0	0x34800
108	blk.6.attn_k.weight	0x51c80ca0	0x69000
109	blk.6.attn_k_norm.weight	0x51ce9ca0	0x400
110	blk.6.attn_norm.weight	0x51cea0a0	0x2800
111	blk.6.attn_output.weight	0x51cec8a0	0x1ea000
112	blk.6.attn_q.weight	0x51ed68a0	0x1a4000
113	blk.6.attn_q_norm.weight	0x5207a8a0	0x400
114	blk.6.attn_v.weight	0x5207aca0	0x69000
115	blk.6.ffn_down.weight	0x520e3ca0	0x672000
116	blk.6.ffn_gate.weight	0x52755ca0	0x578000
117	blk.6.ffn_norm.weight	0x52ccdca0	0x2800
118	blk.6.ffn_up.weight	0x52cd04a0	0x834000
119	blk.6.inp_gate.weight	0x535044a0	0x34800
120	blk.6.layer_output_scale.weight	0x53538ca0	0x4
121	blk.6.post_attention_norm.weight	0x53538cc0	0x2800
122	blk.6.post_ffw_norm.weight	0x5353b4c0	0x2800
123	blk.6.post_norm.weight	0x5353dcc0	0x2800
124	blk.6.proj.weight	0x535404c0	0x34800
125	blk.7.attn_k.weight	0x53574cc0	0x5c800
126	blk.7.attn_k_norm.weight	0x535d14c0	0x400
127	blk.7.attn_norm.weight	0x535d18c0	0x2800
128	blk.7.attn_output.weight	0x535d40c0	0x226000
129	blk.7.attn_q.weight	0x537fa0c0	0x172000
130	blk.7.attn_q_norm.weight	0x5396c0c0	0x400
131	blk.7.attn_v.weight	0x5396c4c0	0x52800
132	blk.7.ffn_down.weight	0x539becc0	0x672000
133	blk.7.ffn_gate.weight	0x54030cc0	0x578000
134	blk.7.ffn_norm.weight	0x545a8cc0	0x2800
135	blk.7.ffn_up.weight	0x545ab4c0	0x672000
136	blk.7.inp_gate.weight	0x54c1d4c0	0x23000
137	blk.7.layer_output_scale.weight	0x54c404c0	0x4
138	blk.7.post_attention_norm.weight	0x54c404e0	0x2800
139	blk.7.post_ffw_norm.weight	0x54c42ce0	0x2800
140	blk.7.post_norm.weight	0x54c454e0	0x2800
141	blk.7.proj.weight	0x54c47ce0	0x34800
142	blk.8.attn_k.weight	0x54c7c4e0	0x5c800
143	blk.8.attn_k_norm.weight	0x54cd8ce0	0x400
144	blk.8.attn_norm.weight	0x54cd90e0	0x2800
145	blk.8.attn_output.weight	0x54cdb8e0	0x226000
146	blk.8.attn_q.weight	0x54f018e0	0x172000
147	blk.8.attn_q_norm.weight	0x550738e0	0x400
148	blk.8.attn_v.weight	0x55073ce0	0x52800
149	blk.8.ffn_down.weight	0x550c64e0	0x672000
150	blk.8.ffn_gate.weight	0x557384e0	0x672000
151	blk.8.ffn_norm.weight	0x55daa4e0	0x2800
152	blk.8.ffn_up.weight	0x55dacce0	0x672000
153	blk.8.inp_gate.weight	0x5641ece0	0x23000
154	blk.8.layer_output_scale.weight	0x56441ce0	0x4
155	blk.8.post_attention_norm.weight	0x56441d00	0x2800
156	blk.8.post_ffw_norm.weight	0x56444500	0x2800
157	blk.8.post_norm.weight	0x56446d00	0x2800
158	blk.8.proj.weight	0x56449500	0x34800
159	blk.9.attn_k.weight	0x5647dd00	0x5c800
160	blk.9.attn_k_norm.weight	0x564da500	0x400
161	blk.9.attn_norm.weight	0x564da900	0x2800
162	blk.9.attn_output.weight	0x564dd100	0x226000
163	blk.9.attn_q.weight	0x56703100	0x172000
164	blk.9.attn_q_norm.weight	0x56875100	0x400
165	blk.9.attn_v.weight	0x56875500	0x52800
166	blk.9.ffn_down.weight	0x568c7d00	0x672000
167	blk.9.ffn_gate.weight	0x56f39d00	0x672000
168	blk.9.ffn_norm.weight	0x575abd00	0x2800
169	blk.9.ffn_up.weight	0x575ae500	0x672000
170	blk.9.inp_gate.weight	0x57c20500	0x23000
171	blk.9.layer_output_scale.weight	0x57c43500	0x4
172	blk.9.post_attention_norm.weight	0x57c43520	0x2800
173	blk.9.post_ffw_norm.weight	0x57c45d20	0x2800
174	blk.9.post_norm.weight	0x57c48520	0x2800
175	blk.9.proj.weight	0x57c4ad20	0x34800
176	blk.10.attn_k.weight	0x57c7f520	0x5c800
177	blk.10.attn_k_norm.weight	0x57cdbd20	0x400
178	blk.10.attn_norm.weight	0x57cdc120	0x2800
179	blk.10.attn_output.weight	0x57cde920	0x226000
180	blk.10.attn_q.weight	0x57f04920	0x1ea000
181	blk.10.attn_q_norm.weight	0x580ee920	0x400
182	blk.10.attn_v.weight	0x580eed20	0x69000
183	blk.10.ffn_down.weight	0x58157d20	0x672000
184	blk.10.ffn_gate.weight	0x587c9d20	0x672000
185	blk.10.ffn_norm.weight	0x58e3bd20	0x2800
186	blk.10.ffn_up.weight	0x58e3e520	0x834000
187	blk.10.inp_gate.weight	0x59672520	0x23000
188	blk.10.layer_output_scale.weight	0x59695520	0x4
189	blk.10.post_attention_norm.weight	0x59695540	0x2800
190	blk.10.post_ffw_norm.weight	0x59697d40	0x2800
191	blk.10.post_norm.weight	0x5969a540	0x2800
192	blk.10.proj.weight	0x5969cd40	0x5a000
193	blk.11.attn_k.weight	0x596f6d40	0xb9000
194	blk.11.attn_k_norm.weight	0x597afd40	0x800
195	blk.11.attn_norm.weight	0x597b0540	0x2800
196	blk.11.attn_output.weight	0x597b2d40	0x348000
197	blk.11.attn_q.weight	0x59afad40	0x3d4000
198	blk.11.attn_q_norm.weight	0x59eced40	0x800
199	blk.11.attn_v.weight	0x59ecf540	0xd2000
200	blk.11.ffn_down.weight	0x59fa1540	0x672000
201	blk.11.ffn_gate.weight	0x5a613540	0x578000
202	blk.11.ffn_norm.weight	0x5ab8b540	0x2800
203	blk.11.ffn_up.weight	0x5ab8dd40	0x672000
204	blk.11.inp_gate.weight	0x5b1ffd40	0x23000
205	blk.11.layer_output_scale.weight	0x5b222d40	0x4
206	blk.11.post_attention_norm.weight	0x5b222d60	0x2800
207	blk.11.post_ffw_norm.weight	0x5b225560	0x2800
208	blk.11.post_norm.weight	0x5b227d60	0x2800
209	blk.11.proj.weight	0x5b22a560	0x34800
210	blk.12.attn_k.weight	0x5b25ed60	0x69000
211	blk.12.attn_k_norm.weight	0x5b2c7d60	0x400
212	blk.12.attn_norm.weight	0x5b2c8160	0x2800
213	blk.12.attn_output.weight	0x5b2ca960	0x1ea000
214	blk.12.attn_q.weight	0x5b4b4960	0x1a4000
215	blk.12.attn_q_norm.weight	0x5b658960	0x400
216	blk.12.attn_v.weight	0x5b658d60	0x69000
217	blk.12.ffn_down.weight	0x5b6c1d60	0x672000
218	blk.12.ffn_gate.weight	0x5bd33d60	0x672000
219	blk.12.ffn_norm.weight	0x5c3a5d60	0x2800
220	blk.12.ffn_up.weight	0x5c3a8560	0x834000
221	blk.12.inp_gate.weight	0x5cbdc560	0x23000
222	blk.12.layer_output_scale.weight	0x5cbff560	0x4
223	blk.12.post_attention_norm.weight	0x5cbff580	0x2800
224	blk.12.post_ffw_norm.weight	0x5cc01d80	0x2800
225	blk.12.post_norm.weight	0x5cc04580	0x2800
226	blk.12.proj.weight	0x5cc06d80	0x34800
227	blk.13.attn_k.weight	0x5cc3b580	0x7a800
228	blk.13.attn_k_norm.weight	0x5ccb5d80	0x400
229	blk.13.attn_norm.weight	0x5ccb6180	0x2800
230	blk.13.attn_output.weight	0x5ccb8980	0x226000
231	blk.13.attn_q.weight	0x5cede980	0x1ea000
232	blk.13.attn_q_norm.weight	0x5d0c8980	0x400
233	blk.13.attn_v.weight	0x5d0c8d80	0x69000
234	blk.13.ffn_down.weight	0x5d131d80	0x672000
235	blk.13.ffn_gate.weight	0x5d7a3d80	0x578000
236	blk.13.ffn_norm.weight	0x5dd1bd80	0x2800
237	blk.13.ffn_up.weight	0x5dd1e580	0x672000
238	blk.13.inp_gate.weight	0x5e390580	0x23000
239	blk.13.layer_output_scale.weight	0x5e3b3580	0x4
240	blk.13.post_attention_norm.weight	0x5e3b35a0	0x2800
241	blk.13.post_ffw_norm.weight	0x5e3b5da0	0x2800
242	blk.13.post_norm.weight	0x5e3b85a0	0x2800
243	blk.13.proj.weight	0x5e3bada0	0x6e000
244	blk.14.attn_k.weight	0x5e428da0	0x7a800
245	blk.14.attn_k_norm.weight	0x5e4a35a0	0x400
246	blk.14.attn_norm.weight	0x5e4a39a0	0x2800
247	blk.14.attn_output.weight	0x5e4a61a0	0x226000
248	blk.14.attn_q.weight	0x5e6cc1a0	0x172000
249	blk.14.attn_q_norm.weight	0x5e83e1a0	0x400
250	blk.14.attn_v.weight	0x5e83e5a0	0x69000
251	blk.14.ffn_down.weight	0x5e8a75a0	0x672000
252	blk.14.ffn_gate.weight	0x5ef195a0	0x672000
253	blk.14.ffn_norm.weight	0x5f58b5a0	0x2800
254	blk.14.ffn_up.weight	0x5f58dda0	0x672000
255	blk.14.inp_gate.weight	0x5fbffda0	0x29400
256	blk.14.layer_output_scale.weight	0x5fc291a0	0x4
257	blk.14.post_attention_norm.weight	0x5fc291c0	0x2800
258	blk.14.post_ffw_norm.weight	0x5fc2b9c0	0x2800
259	blk.14.post_norm.weight	0x5fc2e1c0	0x2800
260	blk.14.proj.weight	0x5fc309c0	0x5a000
261	blk.15.attn_k.weight	0x5fc8a9c0	0x7a800
262	blk.15.attn_k_norm.weight	0x5fd051c0	0x400
263	blk.15.attn_norm.weight	0x5fd055c0	0x2800
264	blk.15.attn_output.weight	0x5fd07dc0	0x226000
265	blk.15.attn_q.weight	0x5ff2ddc0	0x172000
266	blk.15.attn_q_norm.weight	0x6009fdc0	0x400
267	blk.15.attn_v.weight	0x600a01c0	0x69000
268	blk.15.ffn_down.weight	0x601091c0	0x672000
269	blk.15.ffn_gate.weight	0x6077b1c0	0x672000
270	blk.15.ffn_norm.weight	0x60ded1c0	0x2800
271	blk.15.ffn_up.weight	0x60def9c0	0x834000
272	blk.15.inp_gate.weight	0x616239c0	0x23000
273	blk.15.layer_output_scale.weight	0x616469c0	0x4
274	blk.15.post_attention_norm.weight	0x616469e0	0x2800
275	blk.15.post_ffw_norm.weight	0x616491e0	0x2800
276	blk.15.post_norm.weight	0x6164b9e0	0x2800
277	blk.15.proj.weight	0x6164e1e0	0x6e000
278	blk.16.attn_k.weight	0x616bc1e0	0x69000
279	blk.16.attn_k_norm.weight	0x617251e0	0x400
280	blk.16.attn_norm.weight	0x617255e0	0x2800
281	blk.16.attn_output.weight	0x61727de0	0x226000
282	blk.16.attn_q.weight	0x6194dde0	0x172000
283	blk.16.attn_q_norm.weight	0x61abfde0	0x400
284	blk.16.attn_v.weight	0x61ac01e0	0x69000
285	blk.16.ffn_down.weight	0x61b291e0	0x73a000
286	blk.16.ffn_gate.weight	0x622631e0	0x672000
287	blk.16.ffn_norm.weight	0x628d51e0	0x2800
288	blk.16.ffn_up.weight	0x628d79e0	0x834000
289	blk.16.inp_gate.weight	0x6310b9e0	0x23000
290	blk.16.layer_output_scale.weight	0x6312e9e0	0x4
291	blk.16.post_attention_norm.weight	0x6312ea00	0x2800
292	blk.16.post_ffw_norm.weight	0x63131200	0x2800
293	blk.16.post_norm.weight	0x63133a00	0x2800
294	blk.16.proj.weight	0x63136200	0x64000
295	blk.17.attn_k.weight	0x6319a200	0xb9000
296	blk.17.attn_k_norm.weight	0x63253200	0x800
297	blk.17.attn_norm.weight	0x63253a00	0x2800
298	blk.17.attn_output.weight	0x63256200	0x348000
299	blk.17.attn_q.weight	0x6359e200	0x2e4000
300	blk.17.attn_q_norm.weight	0x63882200	0x800
301	blk.17.attn_v.weight	0x63882a00	0xd2000
302	blk.17.ffn_down.weight	0x63954a00	0x672000
303	blk.17.ffn_gate.weight	0x63fc6a00	0x672000
304	blk.17.ffn_norm.weight	0x64638a00	0x2800
305	blk.17.ffn_up.weight	0x6463b200	0x834000
306	blk.17.inp_gate.weight	0x64e6f200	0x23000
307	blk.17.layer_output_scale.weight	0x64e92200	0x4
308	blk.17.post_attention_norm.weight	0x64e92220	0x2800
309	blk.17.post_ffw_norm.weight	0x64e94a20	0x2800
310	blk.17.post_norm.weight	0x64e97220	0x2800
311	blk.17.proj.weight	0x64e99a20	0x5a000
312	blk.18.attn_k.weight	0x64ef3a20	0x7a800
313	blk.18.attn_k_norm.weight	0x64f6e220	0x400
314	blk.18.attn_norm.weight	0x64f6e620	0x2800
315	blk.18.attn_output.weight	0x64f70e20	0x226000
316	blk.18.attn_q.weight	0x65196e20	0x1ea000
317	blk.18.attn_q_norm.weight	0x65380e20	0x400
318	blk.18.attn_v.weight	0x65381220	0x69000
319	blk.18.ffn_down.weight	0x653ea220	0x73a000
320	blk.18.ffn_gate.weight	0x65b24220	0x672000
321	blk.18.ffn_norm.weight	0x66196220	0x2800
322	blk.18.ffn_up.weight	0x66198a20	0x672000
323	blk.18.inp_gate.weight	0x6680aa20	0x23000
324	blk.18.layer_output_scale.weight	0x6682da20	0x4
325	blk.18.post_attention_norm.weight	0x6682da40	0x2800
326	blk.18.post_ffw_norm.weight	0x66830240	0x2800
327	blk.18.post_norm.weight	0x66832a40	0x2800
328	blk.18.proj.weight	0x66835240	0x64000
329	blk.19.attn_k.weight	0x66899240	0x7a800
330	blk.19.attn_k_norm.weight	0x66913a40	0x400
331	blk.19.attn_norm.weight	0x66913e40	0x2800
332	blk.19.attn_output.weight	0x66916640	0x226000
333	blk.19.attn_q.weight	0x66b3c640	0x1ea000
334	blk.19.attn_q_norm.weight	0x66d26640	0x400
335	blk.19.attn_v.weight	0x66d26a40	0x69000
336	blk.19.ffn_down.weight	0x66d8fa40	0x73a000
337	blk.19.ffn_gate.weight	0x674c9a40	0x672000
338	blk.19.ffn_norm.weight	0x67b3ba40	0x2800
339	blk.19.ffn_up.weight	0x67b3e240	0x672000
340	blk.19.inp_gate.weight	0x681b0240	0x23000
341	blk.19.layer_output_scale.weight	0x681d3240	0x4
342	blk.19.post_attention_norm.weight	0x681d3260	0x2800
343	blk.19.post_ffw_norm.weight	0x681d5a60	0x2800
344	blk.19.post_norm.weight	0x681d8260	0x2800
345	blk.19.proj.weight	0x681daa60	0x55000
346	blk.20.attn_k.weight	0x6822fa60	0x7a800
347	blk.20.attn_k_norm.weight	0x682aa260	0x400
348	blk.20.attn_norm.weight	0x682aa660	0x2800
349	blk.20.attn_output.weight	0x682ace60	0x226000
350	blk.20.attn_q.weight	0x684d2e60	0x1ea000
351	blk.20.attn_q_norm.weight	0x686bce60	0x400
352	blk.20.attn_v.weight	0x686bd260	0x69000
353	blk.20.ffn_down.weight	0x68726260	0x672000
354	blk.20.ffn_gate.weight	0x68d98260	0x672000
355	blk.20.ffn_norm.weight	0x6940a260	0x2800
356	blk.20.ffn_up.weight	0x6940ca60	0x834000
357	blk.20.inp_gate.weight	0x69c40a60	0x23000
358	blk.20.layer_output_scale.weight	0x69c63a60	0x4
359	blk.20.post_attention_norm.weight	0x69c63a80	0x2800
360	blk.20.post_ffw_norm.weight	0x69c66280	0x2800
361	blk.20.post_norm.weight	0x69c68a80	0x2800
362	blk.20.proj.weight	0x69c6b280	0x34800
363	blk.21.attn_k.weight	0x69c9fa80	0x7a800
364	blk.21.attn_k_norm.weight	0x69d1a280	0x400
365	blk.21.attn_norm.weight	0x69d1a680	0x2800
366	blk.21.attn_output.weight	0x69d1ce80	0x226000
367	blk.21.attn_q.weight	0x69f42e80	0x1ea000
368	blk.21.attn_q_norm.weight	0x6a12ce80	0x400
369	blk.21.attn_v.weight	0x6a12d280	0x69000
370	blk.21.ffn_down.weight	0x6a196280	0x73a000
371	blk.21.ffn_gate.weight	0x6a8d0280	0x672000
372	blk.21.ffn_norm.weight	0x6af42280	0x2800
373	blk.21.ffn_up.weight	0x6af44a80	0x834000
374	blk.21.inp_gate.weight	0x6b778a80	0x23000
375	blk.21.layer_output_scale.weight	0x6b79ba80	0x4
376	blk.21.post_attention_norm.weight	0x6b79baa0	0x2800
377	blk.21.post_ffw_norm.weight	0x6b79e2a0	0x2800
378	blk.21.post_norm.weight	0x6b7a0aa0	0x2800
379	blk.21.proj.weight	0x6b7a32a0	0x55000
380	blk.22.attn_k.weight	0x6b7f82a0	0x7a800
381	blk.22.attn_k_norm.weight	0x6b872aa0	0x400
382	blk.22.attn_norm.weight	0x6b872ea0	0x2800
383	blk.22.attn_output.weight	0x6b8756a0	0x226000
384	blk.22.attn_q.weight	0x6ba9b6a0	0x1ea000
385	blk.22.attn_q_norm.weight	0x6bc856a0	0x400
386	blk.22.attn_v.weight	0x6bc85aa0	0x69000
387	blk.22.ffn_down.weight	0x6bceeaa0	0x834000
388	blk.22.ffn_gate.weight	0x6c522aa0	0x672000
389	blk.22.ffn_norm.weight	0x6cb94aa0	0x2800
390	blk.22.ffn_up.weight	0x6cb972a0	0x834000
391	blk.22.inp_gate.weight	0x6d3cb2a0	0x23000
392	blk.22.layer_output_scale.weight	0x6d3ee2a0	0x4
393	blk.22.post_attention_norm.weight	0x6d3ee2c0	0x2800
394	blk.22.post_ffw_norm.weight	0x6d3f0ac0	0x2800
395	blk.22.post_norm.weight	0x6d3f32c0	0x2800
396	blk.22.proj.weight	0x6d3f5ac0	0x34800
397	blk.23.attn_k.weight	0x6d42a2c0	0xf5000
398	blk.23.attn_k_norm.weight	0x6d51f2c0	0x800
399	blk.23.attn_norm.weight	0x6d51fac0	0x2800
400	blk.23.attn_output.weight	0x6d5222c0	0x348000
401	blk.23.attn_q.weight	0x6d86a2c0	0x3d4000
402	blk.23.attn_q_norm.weight	0x6dc3e2c0	0x800
403	blk.23.attn_v.weight	0x6dc3eac0	0xf5000
404	blk.23.ffn_down.weight	0x6dd33ac0	0x672000
405	blk.23.ffn_gate.weight	0x6e3a5ac0	0x672000
406	blk.23.ffn_norm.weight	0x6ea17ac0	0x2800
407	blk.23.ffn_up.weight	0x6ea1a2c0	0x834000
408	blk.23.inp_gate.weight	0x6f24e2c0	0x34800
409	blk.23.layer_output_scale.weight	0x6f282ac0	0x4
410	blk.23.post_attention_norm.weight	0x6f282ae0	0x2800
411	blk.23.post_ffw_norm.weight	0x6f2852e0	0x2800
412	blk.23.post_norm.weight	0x6f287ae0	0x2800
413	blk.23.proj.weight	0x6f28a2e0	0x5a000
414	blk.24.attn_k.weight	0x6f2e42e0	0x89800
415	blk.24.attn_k_norm.weight	0x6f36dae0	0x400
416	blk.24.attn_norm.weight	0x6f36dee0	0x2800
417	blk.24.attn_output.weight	0x6f3706e0	0x1ea000
418	blk.24.attn_q.weight	0x6f55a6e0	0x1ea000
419	blk.24.attn_q_norm.weight	0x6f7446e0	0x400
420	blk.24.attn_v.weight	0x6f744ae0	0x69000
421	blk.24.ffn_down.weight	0x6f7adae0	0x672000
422	blk.24.ffn_gate.weight	0x6fe1fae0	0x672000
423	blk.24.ffn_norm.weight	0x70491ae0	0x2800
424	blk.24.ffn_up.weight	0x704942e0	0x834000
425	blk.24.inp_gate.weight	0x70cc82e0	0x34800
426	blk.24.layer_output_scale.weight	0x70cfcae0	0x4
427	blk.24.post_attention_norm.weight	0x70cfcb00	0x2800
428	blk.24.post_ffw_norm.weight	0x70cff300	0x2800
429	blk.24.post_norm.weight	0x70d01b00	0x2800
430	blk.24.proj.weight	0x70d04300	0x34800
431	blk.25.attn_k.weight	0x70d38b00	0x89800
432	blk.25.attn_k_norm.weight	0x70dc2300	0x400
433	blk.25.attn_norm.weight	0x70dc2700	0x2800
434	blk.25.attn_output.weight	0x70dc4f00	0x226000
435	blk.25.attn_q.weight	0x70feaf00	0x1a4000
436	blk.25.attn_q_norm.weight	0x7118ef00	0x400
437	blk.25.attn_v.weight	0x7118f300	0x69000
438	blk.25.ffn_down.weight	0x711f8300	0x672000
439	blk.25.ffn_gate.weight	0x7186a300	0x672000
440	blk.25.ffn_norm.weight	0x71edc300	0x2800
441	blk.25.ffn_up.weight	0x71edeb00	0x834000
442	blk.25.inp_gate.weight	0x72712b00	0x29400
443	blk.25.layer_output_scale.weight	0x7273bf00	0x4
444	blk.25.post_attention_norm.weight	0x7273bf20	0x2800
445	blk.25.post_ffw_norm.weight	0x7273e720	0x2800
446	blk.25.post_norm.weight	0x72740f20	0x2800
447	blk.25.proj.weight	0x72743720	0x5a000
448	blk.26.attn_k.weight	0x7279d720	0x89800
449	blk.26.attn_k_norm.weight	0x72826f20	0x400
450	blk.26.attn_norm.weight	0x72827320	0x2800
451	blk.26.attn_output.weight	0x72829b20	0x1ea000
452	blk.26.attn_q.weight	0x72a13b20	0x1a4000
453	blk.26.attn_q_norm.weight	0x72bb7b20	0x400
454	blk.26.attn_v.weight	0x72bb7f20	0x69000
455	blk.26.ffn_down.weight	0x72c20f20	0x672000
456	blk.26.ffn_gate.weight	0x73292f20	0x672000
457	blk.26.ffn_norm.weight	0x73904f20	0x2800
458	blk.26.ffn_up.weight	0x73907720	0x672000
459	blk.26.inp_gate.weight	0x73f79720	0x23000
460	blk.26.layer_output_scale.weight	0x73f9c720	0x4
461	blk.26.post_attention_norm.weight	0x73f9c740	0x2800
462	blk.26.post_ffw_norm.weight	0x73f9ef40	0x2800
463	blk.26.post_norm.weight	0x73fa1740	0x2800
464	blk.26.proj.weight	0x73fa3f40	0x5a000
465	blk.27.attn_k.weight	0x73ffdf40	0x89800
466	blk.27.attn_k_norm.weight	0x74087740	0x400
467	blk.27.attn_norm.weight	0x74087b40	0x2800
468	blk.27.attn_output.weight	0x7408a340	0x1ea000
469	blk.27.attn_q.weight	0x74274340	0x1a4000
470	blk.27.attn_q_norm.weight	0x74418340	0x400
471	blk.27.attn_v.weight	0x74418740	0x69000
472	blk.27.ffn_down.weight	0x74481740	0x672000
473	blk.27.ffn_gate.weight	0x74af3740	0x672000
474	blk.27.ffn_norm.weight	0x75165740	0x2800
475	blk.27.ffn_up.weight	0x75167f40	0x834000
476	blk.27.inp_gate.weight	0x7599bf40	0x23000
477	blk.27.layer_output_scale.weight	0x759bef40	0x4
478	blk.27.post_attention_norm.weight	0x759bef60	0x2800
479	blk.27.post_ffw_norm.weight	0x759c1760	0x2800
480	blk.27.post_norm.weight	0x759c3f60	0x2800
481	blk.27.proj.weight	0x759c6760	0x5a000
482	blk.28.attn_k.weight	0x75a20760	0x89800
483	blk.28.attn_k_norm.weight	0x75aa9f60	0x400
484	blk.28.attn_norm.weight	0x75aaa360	0x2800
485	blk.28.attn_output.weight	0x75aacb60	0x1ea000
486	blk.28.attn_q.weight	0x75c96b60	0x1a4000
487	blk.28.attn_q_norm.weight	0x75e3ab60	0x400
488	blk.28.attn_v.weight	0x75e3af60	0x69000
489	blk.28.ffn_down.weight	0x75ea3f60	0x672000
490	blk.28.ffn_gate.weight	0x76515f60	0x672000
491	blk.28.ffn_norm.weight	0x76b87f60	0x2800
492	blk.28.ffn_up.weight	0x76b8a760	0x672000
493	blk.28.inp_gate.weight	0x771fc760	0x23000
494	blk.28.layer_output_scale.weight	0x7721f760	0x4
495	blk.28.post_attention_norm.weight	0x7721f780	0x2800
496	blk.28.post_ffw_norm.weight	0x77221f80	0x2800
497	blk.28.post_norm.weight	0x77224780	0x2800
498	blk.28.proj.weight	0x77226f80	0x5a000
499	blk.29.attn_k.weight	0x77280f80	0x113000
500	blk.29.attn_k_norm.weight	0x77393f80	0x800
501	blk.29.attn_norm.weight	0x77394780	0x2800
502	blk.29.attn_output.weight	0x77396f80	0x348000
503	blk.29.attn_q.weight	0x776def80	0x348000
504	blk.29.attn_q_norm.weight	0x77a26f80	0x800
505	blk.29.attn_v.weight	0x77a27780	0xd2000
506	blk.29.ffn_down.weight	0x77af9780	0x672000
507	blk.29.ffn_gate.weight	0x7816b780	0x672000
508	blk.29.ffn_norm.weight	0x787dd780	0x2800
509	blk.29.ffn_up.weight	0x787dff80	0x834000
510	blk.29.inp_gate.weight	0x79013f80	0x29400
511	blk.29.layer_output_scale.weight	0x7903d380	0x4
512	blk.29.post_attention_norm.weight	0x7903d3a0	0x2800
513	blk.29.post_ffw_norm.weight	0x7903fba0	0x2800
514	blk.29.post_norm.weight	0x790423a0	0x2800
515	blk.29.proj.weight	0x79044ba0	0x5a000
516	blk.30.attn_k.weight	0x7909eba0	0x89800
517	blk.30.attn_k_norm.weight	0x791283a0	0x400
518	blk.30.attn_norm.weight	0x791287a0	0x2800
519	blk.30.attn_output.weight	0x7912afa0	0x1ea000
520	blk.30.attn_q.weight	0x79314fa0	0x1ea000
521	blk.30.attn_q_norm.weight	0x794fefa0	0x400
522	blk.30.attn_v.weight	0x794ff3a0	0x69000
523	blk.30.ffn_down.weight	0x795683a0	0x73a000
524	blk.30.ffn_gate.weight	0x79ca23a0	0x672000
525	blk.30.ffn_norm.weight	0x7a3143a0	0x2800
526	blk.30.ffn_up.weight	0x7a316ba0	0x672000
527	blk.30.inp_gate.weight	0x7a988ba0	0x29400
528	blk.30.layer_output_scale.weight	0x7a9b1fa0	0x4
529	blk.30.post_attention_norm.weight	0x7a9b1fc0	0x2800
530	blk.30.post_ffw_norm.weight	0x7a9b47c0	0x2800
531	blk.30.post_norm.weight	0x7a9b6fc0	0x2800
532	blk.30.proj.weight	0x7a9b97c0	0x5a000
533	blk.31.attn_k.weight	0x7aa137c0	0x89800
534	blk.31.attn_k_norm.weight	0x7aa9cfc0	0x400
535	blk.31.attn_norm.weight	0x7aa9d3c0	0x2800
536	blk.31.attn_output.weight	0x7aa9fbc0	0x1ea000
537	blk.31.attn_q.weight	0x7ac89bc0	0x1ea000
538	blk.31.attn_q_norm.weight	0x7ae73bc0	0x400
539	blk.31.attn_v.weight	0x7ae73fc0	0x69000
540	blk.31.ffn_down.weight	0x7aedcfc0	0x834000
541	blk.31.ffn_gate.weight	0x7b710fc0	0x672000
542	blk.31.ffn_norm.weight	0x7bd82fc0	0x2800
543	blk.31.ffn_up.weight	0x7bd857c0	0x672000
544	blk.31.inp_gate.weight	0x7c3f77c0	0x29400
545	blk.31.layer_output_scale.weight	0x7c420bc0	0x4
546	blk.31.post_attention_norm.weight	0x7c420be0	0x2800
547	blk.31.post_ffw_norm.weight	0x7c4233e0	0x2800
548	blk.31.post_norm.weight	0x7c425be0	0x2800
549	blk.31.proj.weight	0x7c4283e0	0x64000
550	blk.32.attn_k.weight	0x7c48c3e0	0x89800
551	blk.32.attn_k_norm.weight	0x7c515be0	0x400
552	blk.32.attn_norm.weight	0x7c515fe0	0x2800
553	blk.32.attn_output.weight	0x7c5187e0	0x1ea000
554	blk.32.attn_q.weight	0x7c7027e0	0x172000
555	blk.32.attn_q_norm.weight	0x7c8747e0	0x400
556	blk.32.attn_v.weight	0x7c874be0	0x69000
557	blk.32.ffn_down.weight	0x7c8ddbe0	0x834000
558	blk.32.ffn_gate.weight	0x7d111be0	0x578000
559	blk.32.ffn_norm.weight	0x7d689be0	0x2800
560	blk.32.ffn_up.weight	0x7d68c3e0	0x672000
561	blk.32.inp_gate.weight	0x7dcfe3e0	0x29400
562	blk.32.layer_output_scale.weight	0x7dd277e0	0x4
563	blk.32.post_attention_norm.weight	0x7dd27800	0x2800
564	blk.32.post_ffw_norm.weight	0x7dd2a000	0x2800
565	blk.32.post_norm.weight	0x7dd2c800	0x2800
566	blk.32.proj.weight	0x7dd2f000	0x64000
567	blk.33.attn_k.weight	0x7dd93000	0x89800
568	blk.33.attn_k_norm.weight	0x7de1c800	0x400
569	blk.33.attn_norm.weight	0x7de1cc00	0x2800
570	blk.33.attn_output.weight	0x7de1f400	0x1ea000
571	blk.33.attn_q.weight	0x7e009400	0x172000
572	blk.33.attn_q_norm.weight	0x7e17b400	0x400
573	blk.33.attn_v.weight	0x7e17b800	0x69000
574	blk.33.ffn_down.weight	0x7e1e4800	0x834000
575	blk.33.ffn_gate.weight	0x7ea18800	0x578000
576	blk.33.ffn_norm.weight	0x7ef90800	0x2800
577	blk.33.ffn_up.weight	0x7ef93000	0x672000
578	blk.33.inp_gate.weight	0x7f605000	0x29400
579	blk.33.layer_output_scale.weight	0x7f62e400	0x4
580	blk.33.post_attention_norm.weight	0x7f62e420	0x2800
581	blk.33.post_ffw_norm.weight	0x7f630c20	0x2800
582	blk.33.post_norm.weight	0x7f633420	0x2800
583	blk.33.proj.weight	0x7f635c20	0x64000
584	blk.34.attn_k.weight	0x7f699c20	0x89800
585	blk.34.attn_k_norm.weight	0x7f723420	0x400
586	blk.34.attn_norm.weight	0x7f723820	0x2800
587	blk.34.attn_output.weight	0x7f726020	0x1ea000
588	blk.34.attn_q.weight	0x7f910020	0x1a4000
589	blk.34.attn_q_norm.weight	0x7fab4020	0x400
590	blk.34.attn_v.weight	0x7fab4420	0x69000
591	blk.34.ffn_down.weight	0x7fb1d420	0x834000
592	blk.34.ffn_gate.weight	0x80351420	0x672000
593	blk.34.ffn_norm.weight	0x809c3420	0x2800
594	blk.34.ffn_up.weight	0x809c5c20	0x672000
595	blk.34.inp_gate.weight	0x81037c20	0x29400
596	blk.34.layer_output_scale.weight	0x81061020	0x4
597	blk.34.post_attention_norm.weight	0x81061040	0x2800
598	blk.34.post_ffw_norm.weight	0x81063840	0x2800
599	blk.34.post_norm.weight	0x81066040	0x2800
600	blk.34.proj.weight	0x81068840	0x34800
601	blk.35.attn_k.weight	0x8109d040	0x113000
602	blk.35.attn_k_norm.weight	0x811b0040	0x800
603	blk.35.attn_norm.weight	0x811b0840	0x2800
604	blk.35.attn_output.weight	0x811b3040	0x348000
605	blk.35.attn_q.weight	0x814fb040	0x348000
606	blk.35.attn_q_norm.weight	0x81843040	0x800
607	blk.35.attn_v.weight	0x81843840	0xd2000
608	blk.35.ffn_down.weight	0x81915840	0x834000
609	blk.35.ffn_gate.weight	0x82149840	0x672000
610	blk.35.ffn_norm.weight	0x827bb840	0x2800
611	blk.35.ffn_up.weight	0x827be040	0x834000
612	blk.35.inp_gate.weight	0x82ff2040	0x29400
613	blk.35.layer_output_scale.weight	0x8301b440	0x4
614	blk.35.post_attention_norm.weight	0x8301b460	0x2800
615	blk.35.post_ffw_norm.weight	0x8301dc60	0x2800
616	blk.35.post_norm.weight	0x83020460	0x2800
617	blk.35.proj.weight	0x83022c60	0x64000
618	blk.36.attn_k.weight	0x83086c60	0x89800
619	blk.36.attn_k_norm.weight	0x83110460	0x400
620	blk.36.attn_norm.weight	0x83110860	0x2800
621	blk.36.attn_output.weight	0x83113060	0x1a4000
622	blk.36.attn_q.weight	0x832b7060	0x172000
623	blk.36.attn_q_norm.weight	0x83429060	0x400
624	blk.36.attn_v.weight	0x83429460	0x69000
625	blk.36.ffn_down.weight	0x83492460	0x834000
626	blk.36.ffn_gate.weight	0x83cc6460	0x672000
627	blk.36.ffn_norm.weight	0x84338460	0x2800
628	blk.36.ffn_up.weight	0x8433ac60	0x672000
629	blk.36.inp_gate.weight	0x849acc60	0x29400
630	blk.36.layer_output_scale.weight	0x849d6060	0x4
631	blk.36.post_attention_norm.weight	0x849d6080	0x2800
632	blk.36.post_ffw_norm.weight	0x849d8880	0x2800
633	blk.36.post_norm.weight	0x849db080	0x2800
634	blk.36.proj.weight	0x849dd880	0x34800
635	blk.37.attn_k.weight	0x84a12080	0x89800
636	blk.37.attn_k_norm.weight	0x84a9b880	0x400
637	blk.37.attn_norm.weight	0x84a9bc80	0x2800
638	blk.37.attn_output.weight	0x84a9e480	0x1ea000
639	blk.37.attn_q.weight	0x84c88480	0x172000
640	blk.37.attn_q_norm.weight	0x84dfa480	0x400
641	blk.37.attn_v.weight	0x84dfa880	0x69000
642	blk.37.ffn_down.weight	0x84e63880	0x834000
643	blk.37.ffn_gate.weight	0x85697880	0x672000
644	blk.37.ffn_norm.weight	0x85d09880	0x2800
645	blk.37.ffn_up.weight	0x85d0c080	0x672000
646	blk.37.inp_gate.weight	0x8637e080	0x29400
647	blk.37.layer_output_scale.weight	0x863a7480	0x4
648	blk.37.post_attention_norm.weight	0x863a74a0	0x2800
649	blk.37.post_ffw_norm.weight	0x863a9ca0	0x2800
650	blk.37.post_norm.weight	0x863ac4a0	0x2800
651	blk.37.proj.weight	0x863aeca0	0x5a000
652	blk.38.attn_k.weight	0x86408ca0	0x89800
653	blk.38.attn_k_norm.weight	0x864924a0	0x400
654	blk.38.attn_norm.weight	0x864928a0	0x2800
655	blk.38.attn_output.weight	0x864950a0	0x1ea000
656	blk.38.attn_q.weight	0x8667f0a0	0x172000
657	blk.38.attn_q_norm.weight	0x867f10a0	0x400
658	blk.38.attn_v.weight	0x867f14a0	0x69000
659	blk.38.ffn_down.weight	0x8685a4a0	0x834000
660	blk.38.ffn_gate.weight	0x8708e4a0	0x672000
661	blk.38.ffn_norm.weight	0x877004a0	0x2800
662	blk.38.ffn_up.weight	0x87702ca0	0x672000
663	blk.38.inp_gate.weight	0x87d74ca0	0x29400
664	blk.38.layer_output_scale.weight	0x87d9e0a0	0x4
665	blk.38.post_attention_norm.weight	0x87d9e0c0	0x2800
666	blk.38.post_ffw_norm.weight	0x87da08c0	0x2800
667	blk.38.post_norm.weight	0x87da30c0	0x2800
668	blk.38.proj.weight	0x87da58c0	0x5a000
669	blk.39.attn_k.weight	0x87dff8c0	0x89800
670	blk.39.attn_k_norm.weight	0x87e890c0	0x400
671	blk.39.attn_norm.weight	0x87e894c0	0x2800
672	blk.39.attn_output.weight	0x87e8bcc0	0x1ea000
673	blk.39.attn_q.weight	0x88075cc0	0x172000
674	blk.39.attn_q_norm.weight	0x881e7cc0	0x400
675	blk.39.attn_v.weight	0x881e80c0	0x69000
676	blk.39.ffn_down.weight	0x882510c0	0x834000
677	blk.39.ffn_gate.weight	0x88a850c0	0x672000
678	blk.39.ffn_norm.weight	0x890f70c0	0x2800
679	blk.39.ffn_up.weight	0x890f98c0	0x672000
680	blk.39.inp_gate.weight	0x8976b8c0	0x29400
681	blk.39.layer_output_scale.weight	0x89794cc0	0x4
682	blk.39.post_attention_norm.weight	0x89794ce0	0x2800
683	blk.39.post_ffw_norm.weight	0x897974e0	0x2800
684	blk.39.post_norm.weight	0x89799ce0	0x2800
685	blk.39.proj.weight	0x8979c4e0	0x5a000
686	blk.40.attn_k.weight	0x897f64e0	0x89800
687	blk.40.attn_k_norm.weight	0x8987fce0	0x400
688	blk.40.attn_norm.weight	0x898800e0	0x2800
689	blk.40.attn_output.weight	0x898828e0	0x1ea000
690	blk.40.attn_q.weight	0x89a6c8e0	0x172000
691	blk.40.attn_q_norm.weight	0x89bde8e0	0x400
692	blk.40.attn_v.weight	0x89bdece0	0x69000
693	blk.40.ffn_down.weight	0x89c47ce0	0x672000
694	blk.40.ffn_gate.weight	0x8a2b9ce0	0x834000
695	blk.40.ffn_norm.weight	0x8aaedce0	0x2800
696	blk.40.ffn_up.weight	0x8aaf04e0	0x834000
697	blk.40.inp_gate.weight	0x8b3244e0	0x23000
698	blk.40.layer_output_scale.weight	0x8b3474e0	0x4
699	blk.40.post_attention_norm.weight	0x8b347500	0x2800
700	blk.40.post_ffw_norm.weight	0x8b349d00	0x2800
701	blk.40.post_norm.weight	0x8b34c500	0x2800
702	blk.40.proj.weight	0x8b34ed00	0x64000
703	blk.41.attn_k.weight	0x8b3b2d00	0x113000
704	blk.41.attn_k_norm.weight	0x8b4c5d00	0x800
705	blk.41.attn_norm.weight	0x8b4c6500	0x2800
706	blk.41.attn_output.weight	0x8b4c8d00	0x348000
707	blk.41.attn_q.weight	0x8b810d00	0x2e4000
708	blk.41.attn_q_norm.weight	0x8baf4d00	0x800
709	blk.41.attn_v.weight	0x8baf5500	0xd2000
710	blk.41.ffn_down.weight	0x8bbc7500	0x672000
711	blk.41.ffn_gate.weight	0x8c239500	0x672000
712	blk.41.ffn_norm.weight	0x8c8ab500	0x2800
713	blk.41.ffn_up.weight	0x8c8add00	0x672000
714	blk.41.inp_gate.weight	0x8cf1fd00	0x23000
715	blk.41.layer_output_scale.weight	0x8cf42d00	0x4
716	blk.41.post_attention_norm.weight	0x8cf42d20	0x2800
717	blk.41.post_ffw_norm.weight	0x8cf45520	0x2800
718	blk.41.post_norm.weight	0x8cf47d20	0x2800
719	blk.41.proj.weight	0x8cf4a520	0x5a000

Base Tensor Group : ~4B Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
0	output_norm.weight	Output Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
1	per_layer_model_proj.weight	Per_Layer_Model_Proj (W)	( ~28M) 27525120	2560 x 10752 x 1 x 1	F16	16.0000
2	per_layer_proj_norm.weight	Per_Layer_Proj_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
3	per_layer_token_embd.weight	Per_Layer_Token_Embd (W)	( ~3B) 2818572288	10752 x 262144 x 1 x 1	Q2_K	2.6250
4	rope_freqs.weight	Rope_Freqs (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
5	token_embd.weight	Token Embedding (W)	(~671M) 671088640	2560 x 262144 x 1 x 1	Q2_K	2.6250

Total elements in base: ( ~4B) 3517189120
Percentage of total elements: 46.78%
Bits per Weight (BPW) for base: 2.7297 bits

Block 0 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
6	blk.0.attn_k.weight	Block 0 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
7	blk.0.attn_k_norm.weight	Block 0 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
8	blk.0.attn_norm.weight	Block 0 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
9	blk.0.attn_output.weight	Block 0 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q2_K	2.6250
10	blk.0.attn_q.weight	Block 0 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
11	blk.0.attn_q_norm.weight	Block 0 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
12	blk.0.attn_v.weight	Block 0 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XXS	2.0625
13	blk.0.ffn_down.weight	Block 0 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ1_M	1.7500
14	blk.0.ffn_gate.weight	Block 0 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
15	blk.0.ffn_norm.weight	Block 0 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
16	blk.0.ffn_up.weight	Block 0 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
17	blk.0.inp_gate.weight	Block 0 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
18	blk.0.layer_output_scale.weight	Block 0 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
19	blk.0.post_attention_norm.weight	Block 0 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
20	blk.0.post_ffw_norm.weight	Block 0 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
21	blk.0.post_norm.weight	Block 0 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
22	blk.0.proj.weight	Block 0 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.0: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.0: 1.9770 bits

Block 1 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
23	blk.1.attn_k.weight	Block 1 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
24	blk.1.attn_k_norm.weight	Block 1 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
25	blk.1.attn_norm.weight	Block 1 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
26	blk.1.attn_output.weight	Block 1 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
27	blk.1.attn_q.weight	Block 1 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
28	blk.1.attn_q_norm.weight	Block 1 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
29	blk.1.attn_v.weight	Block 1 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
30	blk.1.ffn_down.weight	Block 1 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
31	blk.1.ffn_gate.weight	Block 1 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
32	blk.1.ffn_norm.weight	Block 1 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
33	blk.1.ffn_up.weight	Block 1 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
34	blk.1.inp_gate.weight	Block 1 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	Q2_K	2.6250
35	blk.1.layer_output_scale.weight	Block 1 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
36	blk.1.post_attention_norm.weight	Block 1 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
37	blk.1.post_ffw_norm.weight	Block 1 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
38	blk.1.post_norm.weight	Block 1 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
39	blk.1.proj.weight	Block 1 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.1: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.1: 2.3185 bits

Block 2 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
40	blk.2.attn_k.weight	Block 2 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
41	blk.2.attn_k_norm.weight	Block 2 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
42	blk.2.attn_norm.weight	Block 2 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
43	blk.2.attn_output.weight	Block 2 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
44	blk.2.attn_q.weight	Block 2 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
45	blk.2.attn_q_norm.weight	Block 2 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
46	blk.2.attn_v.weight	Block 2 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
47	blk.2.ffn_down.weight	Block 2 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
48	blk.2.ffn_gate.weight	Block 2 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
49	blk.2.ffn_norm.weight	Block 2 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
50	blk.2.ffn_up.weight	Block 2 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
51	blk.2.inp_gate.weight	Block 2 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	Q2_K	2.6250
52	blk.2.layer_output_scale.weight	Block 2 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
53	blk.2.post_attention_norm.weight	Block 2 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
54	blk.2.post_ffw_norm.weight	Block 2 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
55	blk.2.post_norm.weight	Block 2 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
56	blk.2.proj.weight	Block 2 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.2: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.2: 2.2023 bits

Block 3 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
57	blk.3.attn_k.weight	Block 3 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
58	blk.3.attn_k_norm.weight	Block 3 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
59	blk.3.attn_norm.weight	Block 3 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
60	blk.3.attn_output.weight	Block 3 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
61	blk.3.attn_q.weight	Block 3 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
62	blk.3.attn_q_norm.weight	Block 3 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
63	blk.3.attn_v.weight	Block 3 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
64	blk.3.ffn_down.weight	Block 3 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
65	blk.3.ffn_gate.weight	Block 3 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
66	blk.3.ffn_norm.weight	Block 3 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
67	blk.3.ffn_up.weight	Block 3 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
68	blk.3.inp_gate.weight	Block 3 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
69	blk.3.layer_output_scale.weight	Block 3 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
70	blk.3.post_attention_norm.weight	Block 3 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
71	blk.3.post_ffw_norm.weight	Block 3 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
72	blk.3.post_norm.weight	Block 3 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
73	blk.3.proj.weight	Block 3 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.3: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.3: 2.2419 bits

Block 4 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
74	blk.4.attn_k.weight	Block 4 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
75	blk.4.attn_k_norm.weight	Block 4 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
76	blk.4.attn_norm.weight	Block 4 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
77	blk.4.attn_output.weight	Block 4 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
78	blk.4.attn_q.weight	Block 4 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
79	blk.4.attn_q_norm.weight	Block 4 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
80	blk.4.attn_v.weight	Block 4 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XXS	2.0625
81	blk.4.ffn_down.weight	Block 4 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
82	blk.4.ffn_gate.weight	Block 4 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
83	blk.4.ffn_norm.weight	Block 4 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
84	blk.4.ffn_up.weight	Block 4 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
85	blk.4.inp_gate.weight	Block 4 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
86	blk.4.layer_output_scale.weight	Block 4 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
87	blk.4.post_attention_norm.weight	Block 4 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
88	blk.4.post_ffw_norm.weight	Block 4 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
89	blk.4.post_norm.weight	Block 4 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
90	blk.4.proj.weight	Block 4 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.4: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.4: 2.0756 bits

Block 5 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
91	blk.5.attn_k.weight	Block 5 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	IQ2_XS	2.3125
92	blk.5.attn_k_norm.weight	Block 5 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
93	blk.5.attn_norm.weight	Block 5 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
94	blk.5.attn_output.weight	Block 5 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
95	blk.5.attn_q.weight	Block 5 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	IQ2_XS	2.3125
96	blk.5.attn_q_norm.weight	Block 5 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
97	blk.5.attn_v.weight	Block 5 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	IQ2_XXS	2.0625
98	blk.5.ffn_down.weight	Block 5 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
99	blk.5.ffn_gate.weight	Block 5 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
100	blk.5.ffn_norm.weight	Block 5 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
101	blk.5.ffn_up.weight	Block 5 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
102	blk.5.inp_gate.weight	Block 5 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	Q2_K	2.6250
103	blk.5.layer_output_scale.weight	Block 5 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
104	blk.5.post_attention_norm.weight	Block 5 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
105	blk.5.post_ffw_norm.weight	Block 5 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
106	blk.5.post_norm.weight	Block 5 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
107	blk.5.proj.weight	Block 5 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.5: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.5: 2.2832 bits

Block 6 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
108	blk.6.attn_k.weight	Block 6 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
109	blk.6.attn_k_norm.weight	Block 6 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
110	blk.6.attn_norm.weight	Block 6 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
111	blk.6.attn_output.weight	Block 6 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
112	blk.6.attn_q.weight	Block 6 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
113	blk.6.attn_q_norm.weight	Block 6 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
114	blk.6.attn_v.weight	Block 6 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
115	blk.6.ffn_down.weight	Block 6 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
116	blk.6.ffn_gate.weight	Block 6 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
117	blk.6.ffn_norm.weight	Block 6 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
118	blk.6.ffn_up.weight	Block 6 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
119	blk.6.inp_gate.weight	Block 6 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	Q2_K	2.6250
120	blk.6.layer_output_scale.weight	Block 6 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
121	blk.6.post_attention_norm.weight	Block 6 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
122	blk.6.post_ffw_norm.weight	Block 6 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
123	blk.6.post_norm.weight	Block 6 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
124	blk.6.proj.weight	Block 6 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.6: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.6: 2.2490 bits

Block 7 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
125	blk.7.attn_k.weight	Block 7 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
126	blk.7.attn_k_norm.weight	Block 7 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
127	blk.7.attn_norm.weight	Block 7 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
128	blk.7.attn_output.weight	Block 7 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
129	blk.7.attn_q.weight	Block 7 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
130	blk.7.attn_q_norm.weight	Block 7 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
131	blk.7.attn_v.weight	Block 7 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XXS	2.0625
132	blk.7.ffn_down.weight	Block 7 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
133	blk.7.ffn_gate.weight	Block 7 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
134	blk.7.ffn_norm.weight	Block 7 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
135	blk.7.ffn_up.weight	Block 7 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
136	blk.7.inp_gate.weight	Block 7 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
137	blk.7.layer_output_scale.weight	Block 7 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
138	blk.7.post_attention_norm.weight	Block 7 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
139	blk.7.post_ffw_norm.weight	Block 7 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
140	blk.7.post_norm.weight	Block 7 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
141	blk.7.proj.weight	Block 7 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.7: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.7: 2.0756 bits

Block 8 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
142	blk.8.attn_k.weight	Block 8 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
143	blk.8.attn_k_norm.weight	Block 8 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
144	blk.8.attn_norm.weight	Block 8 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
145	blk.8.attn_output.weight	Block 8 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
146	blk.8.attn_q.weight	Block 8 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
147	blk.8.attn_q_norm.weight	Block 8 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
148	blk.8.attn_v.weight	Block 8 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XXS	2.0625
149	blk.8.ffn_down.weight	Block 8 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
150	blk.8.ffn_gate.weight	Block 8 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
151	blk.8.ffn_norm.weight	Block 8 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
152	blk.8.ffn_up.weight	Block 8 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
153	blk.8.inp_gate.weight	Block 8 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
154	blk.8.layer_output_scale.weight	Block 8 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
155	blk.8.post_attention_norm.weight	Block 8 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
156	blk.8.post_ffw_norm.weight	Block 8 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
157	blk.8.post_norm.weight	Block 8 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
158	blk.8.proj.weight	Block 8 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.8: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.8: 2.1636 bits

Block 9 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
159	blk.9.attn_k.weight	Block 9 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
160	blk.9.attn_k_norm.weight	Block 9 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
161	blk.9.attn_norm.weight	Block 9 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
162	blk.9.attn_output.weight	Block 9 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
163	blk.9.attn_q.weight	Block 9 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
164	blk.9.attn_q_norm.weight	Block 9 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
165	blk.9.attn_v.weight	Block 9 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XXS	2.0625
166	blk.9.ffn_down.weight	Block 9 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
167	blk.9.ffn_gate.weight	Block 9 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
168	blk.9.ffn_norm.weight	Block 9 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
169	blk.9.ffn_up.weight	Block 9 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
170	blk.9.inp_gate.weight	Block 9 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
171	blk.9.layer_output_scale.weight	Block 9 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
172	blk.9.post_attention_norm.weight	Block 9 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
173	blk.9.post_ffw_norm.weight	Block 9 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
174	blk.9.post_norm.weight	Block 9 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
175	blk.9.proj.weight	Block 9 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.9: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.9: 2.1636 bits

Block 10 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
176	blk.10.attn_k.weight	Block 10 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ2_XS	2.3125
177	blk.10.attn_k_norm.weight	Block 10 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
178	blk.10.attn_norm.weight	Block 10 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
179	blk.10.attn_output.weight	Block 10 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
180	blk.10.attn_q.weight	Block 10 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
181	blk.10.attn_q_norm.weight	Block 10 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
182	blk.10.attn_v.weight	Block 10 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
183	blk.10.ffn_down.weight	Block 10 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
184	blk.10.ffn_gate.weight	Block 10 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
185	blk.10.ffn_norm.weight	Block 10 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
186	blk.10.ffn_up.weight	Block 10 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
187	blk.10.inp_gate.weight	Block 10 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
188	blk.10.layer_output_scale.weight	Block 10 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
189	blk.10.post_attention_norm.weight	Block 10 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
190	blk.10.post_ffw_norm.weight	Block 10 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
191	blk.10.post_norm.weight	Block 10 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
192	blk.10.proj.weight	Block 10 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.10: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.10: 2.3854 bits

Block 11 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
193	blk.11.attn_k.weight	Block 11 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	IQ2_XS	2.3125
194	blk.11.attn_k_norm.weight	Block 11 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
195	blk.11.attn_norm.weight	Block 11 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
196	blk.11.attn_output.weight	Block 11 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
197	blk.11.attn_q.weight	Block 11 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	IQ3_XXS	3.0625
198	blk.11.attn_q_norm.weight	Block 11 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
199	blk.11.attn_v.weight	Block 11 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q2_K	2.6250
200	blk.11.ffn_down.weight	Block 11 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
201	blk.11.ffn_gate.weight	Block 11 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
202	blk.11.ffn_norm.weight	Block 11 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
203	blk.11.ffn_up.weight	Block 11 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
204	blk.11.inp_gate.weight	Block 11 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
205	blk.11.layer_output_scale.weight	Block 11 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
206	blk.11.post_attention_norm.weight	Block 11 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
207	blk.11.post_ffw_norm.weight	Block 11 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
208	blk.11.post_norm.weight	Block 11 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
209	blk.11.proj.weight	Block 11 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.11: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.11: 2.1652 bits

Block 12 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
210	blk.12.attn_k.weight	Block 12 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
211	blk.12.attn_k_norm.weight	Block 12 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
212	blk.12.attn_norm.weight	Block 12 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
213	blk.12.attn_output.weight	Block 12 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
214	blk.12.attn_q.weight	Block 12 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
215	blk.12.attn_q_norm.weight	Block 12 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
216	blk.12.attn_v.weight	Block 12 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
217	blk.12.ffn_down.weight	Block 12 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
218	blk.12.ffn_gate.weight	Block 12 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
219	blk.12.ffn_norm.weight	Block 12 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
220	blk.12.ffn_up.weight	Block 12 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
221	blk.12.inp_gate.weight	Block 12 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
222	blk.12.layer_output_scale.weight	Block 12 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
223	blk.12.post_attention_norm.weight	Block 12 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
224	blk.12.post_ffw_norm.weight	Block 12 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
225	blk.12.post_norm.weight	Block 12 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
226	blk.12.proj.weight	Block 12 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.12: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.12: 2.3308 bits

Block 13 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
227	blk.13.attn_k.weight	Block 13 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
228	blk.13.attn_k_norm.weight	Block 13 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
229	blk.13.attn_norm.weight	Block 13 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
230	blk.13.attn_output.weight	Block 13 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
231	blk.13.attn_q.weight	Block 13 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
232	blk.13.attn_q_norm.weight	Block 13 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
233	blk.13.attn_v.weight	Block 13 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
234	blk.13.ffn_down.weight	Block 13 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
235	blk.13.ffn_gate.weight	Block 13 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
236	blk.13.ffn_norm.weight	Block 13 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
237	blk.13.ffn_up.weight	Block 13 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
238	blk.13.inp_gate.weight	Block 13 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
239	blk.13.layer_output_scale.weight	Block 13 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
240	blk.13.post_attention_norm.weight	Block 13 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
241	blk.13.post_ffw_norm.weight	Block 13 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
242	blk.13.post_norm.weight	Block 13 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
243	blk.13.proj.weight	Block 13 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q5_K	5.5000

Total elements in blk.13: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.13: 2.1566 bits

Block 14 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
244	blk.14.attn_k.weight	Block 14 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
245	blk.14.attn_k_norm.weight	Block 14 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
246	blk.14.attn_norm.weight	Block 14 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
247	blk.14.attn_output.weight	Block 14 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
248	blk.14.attn_q.weight	Block 14 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
249	blk.14.attn_q_norm.weight	Block 14 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
250	blk.14.attn_v.weight	Block 14 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
251	blk.14.ffn_down.weight	Block 14 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
252	blk.14.ffn_gate.weight	Block 14 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
253	blk.14.ffn_norm.weight	Block 14 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
254	blk.14.ffn_up.weight	Block 14 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
255	blk.14.inp_gate.weight	Block 14 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
256	blk.14.layer_output_scale.weight	Block 14 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
257	blk.14.post_attention_norm.weight	Block 14 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
258	blk.14.post_ffw_norm.weight	Block 14 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
259	blk.14.post_norm.weight	Block 14 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
260	blk.14.proj.weight	Block 14 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.14: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.14: 2.1975 bits

Block 15 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
261	blk.15.attn_k.weight	Block 15 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
262	blk.15.attn_k_norm.weight	Block 15 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
263	blk.15.attn_norm.weight	Block 15 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
264	blk.15.attn_output.weight	Block 15 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
265	blk.15.attn_q.weight	Block 15 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
266	blk.15.attn_q_norm.weight	Block 15 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
267	blk.15.attn_v.weight	Block 15 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
268	blk.15.ffn_down.weight	Block 15 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
269	blk.15.ffn_gate.weight	Block 15 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
270	blk.15.ffn_norm.weight	Block 15 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
271	blk.15.ffn_up.weight	Block 15 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
272	blk.15.inp_gate.weight	Block 15 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
273	blk.15.layer_output_scale.weight	Block 15 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
274	blk.15.post_attention_norm.weight	Block 15 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
275	blk.15.post_ffw_norm.weight	Block 15 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
276	blk.15.post_norm.weight	Block 15 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
277	blk.15.proj.weight	Block 15 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q5_K	5.5000

Total elements in blk.15: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.15: 2.3608 bits

Block 16 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
278	blk.16.attn_k.weight	Block 16 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
279	blk.16.attn_k_norm.weight	Block 16 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
280	blk.16.attn_norm.weight	Block 16 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
281	blk.16.attn_output.weight	Block 16 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
282	blk.16.attn_q.weight	Block 16 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
283	blk.16.attn_q_norm.weight	Block 16 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
284	blk.16.attn_v.weight	Block 16 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
285	blk.16.ffn_down.weight	Block 16 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
286	blk.16.ffn_gate.weight	Block 16 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
287	blk.16.ffn_norm.weight	Block 16 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
288	blk.16.ffn_up.weight	Block 16 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
289	blk.16.inp_gate.weight	Block 16 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
290	blk.16.layer_output_scale.weight	Block 16 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
291	blk.16.post_attention_norm.weight	Block 16 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
292	blk.16.post_ffw_norm.weight	Block 16 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
293	blk.16.post_norm.weight	Block 16 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
294	blk.16.proj.weight	Block 16 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.16: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.16: 2.4215 bits

Block 17 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
295	blk.17.attn_k.weight	Block 17 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	IQ2_XS	2.3125
296	blk.17.attn_k_norm.weight	Block 17 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
297	blk.17.attn_norm.weight	Block 17 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
298	blk.17.attn_output.weight	Block 17 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
299	blk.17.attn_q.weight	Block 17 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	IQ2_XS	2.3125
300	blk.17.attn_q_norm.weight	Block 17 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
301	blk.17.attn_v.weight	Block 17 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q2_K	2.6250
302	blk.17.ffn_down.weight	Block 17 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
303	blk.17.ffn_gate.weight	Block 17 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
304	blk.17.ffn_norm.weight	Block 17 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
305	blk.17.ffn_up.weight	Block 17 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
306	blk.17.inp_gate.weight	Block 17 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
307	blk.17.layer_output_scale.weight	Block 17 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
308	blk.17.post_attention_norm.weight	Block 17 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
309	blk.17.post_ffw_norm.weight	Block 17 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
310	blk.17.post_norm.weight	Block 17 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
311	blk.17.proj.weight	Block 17 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.17: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.17: 2.3187 bits

Block 18 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
312	blk.18.attn_k.weight	Block 18 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
313	blk.18.attn_k_norm.weight	Block 18 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
314	blk.18.attn_norm.weight	Block 18 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
315	blk.18.attn_output.weight	Block 18 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
316	blk.18.attn_q.weight	Block 18 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
317	blk.18.attn_q_norm.weight	Block 18 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
318	blk.18.attn_v.weight	Block 18 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
319	blk.18.ffn_down.weight	Block 18 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
320	blk.18.ffn_gate.weight	Block 18 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
321	blk.18.ffn_norm.weight	Block 18 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
322	blk.18.ffn_up.weight	Block 18 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
323	blk.18.inp_gate.weight	Block 18 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
324	blk.18.layer_output_scale.weight	Block 18 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
325	blk.18.post_attention_norm.weight	Block 18 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
326	blk.18.post_ffw_norm.weight	Block 18 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
327	blk.18.post_norm.weight	Block 18 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
328	blk.18.proj.weight	Block 18 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.18: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.18: 2.3115 bits

Block 19 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
329	blk.19.attn_k.weight	Block 19 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
330	blk.19.attn_k_norm.weight	Block 19 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
331	blk.19.attn_norm.weight	Block 19 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
332	blk.19.attn_output.weight	Block 19 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
333	blk.19.attn_q.weight	Block 19 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
334	blk.19.attn_q_norm.weight	Block 19 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
335	blk.19.attn_v.weight	Block 19 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
336	blk.19.ffn_down.weight	Block 19 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
337	blk.19.ffn_gate.weight	Block 19 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
338	blk.19.ffn_norm.weight	Block 19 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
339	blk.19.ffn_up.weight	Block 19 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
340	blk.19.inp_gate.weight	Block 19 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
341	blk.19.layer_output_scale.weight	Block 19 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
342	blk.19.post_attention_norm.weight	Block 19 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
343	blk.19.post_ffw_norm.weight	Block 19 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
344	blk.19.post_norm.weight	Block 19 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
345	blk.19.proj.weight	Block 19 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	IQ4_XS	4.2500

Total elements in blk.19: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.19: 2.3062 bits

Block 20 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
346	blk.20.attn_k.weight	Block 20 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
347	blk.20.attn_k_norm.weight	Block 20 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
348	blk.20.attn_norm.weight	Block 20 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
349	blk.20.attn_output.weight	Block 20 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
350	blk.20.attn_q.weight	Block 20 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
351	blk.20.attn_q_norm.weight	Block 20 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
352	blk.20.attn_v.weight	Block 20 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
353	blk.20.ffn_down.weight	Block 20 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
354	blk.20.ffn_gate.weight	Block 20 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
355	blk.20.ffn_norm.weight	Block 20 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
356	blk.20.ffn_up.weight	Block 20 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
357	blk.20.inp_gate.weight	Block 20 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
358	blk.20.layer_output_scale.weight	Block 20 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
359	blk.20.post_attention_norm.weight	Block 20 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
360	blk.20.post_ffw_norm.weight	Block 20 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
361	blk.20.post_norm.weight	Block 20 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
362	blk.20.proj.weight	Block 20 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.20: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.20: 2.3828 bits

Block 21 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
363	blk.21.attn_k.weight	Block 21 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
364	blk.21.attn_k_norm.weight	Block 21 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
365	blk.21.attn_norm.weight	Block 21 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
366	blk.21.attn_output.weight	Block 21 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
367	blk.21.attn_q.weight	Block 21 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
368	blk.21.attn_q_norm.weight	Block 21 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
369	blk.21.attn_v.weight	Block 21 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
370	blk.21.ffn_down.weight	Block 21 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
371	blk.21.ffn_gate.weight	Block 21 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
372	blk.21.ffn_norm.weight	Block 21 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
373	blk.21.ffn_up.weight	Block 21 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
374	blk.21.inp_gate.weight	Block 21 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
375	blk.21.layer_output_scale.weight	Block 21 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
376	blk.21.post_attention_norm.weight	Block 21 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
377	blk.21.post_ffw_norm.weight	Block 21 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
378	blk.21.post_norm.weight	Block 21 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
379	blk.21.proj.weight	Block 21 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	IQ4_XS	4.2500

Total elements in blk.21: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.21: 2.4646 bits

Block 22 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
380	blk.22.attn_k.weight	Block 22 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	IQ3_XXS	3.0625
381	blk.22.attn_k_norm.weight	Block 22 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
382	blk.22.attn_norm.weight	Block 22 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
383	blk.22.attn_output.weight	Block 22 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
384	blk.22.attn_q.weight	Block 22 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
385	blk.22.attn_q_norm.weight	Block 22 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
386	blk.22.attn_v.weight	Block 22 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
387	blk.22.ffn_down.weight	Block 22 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
388	blk.22.ffn_gate.weight	Block 22 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
389	blk.22.ffn_norm.weight	Block 22 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
390	blk.22.ffn_up.weight	Block 22 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
391	blk.22.inp_gate.weight	Block 22 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
392	blk.22.layer_output_scale.weight	Block 22 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
393	blk.22.post_attention_norm.weight	Block 22 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
394	blk.22.post_ffw_norm.weight	Block 22 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
395	blk.22.post_norm.weight	Block 22 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
396	blk.22.proj.weight	Block 22 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.22: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.22: 2.5412 bits

Block 23 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
397	blk.23.attn_k.weight	Block 23 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	IQ3_XXS	3.0625
398	blk.23.attn_k_norm.weight	Block 23 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
399	blk.23.attn_norm.weight	Block 23 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
400	blk.23.attn_output.weight	Block 23 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
401	blk.23.attn_q.weight	Block 23 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	IQ3_XXS	3.0625
402	blk.23.attn_q_norm.weight	Block 23 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
403	blk.23.attn_v.weight	Block 23 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	IQ3_XXS	3.0625
404	blk.23.ffn_down.weight	Block 23 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
405	blk.23.ffn_gate.weight	Block 23 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
406	blk.23.ffn_norm.weight	Block 23 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
407	blk.23.ffn_up.weight	Block 23 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
408	blk.23.inp_gate.weight	Block 23 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	Q2_K	2.6250
409	blk.23.layer_output_scale.weight	Block 23 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
410	blk.23.post_attention_norm.weight	Block 23 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
411	blk.23.post_ffw_norm.weight	Block 23 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
412	blk.23.post_norm.weight	Block 23 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
413	blk.23.proj.weight	Block 23 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.23: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.23: 2.4275 bits

Block 24 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
414	blk.24.attn_k.weight	Block 24 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
415	blk.24.attn_k_norm.weight	Block 24 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
416	blk.24.attn_norm.weight	Block 24 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
417	blk.24.attn_output.weight	Block 24 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
418	blk.24.attn_q.weight	Block 24 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
419	blk.24.attn_q_norm.weight	Block 24 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
420	blk.24.attn_v.weight	Block 24 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
421	blk.24.ffn_down.weight	Block 24 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
422	blk.24.ffn_gate.weight	Block 24 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
423	blk.24.ffn_norm.weight	Block 24 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
424	blk.24.ffn_up.weight	Block 24 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
425	blk.24.inp_gate.weight	Block 24 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	Q2_K	2.6250
426	blk.24.layer_output_scale.weight	Block 24 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
427	blk.24.post_attention_norm.weight	Block 24 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
428	blk.24.post_ffw_norm.weight	Block 24 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
429	blk.24.post_norm.weight	Block 24 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
430	blk.24.proj.weight	Block 24 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.24: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.24: 2.3731 bits

Block 25 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
431	blk.25.attn_k.weight	Block 25 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
432	blk.25.attn_k_norm.weight	Block 25 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
433	blk.25.attn_norm.weight	Block 25 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
434	blk.25.attn_output.weight	Block 25 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q3_K	3.4375
435	blk.25.attn_q.weight	Block 25 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
436	blk.25.attn_q_norm.weight	Block 25 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
437	blk.25.attn_v.weight	Block 25 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
438	blk.25.ffn_down.weight	Block 25 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
439	blk.25.ffn_gate.weight	Block 25 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
440	blk.25.ffn_norm.weight	Block 25 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
441	blk.25.ffn_up.weight	Block 25 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
442	blk.25.inp_gate.weight	Block 25 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
443	blk.25.layer_output_scale.weight	Block 25 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
444	blk.25.post_attention_norm.weight	Block 25 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
445	blk.25.post_ffw_norm.weight	Block 25 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
446	blk.25.post_norm.weight	Block 25 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
447	blk.25.proj.weight	Block 25 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.25: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.25: 2.3788 bits

Block 26 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
448	blk.26.attn_k.weight	Block 26 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
449	blk.26.attn_k_norm.weight	Block 26 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
450	blk.26.attn_norm.weight	Block 26 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
451	blk.26.attn_output.weight	Block 26 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
452	blk.26.attn_q.weight	Block 26 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
453	blk.26.attn_q_norm.weight	Block 26 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
454	blk.26.attn_v.weight	Block 26 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
455	blk.26.ffn_down.weight	Block 26 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
456	blk.26.ffn_gate.weight	Block 26 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
457	blk.26.ffn_norm.weight	Block 26 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
458	blk.26.ffn_up.weight	Block 26 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
459	blk.26.inp_gate.weight	Block 26 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
460	blk.26.layer_output_scale.weight	Block 26 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
461	blk.26.post_attention_norm.weight	Block 26 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
462	blk.26.post_ffw_norm.weight	Block 26 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
463	blk.26.post_norm.weight	Block 26 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
464	blk.26.proj.weight	Block 26 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.26: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.26: 2.1970 bits

Block 27 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
465	blk.27.attn_k.weight	Block 27 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
466	blk.27.attn_k_norm.weight	Block 27 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
467	blk.27.attn_norm.weight	Block 27 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
468	blk.27.attn_output.weight	Block 27 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
469	blk.27.attn_q.weight	Block 27 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
470	blk.27.attn_q_norm.weight	Block 27 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
471	blk.27.attn_v.weight	Block 27 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
472	blk.27.ffn_down.weight	Block 27 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
473	blk.27.ffn_gate.weight	Block 27 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
474	blk.27.ffn_norm.weight	Block 27 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
475	blk.27.ffn_up.weight	Block 27 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
476	blk.27.inp_gate.weight	Block 27 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
477	blk.27.layer_output_scale.weight	Block 27 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
478	blk.27.post_attention_norm.weight	Block 27 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
479	blk.27.post_ffw_norm.weight	Block 27 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
480	blk.27.post_norm.weight	Block 27 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
481	blk.27.proj.weight	Block 27 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.27: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.27: 2.3555 bits

Block 28 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
482	blk.28.attn_k.weight	Block 28 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
483	blk.28.attn_k_norm.weight	Block 28 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
484	blk.28.attn_norm.weight	Block 28 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
485	blk.28.attn_output.weight	Block 28 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
486	blk.28.attn_q.weight	Block 28 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
487	blk.28.attn_q_norm.weight	Block 28 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
488	blk.28.attn_v.weight	Block 28 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
489	blk.28.ffn_down.weight	Block 28 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
490	blk.28.ffn_gate.weight	Block 28 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
491	blk.28.ffn_norm.weight	Block 28 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
492	blk.28.ffn_up.weight	Block 28 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
493	blk.28.inp_gate.weight	Block 28 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
494	blk.28.layer_output_scale.weight	Block 28 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
495	blk.28.post_attention_norm.weight	Block 28 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
496	blk.28.post_ffw_norm.weight	Block 28 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
497	blk.28.post_norm.weight	Block 28 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
498	blk.28.proj.weight	Block 28 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.28: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.28: 2.1970 bits

Block 29 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
499	blk.29.attn_k.weight	Block 29 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q3_K	3.4375
500	blk.29.attn_k_norm.weight	Block 29 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
501	blk.29.attn_norm.weight	Block 29 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
502	blk.29.attn_output.weight	Block 29 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
503	blk.29.attn_q.weight	Block 29 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	Q2_K	2.6250
504	blk.29.attn_q_norm.weight	Block 29 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
505	blk.29.attn_v.weight	Block 29 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q2_K	2.6250
506	blk.29.ffn_down.weight	Block 29 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
507	blk.29.ffn_gate.weight	Block 29 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
508	blk.29.ffn_norm.weight	Block 29 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
509	blk.29.ffn_up.weight	Block 29 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
510	blk.29.inp_gate.weight	Block 29 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
511	blk.29.layer_output_scale.weight	Block 29 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
512	blk.29.post_attention_norm.weight	Block 29 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
513	blk.29.post_ffw_norm.weight	Block 29 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
514	blk.29.post_norm.weight	Block 29 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
515	blk.29.proj.weight	Block 29 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.29: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.29: 2.3792 bits

Block 30 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
516	blk.30.attn_k.weight	Block 30 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
517	blk.30.attn_k_norm.weight	Block 30 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
518	blk.30.attn_norm.weight	Block 30 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
519	blk.30.attn_output.weight	Block 30 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
520	blk.30.attn_q.weight	Block 30 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
521	blk.30.attn_q_norm.weight	Block 30 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
522	blk.30.attn_v.weight	Block 30 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
523	blk.30.ffn_down.weight	Block 30 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XS	2.3125
524	blk.30.ffn_gate.weight	Block 30 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
525	blk.30.ffn_norm.weight	Block 30 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
526	blk.30.ffn_up.weight	Block 30 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
527	blk.30.inp_gate.weight	Block 30 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
528	blk.30.layer_output_scale.weight	Block 30 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
529	blk.30.post_attention_norm.weight	Block 30 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
530	blk.30.post_ffw_norm.weight	Block 30 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
531	blk.30.post_norm.weight	Block 30 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
532	blk.30.proj.weight	Block 30 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.30: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.30: 2.2943 bits

Block 31 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
533	blk.31.attn_k.weight	Block 31 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
534	blk.31.attn_k_norm.weight	Block 31 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
535	blk.31.attn_norm.weight	Block 31 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
536	blk.31.attn_output.weight	Block 31 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
537	blk.31.attn_q.weight	Block 31 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ3_XXS	3.0625
538	blk.31.attn_q_norm.weight	Block 31 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
539	blk.31.attn_v.weight	Block 31 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
540	blk.31.ffn_down.weight	Block 31 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
541	blk.31.ffn_gate.weight	Block 31 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
542	blk.31.ffn_norm.weight	Block 31 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
543	blk.31.ffn_up.weight	Block 31 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
544	blk.31.inp_gate.weight	Block 31 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
545	blk.31.layer_output_scale.weight	Block 31 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
546	blk.31.post_attention_norm.weight	Block 31 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
547	blk.31.post_ffw_norm.weight	Block 31 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
548	blk.31.post_norm.weight	Block 31 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
549	blk.31.proj.weight	Block 31 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.31: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.31: 2.3858 bits

Block 32 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
550	blk.32.attn_k.weight	Block 32 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
551	blk.32.attn_k_norm.weight	Block 32 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
552	blk.32.attn_norm.weight	Block 32 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
553	blk.32.attn_output.weight	Block 32 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
554	blk.32.attn_q.weight	Block 32 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
555	blk.32.attn_q_norm.weight	Block 32 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
556	blk.32.attn_v.weight	Block 32 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
557	blk.32.ffn_down.weight	Block 32 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
558	blk.32.ffn_gate.weight	Block 32 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
559	blk.32.ffn_norm.weight	Block 32 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
560	blk.32.ffn_up.weight	Block 32 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
561	blk.32.inp_gate.weight	Block 32 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
562	blk.32.layer_output_scale.weight	Block 32 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
563	blk.32.post_attention_norm.weight	Block 32 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
564	blk.32.post_ffw_norm.weight	Block 32 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
565	blk.32.post_norm.weight	Block 32 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
566	blk.32.proj.weight	Block 32 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.32: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.32: 2.2556 bits

Block 33 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
567	blk.33.attn_k.weight	Block 33 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
568	blk.33.attn_k_norm.weight	Block 33 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
569	blk.33.attn_norm.weight	Block 33 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
570	blk.33.attn_output.weight	Block 33 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
571	blk.33.attn_q.weight	Block 33 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
572	blk.33.attn_q_norm.weight	Block 33 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
573	blk.33.attn_v.weight	Block 33 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
574	blk.33.ffn_down.weight	Block 33 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
575	blk.33.ffn_gate.weight	Block 33 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ1_M	1.7500
576	blk.33.ffn_norm.weight	Block 33 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
577	blk.33.ffn_up.weight	Block 33 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
578	blk.33.inp_gate.weight	Block 33 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
579	blk.33.layer_output_scale.weight	Block 33 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
580	blk.33.post_attention_norm.weight	Block 33 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
581	blk.33.post_ffw_norm.weight	Block 33 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
582	blk.33.post_norm.weight	Block 33 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
583	blk.33.proj.weight	Block 33 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.33: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.33: 2.2556 bits

Block 34 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
584	blk.34.attn_k.weight	Block 34 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
585	blk.34.attn_k_norm.weight	Block 34 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
586	blk.34.attn_norm.weight	Block 34 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
587	blk.34.attn_output.weight	Block 34 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
588	blk.34.attn_q.weight	Block 34 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	Q2_K	2.6250
589	blk.34.attn_q_norm.weight	Block 34 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
590	blk.34.attn_v.weight	Block 34 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
591	blk.34.ffn_down.weight	Block 34 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
592	blk.34.ffn_gate.weight	Block 34 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
593	blk.34.ffn_norm.weight	Block 34 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
594	blk.34.ffn_up.weight	Block 34 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
595	blk.34.inp_gate.weight	Block 34 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
596	blk.34.layer_output_scale.weight	Block 34 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
597	blk.34.post_attention_norm.weight	Block 34 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
598	blk.34.post_ffw_norm.weight	Block 34 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
599	blk.34.post_norm.weight	Block 34 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
600	blk.34.proj.weight	Block 34 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.34: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.34: 2.3445 bits

Block 35 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
601	blk.35.attn_k.weight	Block 35 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q3_K	3.4375
602	blk.35.attn_k_norm.weight	Block 35 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
603	blk.35.attn_norm.weight	Block 35 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
604	blk.35.attn_output.weight	Block 35 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
605	blk.35.attn_q.weight	Block 35 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	Q2_K	2.6250
606	blk.35.attn_q_norm.weight	Block 35 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
607	blk.35.attn_v.weight	Block 35 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q2_K	2.6250
608	blk.35.ffn_down.weight	Block 35 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
609	blk.35.ffn_gate.weight	Block 35 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
610	blk.35.ffn_norm.weight	Block 35 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
611	blk.35.ffn_up.weight	Block 35 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
612	blk.35.inp_gate.weight	Block 35 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
613	blk.35.layer_output_scale.weight	Block 35 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
614	blk.35.post_attention_norm.weight	Block 35 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
615	blk.35.post_ffw_norm.weight	Block 35 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
616	blk.35.post_norm.weight	Block 35 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
617	blk.35.proj.weight	Block 35 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.35: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.35: 2.5212 bits

Block 36 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
618	blk.36.attn_k.weight	Block 36 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
619	blk.36.attn_k_norm.weight	Block 36 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
620	blk.36.attn_norm.weight	Block 36 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
621	blk.36.attn_output.weight	Block 36 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	Q2_K	2.6250
622	blk.36.attn_q.weight	Block 36 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
623	blk.36.attn_q_norm.weight	Block 36 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
624	blk.36.attn_v.weight	Block 36 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
625	blk.36.ffn_down.weight	Block 36 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
626	blk.36.ffn_gate.weight	Block 36 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
627	blk.36.ffn_norm.weight	Block 36 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
628	blk.36.ffn_up.weight	Block 36 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
629	blk.36.inp_gate.weight	Block 36 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
630	blk.36.layer_output_scale.weight	Block 36 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
631	blk.36.post_attention_norm.weight	Block 36 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
632	blk.36.post_ffw_norm.weight	Block 36 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
633	blk.36.post_norm.weight	Block 36 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
634	blk.36.proj.weight	Block 36 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q2_K	2.6250

Total elements in blk.36: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.36: 2.3022 bits

Block 37 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
635	blk.37.attn_k.weight	Block 37 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
636	blk.37.attn_k_norm.weight	Block 37 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
637	blk.37.attn_norm.weight	Block 37 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
638	blk.37.attn_output.weight	Block 37 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
639	blk.37.attn_q.weight	Block 37 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
640	blk.37.attn_q_norm.weight	Block 37 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
641	blk.37.attn_v.weight	Block 37 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
642	blk.37.ffn_down.weight	Block 37 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
643	blk.37.ffn_gate.weight	Block 37 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
644	blk.37.ffn_norm.weight	Block 37 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
645	blk.37.ffn_up.weight	Block 37 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
646	blk.37.inp_gate.weight	Block 37 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
647	blk.37.layer_output_scale.weight	Block 37 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
648	blk.37.post_attention_norm.weight	Block 37 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
649	blk.37.post_ffw_norm.weight	Block 37 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
650	blk.37.post_norm.weight	Block 37 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
651	blk.37.proj.weight	Block 37 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.37: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.37: 2.3401 bits

Block 38 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
652	blk.38.attn_k.weight	Block 38 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
653	blk.38.attn_k_norm.weight	Block 38 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
654	blk.38.attn_norm.weight	Block 38 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
655	blk.38.attn_output.weight	Block 38 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
656	blk.38.attn_q.weight	Block 38 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
657	blk.38.attn_q_norm.weight	Block 38 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
658	blk.38.attn_v.weight	Block 38 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
659	blk.38.ffn_down.weight	Block 38 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
660	blk.38.ffn_gate.weight	Block 38 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
661	blk.38.ffn_norm.weight	Block 38 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
662	blk.38.ffn_up.weight	Block 38 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
663	blk.38.inp_gate.weight	Block 38 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
664	blk.38.layer_output_scale.weight	Block 38 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
665	blk.38.post_attention_norm.weight	Block 38 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
666	blk.38.post_ffw_norm.weight	Block 38 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
667	blk.38.post_norm.weight	Block 38 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
668	blk.38.proj.weight	Block 38 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.38: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.38: 2.3401 bits

Block 39 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
669	blk.39.attn_k.weight	Block 39 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
670	blk.39.attn_k_norm.weight	Block 39 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
671	blk.39.attn_norm.weight	Block 39 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
672	blk.39.attn_output.weight	Block 39 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
673	blk.39.attn_q.weight	Block 39 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
674	blk.39.attn_q_norm.weight	Block 39 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
675	blk.39.attn_v.weight	Block 39 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
676	blk.39.ffn_down.weight	Block 39 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	Q2_K	2.6250
677	blk.39.ffn_gate.weight	Block 39 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
678	blk.39.ffn_norm.weight	Block 39 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
679	blk.39.ffn_up.weight	Block 39 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
680	blk.39.inp_gate.weight	Block 39 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ2_XXS	2.0625
681	blk.39.layer_output_scale.weight	Block 39 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
682	blk.39.post_attention_norm.weight	Block 39 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
683	blk.39.post_ffw_norm.weight	Block 39 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
684	blk.39.post_norm.weight	Block 39 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
685	blk.39.proj.weight	Block 39 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.39: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.39: 2.3401 bits

Block 40 Tensor Group : ~93M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
686	blk.40.attn_k.weight	Block 40 Attention Key (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q3_K	3.4375
687	blk.40.attn_k_norm.weight	Block 40 Attn_K_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
688	blk.40.attn_norm.weight	Block 40 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
689	blk.40.attn_output.weight	Block 40 Attention Output (W)	( ~5M) 5242880	2048 x 2560 x 1 x 1	IQ3_XXS	3.0625
690	blk.40.attn_q.weight	Block 40 Attention Query (W)	( ~5M) 5242880	2560 x 2048 x 1 x 1	IQ2_XS	2.3125
691	blk.40.attn_q_norm.weight	Block 40 Attn_Q_Norm (W)	( 256) 256	256 x 1 x 1 x 1	F32	32.0000
692	blk.40.attn_v.weight	Block 40 Attention Value (W)	( ~1M) 1310720	2560 x 512 x 1 x 1	Q2_K	2.6250
693	blk.40.ffn_down.weight	Block 40 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
694	blk.40.ffn_gate.weight	Block 40 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
695	blk.40.ffn_norm.weight	Block 40 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
696	blk.40.ffn_up.weight	Block 40 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	Q2_K	2.6250
697	blk.40.inp_gate.weight	Block 40 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
698	blk.40.layer_output_scale.weight	Block 40 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
699	blk.40.post_attention_norm.weight	Block 40 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
700	blk.40.post_ffw_norm.weight	Block 40 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
701	blk.40.post_norm.weight	Block 40 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
702	blk.40.proj.weight	Block 40 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_1	5.0000

Total elements in blk.40: (~93M) 93074433
Percentage of total elements: 1.24%
Bits per Weight (BPW) for blk.40: 2.4998 bits

Block 41 Tensor Group : ~106M Elements

T_ID	Tensor Layer Name	Human Friendly Tensor Layer Name	Elements	Shape	Type	BPW
703	blk.41.attn_k.weight	Block 41 Attention Key (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q3_K	3.4375
704	blk.41.attn_k_norm.weight	Block 41 Attn_K_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
705	blk.41.attn_norm.weight	Block 41 Attention Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
706	blk.41.attn_output.weight	Block 41 Attention Output (W)	( ~10M) 10485760	4096 x 2560 x 1 x 1	Q2_K	2.6250
707	blk.41.attn_q.weight	Block 41 Attention Query (W)	( ~10M) 10485760	2560 x 4096 x 1 x 1	IQ2_XS	2.3125
708	blk.41.attn_q_norm.weight	Block 41 Attn_Q_Norm (W)	( 512) 512	512 x 1 x 1 x 1	F32	32.0000
709	blk.41.attn_v.weight	Block 41 Attention Value (W)	( ~3M) 2621440	2560 x 1024 x 1 x 1	Q2_K	2.6250
710	blk.41.ffn_down.weight	Block 41 Feed-Forward Network "Down" (W)	( ~26M) 26214400	10240 x 2560 x 1 x 1	IQ2_XXS	2.0625
711	blk.41.ffn_gate.weight	Block 41 Feed-Forward Network "Gate" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
712	blk.41.ffn_norm.weight	Block 41 Feed-Forward Network Normalization (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
713	blk.41.ffn_up.weight	Block 41 Feed-Forward Network "Up" (W)	( ~26M) 26214400	2560 x 10240 x 1 x 1	IQ2_XXS	2.0625
714	blk.41.inp_gate.weight	Block 41 Inp_Gate (W)	(~655K) 655360	2560 x 256 x 1 x 1	IQ1_M	1.7500
715	blk.41.layer_output_scale.weight	Block 41 Layer_Output_Scale (W)	( 1) 1	1 x 1 x 1 x 1	F32	32.0000
716	blk.41.post_attention_norm.weight	Block 41 Post_Attention_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
717	blk.41.post_ffw_norm.weight	Block 41 Post_Ffw_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
718	blk.41.post_norm.weight	Block 41 Post_Norm (W)	( ~3K) 2560	2560 x 1 x 1 x 1	F32	32.0000
719	blk.41.proj.weight	Block 41 Proj (W)	(~655K) 655360	256 x 2560 x 1 x 1	Q4_K	4.5000

Total elements in blk.41: (~106M) 106182145
Percentage of total elements: 1.41%
Bits per Weight (BPW) for blk.41: 2.2076 bits

Total BPW for gemma-4-E4B-it-Q2_K.gguf: 2.5000 bits