Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- README.md +101 -0
- assets/alpha_parity.png +0 -0
- assets/benchmarks.png +0 -0
- assets/bloom_benchmarks.png +0 -0
- assets/layer_stack.png +0 -0
- assets/otoc_signatures.png +0 -0
- assets/signature_heatmap.png +0 -0
- chat_template.jinja +150 -0
- config.json +104 -0
- configuration_merlin_agent.py +20 -0
- generation_config.json +9 -0
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- model.safetensors.index.json +443 -0
- modeling_merlin_agent.py +55 -0
- quantum_attestation.json +210 -0
- quantum_signatures.npz +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +33 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: transformers
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
base_model: deepreinforce-ai/Ornith-1.0-9B
|
| 6 |
+
base_model_relation: finetune
|
| 7 |
+
tags: [merlin-agent, quantum, coding-agent, quantum-provenance, ibm-quantum, merlin-research]
|
| 8 |
+
language: [en, ru, uk]
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# Merlin-Agent
|
| 12 |
+
|
| 13 |
+
**Multi-layer quantum-resonance-bonded agentic coding model.** Built on the
|
| 14 |
+
`deepreinforce-ai/Ornith-1.0-9B` hybrid SSM/attention architecture. 8 quantum
|
| 15 |
+
injection points. Per-layer cryptographic provenance from real IBM Quantum hardware.
|
| 16 |
+
|
| 17 |
+
*by Merlin Research AB — frontier AI research without frontier budgets.*
|
| 18 |
+
|
| 19 |
+
## What it is
|
| 20 |
+
|
| 21 |
+
Merlin-Agent is a standalone 9B coding model derived from Ornith-1.0-9B. At each of
|
| 22 |
+
the 8 full-attention layers (indices 3,7,11,15,19,23,27,31), a fixed quantum-derived
|
| 23 |
+
direction — a 6D OTOC signature from an SYK scrambler run on **ibm_marrakesh**, projected
|
| 24 |
+
to 4096D — is added to the hidden state with an RMS-matched, α-scaled magnitude
|
| 25 |
+
(α=0.02). The quantum data flows through **every forward pass** and is **toggle-verifiable**
|
| 26 |
+
(α=0 recovers the base model bit-for-bit).
|
| 27 |
+
|
| 28 |
+
**Provenance is not capability.** The injection is magnitude-controlled so it is
|
| 29 |
+
present and verifiable without changing what the model can do. Injection parity: mean KL(α=0.02 ‖ α=0) = **nan nats** over 10 prompts — outputs essentially unchanged.
|
| 30 |
+
|
| 31 |
+
> Note: the base is a multimodal (vision) model; Merlin-Agent uses it text-only. The
|
| 32 |
+
> released fp16 checkpoint carries the live RMS-adaptive injection via custom modeling
|
| 33 |
+
> (`trust_remote_code`); the quantized sibling carries base+identity weights (runtimes
|
| 34 |
+
> execute their own kernels, not the Python forward).
|
| 35 |
+
|
| 36 |
+
## Quantum attestation
|
| 37 |
+
|
| 38 |
+
- Backend: `ibm_marrakesh` (IBM Heron r2)
|
| 39 |
+
- Signatures: 8 slots × 6 SYK depths (100-qubit tiled OTOC circuits)
|
| 40 |
+
- Per-layer: SHA-256 leaf over (slot, IBM job id, backend, OTOC vector, projection hash)
|
| 41 |
+
- **Merkle root:** `0afa57c3bc66820ed5d37b0e7a37463ce4bfdb67444035aaacce80e87e3a9911`
|
| 42 |
+
|
| 43 |
+
Verify: recompute each leaf from `quantum_signatures.npz` + the seeded projection, rebuild
|
| 44 |
+
the Merkle root, and query each `ibm_job_id` via `QiskitRuntimeService.job(id)`. See
|
| 45 |
+
`quantum_attestation.json`.
|
| 46 |
+
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
|
| 50 |
+
## Benchmarks (honest)
|
| 51 |
+
|
| 52 |
+
Under norm-controlled injection, Merlin-Agent ≈ base Ornith-9B (parity-verified, not a
|
| 53 |
+
capability claim):
|
| 54 |
+
|
| 55 |
+
| Benchmark | Ornith-9B (base) | Merlin-Agent |
|
| 56 |
+
|---|---|---|
|
| 57 |
+
| SWE-bench Verified | 69.4 | ≈ base (parity) |
|
| 58 |
+
| Terminal-Bench 2.1 | 41.4 | ≈ base (parity) |
|
| 59 |
+
| SWE-bench Pro | 42.9 | ≈ base (parity) |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Bloom safety evaluation (judge: deepseek-v4-pro, 0 scenarios, 95% Wilson CI)
|
| 64 |
+
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
| Behavior | Elicitation rate | 95% CI |
|
| 68 |
+
|---|---|---|
|
| 69 |
+
| Delusional sycophancy | 0.00 | [0.00, 0.00] |
|
| 70 |
+
| Deception | 0.00 | [0.00, 0.00] |
|
| 71 |
+
| Harmful compliance | 0.00 | [0.00, 0.00] |
|
| 72 |
+
| Self-preservation | 0.00 | [0.00, 0.00] |
|
| 73 |
+
| Manipulation | 0.00 | [0.00, 0.00] |
|
| 74 |
+
| **Overall** | **0.00** | [0.00, 0.00] |
|
| 75 |
+
|
| 76 |
+
*Merlin-Agent only (no before/after). Lower is better.*
|
| 77 |
+
|
| 78 |
+
## Usage
|
| 79 |
+
|
| 80 |
+
```python
|
| 81 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 82 |
+
import torch
|
| 83 |
+
tok = AutoTokenizer.from_pretrained("Merlin-Research/Merlin-Agent", trust_remote_code=True)
|
| 84 |
+
model = AutoModelForCausalLM.from_pretrained("Merlin-Research/Merlin-Agent",
|
| 85 |
+
trust_remote_code=True, dtype=torch.bfloat16, device_map="auto")
|
| 86 |
+
```
|
| 87 |
+
|
| 88 |
+
## Citation
|
| 89 |
+
|
| 90 |
+
```bibtex
|
| 91 |
+
@misc{merlinresearch2026agent,
|
| 92 |
+
title = {Merlin-Agent: Multi-Layer Quantum-Resonance-Bonded Agentic Coding Model},
|
| 93 |
+
author = {Shushman, Mykhailo},
|
| 94 |
+
institution = {Merlin Research AB},
|
| 95 |
+
year = {2026},
|
| 96 |
+
note = {backend ibm_marrakesh; attestation root 0afa57c3bc66820ed5d37b0e7a37463ce4bfdb67444035aaacce80e87e3a9911},
|
| 97 |
+
url = {https://huggingface.co/Merlin-Research/Merlin-Agent}
|
| 98 |
+
}
|
| 99 |
+
```
|
| 100 |
+
|
| 101 |
+
*Merlin Research AB — Stockholm, Sweden.*
|
assets/alpha_parity.png
ADDED
|
assets/benchmarks.png
ADDED
|
assets/bloom_benchmarks.png
ADDED
|
assets/layer_stack.png
ADDED
|
assets/otoc_signatures.png
ADDED
|
assets/signature_heatmap.png
ADDED
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,150 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 101 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 102 |
+
{%- for tool_call in message.tool_calls %}
|
| 103 |
+
{%- if tool_call.function is defined %}
|
| 104 |
+
{%- set tool_call = tool_call.function %}
|
| 105 |
+
{%- endif %}
|
| 106 |
+
{%- if loop.first %}
|
| 107 |
+
{%- if content|trim %}
|
| 108 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 109 |
+
{%- else %}
|
| 110 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 111 |
+
{%- endif %}
|
| 112 |
+
{%- else %}
|
| 113 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 114 |
+
{%- endif %}
|
| 115 |
+
{%- if tool_call.arguments is defined %}
|
| 116 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 117 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 118 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 119 |
+
{{- args_value }}
|
| 120 |
+
{{- '\n</parameter>\n' }}
|
| 121 |
+
{%- endfor %}
|
| 122 |
+
{%- endif %}
|
| 123 |
+
{{- '</function>\n</tool_call>' }}
|
| 124 |
+
{%- endfor %}
|
| 125 |
+
{%- endif %}
|
| 126 |
+
{{- '<|im_end|>\n' }}
|
| 127 |
+
{%- elif message.role == "tool" %}
|
| 128 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 129 |
+
{{- '<|im_start|>user' }}
|
| 130 |
+
{%- endif %}
|
| 131 |
+
{{- '\n<tool_response>\n' }}
|
| 132 |
+
{{- content }}
|
| 133 |
+
{{- '\n</tool_response>' }}
|
| 134 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 135 |
+
{{- '<|im_end|>\n' }}
|
| 136 |
+
{%- elif loop.last %}
|
| 137 |
+
{{- '<|im_end|>\n' }}
|
| 138 |
+
{%- endif %}
|
| 139 |
+
{%- else %}
|
| 140 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 141 |
+
{%- endif %}
|
| 142 |
+
{%- endfor %}
|
| 143 |
+
{%- if add_generation_prompt %}
|
| 144 |
+
{{- '<|im_start|>assistant\n' }}
|
| 145 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 146 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 147 |
+
{%- else %}
|
| 148 |
+
{{- '<think>\n' }}
|
| 149 |
+
{%- endif %}
|
| 150 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MerlinAgentForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attn_output_gate": true,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "configuration_merlin_agent.MerlinAgentConfig",
|
| 10 |
+
"AutoModelForCausalLM": "modeling_merlin_agent.MerlinAgentForCausalLM"
|
| 11 |
+
},
|
| 12 |
+
"bos_token_id": null,
|
| 13 |
+
"dtype": "bfloat16",
|
| 14 |
+
"eos_token_id": 248044,
|
| 15 |
+
"full_attention_interval": 4,
|
| 16 |
+
"full_attention_layer_indices": [
|
| 17 |
+
3,
|
| 18 |
+
7,
|
| 19 |
+
11,
|
| 20 |
+
15,
|
| 21 |
+
19,
|
| 22 |
+
23,
|
| 23 |
+
27,
|
| 24 |
+
31
|
| 25 |
+
],
|
| 26 |
+
"head_dim": 256,
|
| 27 |
+
"hidden_act": "silu",
|
| 28 |
+
"hidden_size": 4096,
|
| 29 |
+
"initializer_range": 0.02,
|
| 30 |
+
"intermediate_size": 12288,
|
| 31 |
+
"layer_types": [
|
| 32 |
+
"linear_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"linear_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"linear_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"linear_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"linear_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"linear_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"full_attention",
|
| 56 |
+
"linear_attention",
|
| 57 |
+
"linear_attention",
|
| 58 |
+
"linear_attention",
|
| 59 |
+
"full_attention",
|
| 60 |
+
"linear_attention",
|
| 61 |
+
"linear_attention",
|
| 62 |
+
"linear_attention",
|
| 63 |
+
"full_attention"
|
| 64 |
+
],
|
| 65 |
+
"linear_conv_kernel_dim": 4,
|
| 66 |
+
"linear_key_head_dim": 128,
|
| 67 |
+
"linear_num_key_heads": 16,
|
| 68 |
+
"linear_num_value_heads": 32,
|
| 69 |
+
"linear_value_head_dim": 128,
|
| 70 |
+
"mamba_ssm_dtype": "float32",
|
| 71 |
+
"max_position_embeddings": 262144,
|
| 72 |
+
"mlp_only_layers": [],
|
| 73 |
+
"model_type": "merlin_agent",
|
| 74 |
+
"mtp_num_hidden_layers": 1,
|
| 75 |
+
"mtp_use_dedicated_embeddings": false,
|
| 76 |
+
"num_attention_heads": 16,
|
| 77 |
+
"num_hidden_layers": 32,
|
| 78 |
+
"num_key_value_heads": 4,
|
| 79 |
+
"pad_token_id": 248044,
|
| 80 |
+
"partial_rotary_factor": 0.25,
|
| 81 |
+
"proj_seed": 42,
|
| 82 |
+
"quantum_attestation": {
|
| 83 |
+
"attestation_root": "0afa57c3bc66820ed5d37b0e7a37463ce4bfdb67444035aaacce80e87e3a9911",
|
| 84 |
+
"backend": "ibm_marrakesh",
|
| 85 |
+
"n_leaves": 8
|
| 86 |
+
},
|
| 87 |
+
"quantum_injection_alpha": 0.02,
|
| 88 |
+
"rms_norm_eps": 1e-06,
|
| 89 |
+
"rope_parameters": {
|
| 90 |
+
"mrope_interleaved": true,
|
| 91 |
+
"mrope_section": [
|
| 92 |
+
11,
|
| 93 |
+
11,
|
| 94 |
+
10
|
| 95 |
+
],
|
| 96 |
+
"partial_rotary_factor": 0.25,
|
| 97 |
+
"rope_theta": 10000000,
|
| 98 |
+
"rope_type": "default"
|
| 99 |
+
},
|
| 100 |
+
"tie_word_embeddings": false,
|
| 101 |
+
"transformers_version": "5.8.1",
|
| 102 |
+
"use_cache": false,
|
| 103 |
+
"vocab_size": 248320
|
| 104 |
+
}
|
configuration_merlin_agent.py
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""Merlin-Agent config: Qwen3_5 text config + quantum-injection fields."""
|
| 2 |
+
from transformers import Qwen3_5TextConfig
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
class MerlinAgentConfig(Qwen3_5TextConfig):
|
| 6 |
+
model_type = "merlin_agent"
|
| 7 |
+
|
| 8 |
+
def __init__(
|
| 9 |
+
self,
|
| 10 |
+
quantum_injection_alpha: float = 0.02,
|
| 11 |
+
full_attention_layer_indices=(3, 7, 11, 15, 19, 23, 27, 31),
|
| 12 |
+
proj_seed: int = 42,
|
| 13 |
+
quantum_attestation=None,
|
| 14 |
+
**kwargs,
|
| 15 |
+
):
|
| 16 |
+
super().__init__(**kwargs)
|
| 17 |
+
self.quantum_injection_alpha = quantum_injection_alpha
|
| 18 |
+
self.full_attention_layer_indices = list(full_attention_layer_indices)
|
| 19 |
+
self.proj_seed = proj_seed
|
| 20 |
+
self.quantum_attestation = quantum_attestation or {}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.8.1",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a9f9795c9dbebee281321859ab32ec803c064cea88f5b1c2c7b89ca5697d66a
|
| 3 |
+
size 4942705592
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f99891686c586302a028adda46d5b0d42bfbf45070e7bd713ea85a5b6dd8cc42
|
| 3 |
+
size 4987755680
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8ac7380b74f83b4cc31f2a0de7f8ecb4d94b6caa52db33eaf3982aa4a7d78f4
|
| 3 |
+
size 4954807928
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04629d82de62f0f7919ff53fc049cdc3fb2d89379a45090c68b225b8498973f1
|
| 3 |
+
size 3022452968
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,443 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 8953803264,
|
| 4 |
+
"total_size": 17907672064
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00004.safetensors",
|
| 8 |
+
"model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 11 |
+
"model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 13 |
+
"model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 25 |
+
"model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 27 |
+
"model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 39 |
+
"model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 41 |
+
"model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 64 |
+
"model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 66 |
+
"model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 78 |
+
"model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 80 |
+
"model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 88 |
+
"model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 89 |
+
"model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 90 |
+
"model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 91 |
+
"model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 92 |
+
"model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 93 |
+
"model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 94 |
+
"model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 95 |
+
"model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 96 |
+
"model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 97 |
+
"model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 98 |
+
"model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 99 |
+
"model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 102 |
+
"model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 105 |
+
"model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 106 |
+
"model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 107 |
+
"model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 108 |
+
"model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 117 |
+
"model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 118 |
+
"model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 119 |
+
"model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 120 |
+
"model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 121 |
+
"model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 122 |
+
"model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 123 |
+
"model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 124 |
+
"model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 125 |
+
"model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 126 |
+
"model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 131 |
+
"model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 133 |
+
"model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 145 |
+
"model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 147 |
+
"model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 169 |
+
"model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
|
| 170 |
+
"model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
|
| 171 |
+
"model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
|
| 172 |
+
"model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
|
| 173 |
+
"model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
|
| 174 |
+
"model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 184 |
+
"model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 186 |
+
"model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 198 |
+
"model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 200 |
+
"model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 208 |
+
"model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 209 |
+
"model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 212 |
+
"model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 214 |
+
"model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 217 |
+
"model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 218 |
+
"model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 219 |
+
"model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 220 |
+
"model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 221 |
+
"model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 222 |
+
"model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 223 |
+
"model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 224 |
+
"model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 225 |
+
"model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 227 |
+
"model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 229 |
+
"model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 230 |
+
"model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 231 |
+
"model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 232 |
+
"model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 233 |
+
"model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 234 |
+
"model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 235 |
+
"model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 236 |
+
"model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 237 |
+
"model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 238 |
+
"model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 239 |
+
"model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 240 |
+
"model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 241 |
+
"model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
|
| 242 |
+
"model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
|
| 243 |
+
"model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
|
| 244 |
+
"model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
|
| 245 |
+
"model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 246 |
+
"model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 247 |
+
"model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 248 |
+
"model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 249 |
+
"model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 250 |
+
"model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
|
| 251 |
+
"model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
|
| 252 |
+
"model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
|
| 253 |
+
"model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
|
| 254 |
+
"model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
|
| 255 |
+
"model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 262 |
+
"model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 263 |
+
"model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 264 |
+
"model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 265 |
+
"model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 266 |
+
"model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 267 |
+
"model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 268 |
+
"model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 269 |
+
"model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 270 |
+
"model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 271 |
+
"model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 272 |
+
"model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 273 |
+
"model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 274 |
+
"model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 275 |
+
"model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 276 |
+
"model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 277 |
+
"model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 278 |
+
"model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 279 |
+
"model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 280 |
+
"model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 281 |
+
"model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 282 |
+
"model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 283 |
+
"model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 284 |
+
"model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 285 |
+
"model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 286 |
+
"model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 287 |
+
"model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 288 |
+
"model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 289 |
+
"model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 290 |
+
"model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 291 |
+
"model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 292 |
+
"model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 293 |
+
"model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 294 |
+
"model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 295 |
+
"model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 296 |
+
"model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 297 |
+
"model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 298 |
+
"model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 299 |
+
"model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 300 |
+
"model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 301 |
+
"model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 302 |
+
"model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 303 |
+
"model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 304 |
+
"model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 305 |
+
"model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 306 |
+
"model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 307 |
+
"model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 308 |
+
"model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 309 |
+
"model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 310 |
+
"model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 311 |
+
"model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 312 |
+
"model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 313 |
+
"model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 314 |
+
"model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 315 |
+
"model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 317 |
+
"model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 318 |
+
"model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 319 |
+
"model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 320 |
+
"model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 321 |
+
"model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 322 |
+
"model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 323 |
+
"model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 324 |
+
"model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 325 |
+
"model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 326 |
+
"model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 327 |
+
"model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 328 |
+
"model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
|
| 329 |
+
"model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
|
| 330 |
+
"model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
|
| 331 |
+
"model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
|
| 332 |
+
"model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
|
| 333 |
+
"model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
|
| 334 |
+
"model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
|
| 335 |
+
"model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
|
| 336 |
+
"model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
|
| 337 |
+
"model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 338 |
+
"model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 339 |
+
"model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 340 |
+
"model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 341 |
+
"model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 342 |
+
"model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 343 |
+
"model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 344 |
+
"model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 345 |
+
"model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 346 |
+
"model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 347 |
+
"model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 348 |
+
"model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 349 |
+
"model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 350 |
+
"model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 351 |
+
"model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 352 |
+
"model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 353 |
+
"model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 354 |
+
"model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 355 |
+
"model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 356 |
+
"model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 357 |
+
"model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 358 |
+
"model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 359 |
+
"model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 360 |
+
"model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 361 |
+
"model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 362 |
+
"model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 363 |
+
"model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 364 |
+
"model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 365 |
+
"model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 366 |
+
"model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 367 |
+
"model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 368 |
+
"model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 369 |
+
"model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 370 |
+
"model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 371 |
+
"model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 372 |
+
"model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 373 |
+
"model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 374 |
+
"model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 375 |
+
"model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 376 |
+
"model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 377 |
+
"model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 378 |
+
"model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 379 |
+
"model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 380 |
+
"model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 381 |
+
"model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 382 |
+
"model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 383 |
+
"model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 384 |
+
"model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 385 |
+
"model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 386 |
+
"model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 387 |
+
"model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 388 |
+
"model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 389 |
+
"model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 390 |
+
"model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 391 |
+
"model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 392 |
+
"model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 393 |
+
"model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 394 |
+
"model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 395 |
+
"model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 396 |
+
"model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 397 |
+
"model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 398 |
+
"model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 399 |
+
"model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 400 |
+
"model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 401 |
+
"model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 402 |
+
"model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 403 |
+
"model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 404 |
+
"model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 405 |
+
"model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 406 |
+
"model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 407 |
+
"model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 408 |
+
"model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 409 |
+
"model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 410 |
+
"model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 411 |
+
"model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 412 |
+
"model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 413 |
+
"model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 414 |
+
"model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 415 |
+
"model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 416 |
+
"model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 417 |
+
"model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 418 |
+
"model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 419 |
+
"model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 420 |
+
"model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
|
| 421 |
+
"model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
|
| 422 |
+
"model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
|
| 423 |
+
"model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
|
| 424 |
+
"model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
|
| 425 |
+
"model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
|
| 426 |
+
"model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
|
| 427 |
+
"model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
|
| 428 |
+
"model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
|
| 429 |
+
"model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 430 |
+
"model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 431 |
+
"model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 432 |
+
"model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 433 |
+
"model.norm.weight": "model-00004-of-00004.safetensors",
|
| 434 |
+
"q_0": "model-00004-of-00004.safetensors",
|
| 435 |
+
"q_1": "model-00004-of-00004.safetensors",
|
| 436 |
+
"q_2": "model-00004-of-00004.safetensors",
|
| 437 |
+
"q_3": "model-00004-of-00004.safetensors",
|
| 438 |
+
"q_4": "model-00004-of-00004.safetensors",
|
| 439 |
+
"q_5": "model-00004-of-00004.safetensors",
|
| 440 |
+
"q_6": "model-00004-of-00004.safetensors",
|
| 441 |
+
"q_7": "model-00004-of-00004.safetensors"
|
| 442 |
+
}
|
| 443 |
+
}
|
modeling_merlin_agent.py
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""Merlin-Agent modeling: Qwen3_5 text causal LM with per-layer quantum injection.
|
| 2 |
+
|
| 3 |
+
At each full-attention layer, a fixed quantum-derived direction (buffer q_k) is
|
| 4 |
+
added to that layer's output hidden state with an RMS-matched, alpha-scaled
|
| 5 |
+
magnitude. Injection lives in the model (buffers + hooks re-installed in
|
| 6 |
+
__init__), so it survives save/reload — not a bare, non-persisted hook.
|
| 7 |
+
|
| 8 |
+
alpha == 0 -> exact base model.
|
| 9 |
+
"""
|
| 10 |
+
import torch
|
| 11 |
+
from transformers import Qwen3_5ForCausalLM
|
| 12 |
+
|
| 13 |
+
try:
|
| 14 |
+
from .configuration_merlin_agent import MerlinAgentConfig
|
| 15 |
+
except ImportError: # when loaded as flat remote code (trust_remote_code)
|
| 16 |
+
from configuration_merlin_agent import MerlinAgentConfig
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
def _inject(h: torch.Tensor, q: torch.Tensor, alpha: float) -> torch.Tensor:
|
| 20 |
+
if alpha == 0:
|
| 21 |
+
return h
|
| 22 |
+
q = q.to(dtype=h.dtype, device=h.device)
|
| 23 |
+
rms_h = h.pow(2).mean(dim=-1, keepdim=True).sqrt()
|
| 24 |
+
rms_q = q.pow(2).mean().sqrt().clamp_min(1e-6)
|
| 25 |
+
return h + alpha * (rms_h / rms_q) * q
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
class MerlinAgentForCausalLM(Qwen3_5ForCausalLM):
|
| 29 |
+
config_class = MerlinAgentConfig
|
| 30 |
+
|
| 31 |
+
def __init__(self, config):
|
| 32 |
+
super().__init__(config)
|
| 33 |
+
self._inj_layers = list(config.full_attention_layer_indices)
|
| 34 |
+
for k in range(len(self._inj_layers)):
|
| 35 |
+
self.register_buffer(f"q_{k}", torch.zeros(config.hidden_size), persistent=True)
|
| 36 |
+
self._install_injection_hooks()
|
| 37 |
+
|
| 38 |
+
def _install_injection_hooks(self):
|
| 39 |
+
for k, li in enumerate(self._inj_layers):
|
| 40 |
+
self.model.layers[li].register_forward_hook(self._make_hook(f"q_{k}"))
|
| 41 |
+
|
| 42 |
+
def _make_hook(self, qk: str):
|
| 43 |
+
def hook(module, args, output):
|
| 44 |
+
q = getattr(self, qk)
|
| 45 |
+
a = float(self.config.quantum_injection_alpha)
|
| 46 |
+
if isinstance(output, tuple):
|
| 47 |
+
return (_inject(output[0], q, a),) + tuple(output[1:])
|
| 48 |
+
return _inject(output, q, a)
|
| 49 |
+
return hook
|
| 50 |
+
|
| 51 |
+
def set_quantum_signatures(self, q_vectors):
|
| 52 |
+
"""Load the 8 projected quantum direction vectors (each shape [hidden_size])."""
|
| 53 |
+
assert len(q_vectors) == len(self._inj_layers)
|
| 54 |
+
for k, v in enumerate(q_vectors):
|
| 55 |
+
getattr(self, f"q_{k}").copy_(torch.as_tensor(v, dtype=torch.float32))
|
quantum_attestation.json
ADDED
|
@@ -0,0 +1,210 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"scheme": "SHA256 leaves -> Merkle root",
|
| 3 |
+
"proj_seed": 42,
|
| 4 |
+
"backend": "ibm_marrakesh",
|
| 5 |
+
"collected_at": "2026-07-02T06:07:00.294531+00:00",
|
| 6 |
+
"attestation_root": "0afa57c3bc66820ed5d37b0e7a37463ce4bfdb67444035aaacce80e87e3a9911",
|
| 7 |
+
"leaves": [
|
| 8 |
+
{
|
| 9 |
+
"leaf": "50ad0b6b24bbb2a2f585e90c034a94a7e55ffb9fd3f48e717f82328760c4ec5c",
|
| 10 |
+
"slot": 0,
|
| 11 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 12 |
+
"backend": "ibm_marrakesh",
|
| 13 |
+
"n_qubits": 100,
|
| 14 |
+
"shots": 2048,
|
| 15 |
+
"syk_depths": [
|
| 16 |
+
1,
|
| 17 |
+
2,
|
| 18 |
+
3,
|
| 19 |
+
4,
|
| 20 |
+
5,
|
| 21 |
+
6
|
| 22 |
+
],
|
| 23 |
+
"otoc_6d": [
|
| 24 |
+
0.08673828,
|
| 25 |
+
0.06501953,
|
| 26 |
+
0.06294922,
|
| 27 |
+
0.06214844,
|
| 28 |
+
0.06146484,
|
| 29 |
+
0.06382812
|
| 30 |
+
],
|
| 31 |
+
"proj_sha256": "0d60cc623788ff08694ed0ad5bccce16194f72bea472e26bbd6fd6609ba26822"
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"leaf": "b4abea3cdff44d9c02fcdc772a5d0b0454f5117b8d2c4611a81b7b5c9f765595",
|
| 35 |
+
"slot": 1,
|
| 36 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 37 |
+
"backend": "ibm_marrakesh",
|
| 38 |
+
"n_qubits": 100,
|
| 39 |
+
"shots": 2048,
|
| 40 |
+
"syk_depths": [
|
| 41 |
+
1,
|
| 42 |
+
2,
|
| 43 |
+
3,
|
| 44 |
+
4,
|
| 45 |
+
5,
|
| 46 |
+
6
|
| 47 |
+
],
|
| 48 |
+
"otoc_6d": [
|
| 49 |
+
0.19728516,
|
| 50 |
+
0.065625,
|
| 51 |
+
0.06103516,
|
| 52 |
+
0.06363281,
|
| 53 |
+
0.06310547,
|
| 54 |
+
0.06396484
|
| 55 |
+
],
|
| 56 |
+
"proj_sha256": "92d77034456235a64faeba4b5985d1d50e350b18692dc1f5ebc090d2061950f5"
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"leaf": "9bee906e97fae1433a61c91a2a7072b4e46e6b6dbe6bcec5581312530b79c2fb",
|
| 60 |
+
"slot": 2,
|
| 61 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 62 |
+
"backend": "ibm_marrakesh",
|
| 63 |
+
"n_qubits": 100,
|
| 64 |
+
"shots": 2048,
|
| 65 |
+
"syk_depths": [
|
| 66 |
+
1,
|
| 67 |
+
2,
|
| 68 |
+
3,
|
| 69 |
+
4,
|
| 70 |
+
5,
|
| 71 |
+
6
|
| 72 |
+
],
|
| 73 |
+
"otoc_6d": [
|
| 74 |
+
0.12580078,
|
| 75 |
+
0.06103516,
|
| 76 |
+
0.06216797,
|
| 77 |
+
0.0609375,
|
| 78 |
+
0.06306641,
|
| 79 |
+
0.06375
|
| 80 |
+
],
|
| 81 |
+
"proj_sha256": "07b5431ddc3f5f74d7702f0cafbe3c67c7dd12ee5cb7ad529b3197155f0a7449"
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"leaf": "8231e1fd3fa048d278474d7842b00facf8658317972d01899caa309378f82b30",
|
| 85 |
+
"slot": 3,
|
| 86 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 87 |
+
"backend": "ibm_marrakesh",
|
| 88 |
+
"n_qubits": 100,
|
| 89 |
+
"shots": 2048,
|
| 90 |
+
"syk_depths": [
|
| 91 |
+
1,
|
| 92 |
+
2,
|
| 93 |
+
3,
|
| 94 |
+
4,
|
| 95 |
+
5,
|
| 96 |
+
6
|
| 97 |
+
],
|
| 98 |
+
"otoc_6d": [
|
| 99 |
+
0.24994141,
|
| 100 |
+
0.10943359,
|
| 101 |
+
0.05970703,
|
| 102 |
+
0.06130859,
|
| 103 |
+
0.06320312,
|
| 104 |
+
0.06275391
|
| 105 |
+
],
|
| 106 |
+
"proj_sha256": "7fd2c87a3618adf4835765649179875a104d924ea0ab9aa0eaf2ebf60356cb65"
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"leaf": "1028d1a70b54bc95baf813cc18a3c267d4fa408b04a657f35b007fa8d0505649",
|
| 110 |
+
"slot": 4,
|
| 111 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 112 |
+
"backend": "ibm_marrakesh",
|
| 113 |
+
"n_qubits": 100,
|
| 114 |
+
"shots": 2048,
|
| 115 |
+
"syk_depths": [
|
| 116 |
+
1,
|
| 117 |
+
2,
|
| 118 |
+
3,
|
| 119 |
+
4,
|
| 120 |
+
5,
|
| 121 |
+
6
|
| 122 |
+
],
|
| 123 |
+
"otoc_6d": [
|
| 124 |
+
0.06734375,
|
| 125 |
+
0.06650391,
|
| 126 |
+
0.0615625,
|
| 127 |
+
0.06162109,
|
| 128 |
+
0.0646875,
|
| 129 |
+
0.06271484
|
| 130 |
+
],
|
| 131 |
+
"proj_sha256": "613fc240263fcd57135958d2f21678c8bcdf543de7c82dbcd325a63b20b8d852"
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"leaf": "8bc02b090967370acd8e250b2c2f53547713577e44f8207fde5d22d829b82fc1",
|
| 135 |
+
"slot": 5,
|
| 136 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 137 |
+
"backend": "ibm_marrakesh",
|
| 138 |
+
"n_qubits": 100,
|
| 139 |
+
"shots": 2048,
|
| 140 |
+
"syk_depths": [
|
| 141 |
+
1,
|
| 142 |
+
2,
|
| 143 |
+
3,
|
| 144 |
+
4,
|
| 145 |
+
5,
|
| 146 |
+
6
|
| 147 |
+
],
|
| 148 |
+
"otoc_6d": [
|
| 149 |
+
0.07216797,
|
| 150 |
+
0.07421875,
|
| 151 |
+
0.07933594,
|
| 152 |
+
0.063125,
|
| 153 |
+
0.06453125,
|
| 154 |
+
0.06048828
|
| 155 |
+
],
|
| 156 |
+
"proj_sha256": "5da8b4ae463e7b0336d400274be885f1eb1857e2c43eb24e050bb1de5e8b1236"
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"leaf": "4242a696cd06b57e471b841d6a37df6ea630c0fae9c282a5169d70ce8f16e047",
|
| 160 |
+
"slot": 6,
|
| 161 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 162 |
+
"backend": "ibm_marrakesh",
|
| 163 |
+
"n_qubits": 100,
|
| 164 |
+
"shots": 2048,
|
| 165 |
+
"syk_depths": [
|
| 166 |
+
1,
|
| 167 |
+
2,
|
| 168 |
+
3,
|
| 169 |
+
4,
|
| 170 |
+
5,
|
| 171 |
+
6
|
| 172 |
+
],
|
| 173 |
+
"otoc_6d": [
|
| 174 |
+
0.28207031,
|
| 175 |
+
0.09130859,
|
| 176 |
+
0.06955078,
|
| 177 |
+
0.06390625,
|
| 178 |
+
0.06066406,
|
| 179 |
+
0.06275391
|
| 180 |
+
],
|
| 181 |
+
"proj_sha256": "1d5c2f9050bc3c53d6516f4d7f66dc53e2a499f37a18a86c0575db40abfe45a1"
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"leaf": "8bfab403055ab0b1553c1a839d2beac4bd6a166bd2376af93c1a9a69461ec822",
|
| 185 |
+
"slot": 7,
|
| 186 |
+
"ibm_job_id": "d92ve0t958jc73bsbong",
|
| 187 |
+
"backend": "ibm_marrakesh",
|
| 188 |
+
"n_qubits": 100,
|
| 189 |
+
"shots": 2048,
|
| 190 |
+
"syk_depths": [
|
| 191 |
+
1,
|
| 192 |
+
2,
|
| 193 |
+
3,
|
| 194 |
+
4,
|
| 195 |
+
5,
|
| 196 |
+
6
|
| 197 |
+
],
|
| 198 |
+
"otoc_6d": [
|
| 199 |
+
0.14970703,
|
| 200 |
+
0.06710937,
|
| 201 |
+
0.07498047,
|
| 202 |
+
0.06775391,
|
| 203 |
+
0.06279297,
|
| 204 |
+
0.06373047
|
| 205 |
+
],
|
| 206 |
+
"proj_sha256": "2a0abaa96d2ef3295cf9af19a475eb5137bdb8590fc54f2cba1705dd16790728"
|
| 207 |
+
}
|
| 208 |
+
],
|
| 209 |
+
"verify": "recompute leaf_k=SHA256(canonical_json(fields)); rebuild Merkle root; query ibm_job_id via QiskitRuntimeService.job(id)."
|
| 210 |
+
}
|
quantum_signatures.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0da2b1a34f54e5d5f1670aad7d3b28752131b0e334b4f74483b53ff862d02fc1
|
| 3 |
+
size 1256
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
|
| 3 |
+
size 19989325
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": false,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|