ZTC-Judge-27B v1.0 — 자립형 (가중치+프로브+추론코드)
Browse files- .gitattributes +1 -0
- README.md +100 -0
- chat_template.jinja +154 -0
- config.json +135 -0
- generation_config.json +13 -0
- merges.txt +0 -0
- model.safetensors-00001-of-00011.safetensors +3 -0
- model.safetensors-00002-of-00011.safetensors +3 -0
- model.safetensors-00003-of-00011.safetensors +3 -0
- model.safetensors-00004-of-00011.safetensors +3 -0
- model.safetensors-00005-of-00011.safetensors +3 -0
- model.safetensors-00006-of-00011.safetensors +3 -0
- model.safetensors-00007-of-00011.safetensors +3 -0
- model.safetensors-00008-of-00011.safetensors +3 -0
- model.safetensors-00009-of-00011.safetensors +3 -0
- model.safetensors-00010-of-00011.safetensors +3 -0
- model.safetensors-00011-of-00011.safetensors +3 -0
- model.safetensors.index.json +0 -0
- tokenizer.json +3 -0
- tokenizer_config.json +305 -0
- vocab.json +0 -0
- ztc_config.json +106 -0
- ztc_judge.py +67 -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 +1,100 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: Qwen/Qwen3.5-27B
|
| 4 |
+
language:
|
| 5 |
+
- ko
|
| 6 |
+
- en
|
| 7 |
+
tags:
|
| 8 |
+
- ztc
|
| 9 |
+
- verification
|
| 10 |
+
- hallucination-detection
|
| 11 |
+
- zero-token
|
| 12 |
+
library_name: transformers
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# ZTC-Judge-27B
|
| 16 |
+
|
| 17 |
+
**모델에게 확신을 물으면 동전던지기(AUC 0.50)입니다. 우리는 0.80 로 잡습니다.**
|
| 18 |
+
|
| 19 |
+
어떤 모델이 쓴 답이든 받아, **글자를 한 개도 생성하지 않고** 한 번 읽어
|
| 20 |
+
그 답을 믿어도 되는지 판정합니다.
|
| 21 |
+
|
| 22 |
+
> **ZTC** = Zero-Token Confidence (무토큰 신뢰도)
|
| 23 |
+
> **Judge** = 자기 답이 아니라 **남의 답**을 심판한다
|
| 24 |
+
|
| 25 |
+
## 동작
|
| 26 |
+
|
| 27 |
+
```
|
| 28 |
+
[문제 + 답변] → 27B 에 한 번 통과 → 마지막 층·마지막 토큰의 은닉(5120차원)
|
| 29 |
+
→ 프로브 벡터와 곱셈 한 번 → 점수
|
| 30 |
+
```
|
| 31 |
+
|
| 32 |
+
생성 토큰 **0개**. 판정 대상 모델의 가중치도, 로그확률도 필요 없습니다.
|
| 33 |
+
|
| 34 |
+
## 쓰는 법
|
| 35 |
+
|
| 36 |
+
```python
|
| 37 |
+
from ztc_judge import ZTCJudge
|
| 38 |
+
j = ZTCJudge.from_pretrained("FINAL-Bench/ZTC-Judge-27B")
|
| 39 |
+
j.score("문제…", "어떤 모델이든 쓴 답…", domain="과학추론")
|
| 40 |
+
# {'score': -0.61, 'verdict': '검토 요망', 'domain_auc': 0.753, 'generated_tokens': 0}
|
| 41 |
+
```
|
| 42 |
+
|
| 43 |
+
묶어서 재면 그대로 빨라집니다 — 생성이 없는 작업이라 배치가 곧 가속입니다.
|
| 44 |
+
|
| 45 |
+
```python
|
| 46 |
+
j.score_batch([(q1, a1), (q2, a2)], domain="과학추론")
|
| 47 |
+
```
|
| 48 |
+
|
| 49 |
+
## 실측 (2026-09-19)
|
| 50 |
+
|
| 51 |
+
| | |
|
| 52 |
+
|---|---|
|
| 53 |
+
| 심판자 | Qwen3.5-27B (판본 `fc05daec18b0`) |
|
| 54 |
+
| 자료 | 2018문항 · 오답 508 · 5분야 · 답 쓴 모델 4종 |
|
| 55 |
+
| 판별력 | **AUC 0.7999** · 순열검정 **z = 25.68** |
|
| 56 |
+
| 프로브 | 5120차원 벡터 1개 (59KB) · 가중치 미변경 |
|
| 57 |
+
|
| 58 |
+
### 분야별 — **그 분야를 빼고 배웠을 때** (처음 보는 분야 기준)
|
| 59 |
+
|
| 60 |
+
| 분야 | AUC | 문항 | 오답 |
|
| 61 |
+
|---|---|---|---|
|
| 62 |
+
| 전문시험(법·수학·생물) | 0.833 | 400 | 80 |
|
| 63 |
+
| 과학추론 | 0.753 | 198 | 20 |
|
| 64 |
+
| 생물·의학 | 0.712 | 917 | 169 |
|
| 65 |
+
| 재난·안전 지침 | 0.682 | 225 | 109 |
|
| 66 |
+
| 상식·다단계추론 | 0.615 | 278 | 130 |
|
| 67 |
+
|
| 68 |
+
🔑 같이 배운 뒤의 값이 아니라 **빼고 배운 값**을 적습니다.
|
| 69 |
+
고객의 분야는 사실상 처음 보는 분야이므로, 그게 정직한 기대치입니다.
|
| 70 |
+
|
| 71 |
+
### 검토 예산
|
| 72 |
+
|
| 73 |
+
| 검토 비율 | 오류 발견율 | 무작위 대비 |
|
| 74 |
+
|---|---|---|
|
| 75 |
+
| 5% | 14.0% | 2.8배 |
|
| 76 |
+
| 10% | 28.0% | 2.8배 |
|
| 77 |
+
| 15% | 38.4% | 2.6배 |
|
| 78 |
+
| 20% | 47.2% | 2.4배 |
|
| 79 |
+
| 30% | 62.0% | 2.1배 |
|
| 80 |
+
| 40% | 72.6% | 1.8배 |
|
| 81 |
+
|
| 82 |
+
## 🔴 한계
|
| 83 |
+
|
| 84 |
+
- **위 5분야 밖에는 점수를 주지 않습니다.** `domain_auc` 가 `None` 이면 측정 안 된 분야입니다.
|
| 85 |
+
- 가장 약한 분야는 **0.615** 입니다. 약한 것을 먼저 적습니다.
|
| 86 |
+
- 판본이 고정입니다. 다른 판본의 Qwen3.5-27B 위에서 이 프로브를 쓰면
|
| 87 |
+
**오류 없이 점수만 조용히 틀려집니다.** 이 저장소는 가중치를 함께 담아 그 사고를 막습니다.
|
| 88 |
+
- 이것은 **검증기의 판단**이며 답을 쓴 모델의 확신도가 아닙니다.
|
| 89 |
+
|
| 90 |
+
## 포함 / 미포함
|
| 91 |
+
|
| 92 |
+
| | |
|
| 93 |
+
|---|---|
|
| 94 |
+
| ✅ 포함 | 기반 가중치 · 토크나이저 · 프로브 · 추론코드 · 실측 설정 |
|
| 95 |
+
| 🔴 미포함 | 학습 자료 · 은닉 행렬 · 적합 파이프라인 |
|
| 96 |
+
|
| 97 |
+
## 라이선스
|
| 98 |
+
|
| 99 |
+
기반 모델 Qwen3.5-27B 는 Apache-2.0 이며 재배포가 허용됩니다.
|
| 100 |
+
프로브·추론코드·실측표는 **FINAL-Bench / VIDRAFT** 자산입니다.
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- 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 %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,135 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"image_token_id": 248056,
|
| 6 |
+
"model_type": "qwen3_5",
|
| 7 |
+
"text_config": {
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"attn_output_gate": true,
|
| 11 |
+
"dtype": "bfloat16",
|
| 12 |
+
"eos_token_id": 248044,
|
| 13 |
+
"full_attention_interval": 4,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_act": "silu",
|
| 16 |
+
"hidden_size": 5120,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 17408,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"linear_attention",
|
| 21 |
+
"linear_attention",
|
| 22 |
+
"linear_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"linear_attention",
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"linear_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"full_attention",
|
| 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 |
+
"linear_attention",
|
| 65 |
+
"linear_attention",
|
| 66 |
+
"linear_attention",
|
| 67 |
+
"full_attention",
|
| 68 |
+
"linear_attention",
|
| 69 |
+
"linear_attention",
|
| 70 |
+
"linear_attention",
|
| 71 |
+
"full_attention",
|
| 72 |
+
"linear_attention",
|
| 73 |
+
"linear_attention",
|
| 74 |
+
"linear_attention",
|
| 75 |
+
"full_attention",
|
| 76 |
+
"linear_attention",
|
| 77 |
+
"linear_attention",
|
| 78 |
+
"linear_attention",
|
| 79 |
+
"full_attention",
|
| 80 |
+
"linear_attention",
|
| 81 |
+
"linear_attention",
|
| 82 |
+
"linear_attention",
|
| 83 |
+
"full_attention"
|
| 84 |
+
],
|
| 85 |
+
"linear_conv_kernel_dim": 4,
|
| 86 |
+
"linear_key_head_dim": 128,
|
| 87 |
+
"linear_num_key_heads": 16,
|
| 88 |
+
"linear_num_value_heads": 48,
|
| 89 |
+
"linear_value_head_dim": 128,
|
| 90 |
+
"max_position_embeddings": 262144,
|
| 91 |
+
"mlp_only_layers": [],
|
| 92 |
+
"model_type": "qwen3_5_text",
|
| 93 |
+
"mtp_num_hidden_layers": 1,
|
| 94 |
+
"mtp_use_dedicated_embeddings": false,
|
| 95 |
+
"num_attention_heads": 24,
|
| 96 |
+
"num_hidden_layers": 64,
|
| 97 |
+
"num_key_value_heads": 4,
|
| 98 |
+
"rms_norm_eps": 1e-06,
|
| 99 |
+
"use_cache": true,
|
| 100 |
+
"vocab_size": 248320,
|
| 101 |
+
"mamba_ssm_dtype": "float32",
|
| 102 |
+
"rope_parameters": {
|
| 103 |
+
"mrope_interleaved": true,
|
| 104 |
+
"mrope_section": [
|
| 105 |
+
11,
|
| 106 |
+
11,
|
| 107 |
+
10
|
| 108 |
+
],
|
| 109 |
+
"rope_type": "default",
|
| 110 |
+
"rope_theta": 10000000,
|
| 111 |
+
"partial_rotary_factor": 0.25
|
| 112 |
+
}
|
| 113 |
+
},
|
| 114 |
+
"tie_word_embeddings": false,
|
| 115 |
+
"transformers_version": "4.57.0.dev0",
|
| 116 |
+
"video_token_id": 248057,
|
| 117 |
+
"vision_config": {
|
| 118 |
+
"deepstack_visual_indexes": [],
|
| 119 |
+
"depth": 27,
|
| 120 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 121 |
+
"hidden_size": 1152,
|
| 122 |
+
"in_channels": 3,
|
| 123 |
+
"initializer_range": 0.02,
|
| 124 |
+
"intermediate_size": 4304,
|
| 125 |
+
"model_type": "qwen3_5",
|
| 126 |
+
"num_heads": 16,
|
| 127 |
+
"num_position_embeddings": 2304,
|
| 128 |
+
"out_hidden_size": 5120,
|
| 129 |
+
"patch_size": 16,
|
| 130 |
+
"spatial_merge_size": 2,
|
| 131 |
+
"temporal_patch_size": 2
|
| 132 |
+
},
|
| 133 |
+
"vision_end_token_id": 248054,
|
| 134 |
+
"vision_start_token_id": 248053
|
| 135 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 248044,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
248046,
|
| 6 |
+
248044
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 248044,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_k": 20,
|
| 11 |
+
"top_p": 0.95,
|
| 12 |
+
"transformers_version": "4.57.0.dev0"
|
| 13 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors-00001-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9019228d172c87d5603266c2d56672d119e838facffa164de803a1ebf0d716d2
|
| 3 |
+
size 5263851872
|
model.safetensors-00002-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:890ef00c920b01c1c02755088c8d8ca5cdd6a1faa2a9a729eecd00857648b411
|
| 3 |
+
size 5347741440
|
model.safetensors-00003-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8aca03689ad0717fb91455809a6670eecca818bea8c0b6bd3a591a8807cc3223
|
| 3 |
+
size 5347741504
|
model.safetensors-00004-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20f539430c60fa611b3522b8408749a080b63f1fe034e551a6539ef864ca079a
|
| 3 |
+
size 5347741504
|
model.safetensors-00005-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57a0c074c654f05fc2d6b5112eeec387ac04986e928e42f502993e69aa03a49d
|
| 3 |
+
size 5347741504
|
model.safetensors-00006-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfa4e6fbfc600854ef6c8e5465d0a1e43832a64b5e539f1e880333e3b1086703
|
| 3 |
+
size 5347741504
|
model.safetensors-00007-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d426963325b2319cb9e1442bc8b89b9a7748a8f7a907aa4041c11531cda6b014
|
| 3 |
+
size 5347741496
|
model.safetensors-00008-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe60dbb9d25354c4eb2a9a59d9b1afd07741c8bebab5870f5eea6eadb4cf9a06
|
| 3 |
+
size 5368714128
|
model.safetensors-00009-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71a153d882242734a1fc7e000734727f00f6ec8ca70b044f5e44fcf97736ef8f
|
| 3 |
+
size 5347745520
|
model.safetensors-00010-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:146745698b9f21940e2982beb1816a0eef3b80d77ab9cd884695b9a08c2697eb
|
| 3 |
+
size 5347749200
|
model.safetensors-00011-of-00011.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d947ce7483c4109b55039f1359f4494d22390cf123568100abd89816802f097d
|
| 3 |
+
size 2148512760
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42
|
| 3 |
+
size 12807982
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,305 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"248044": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"248045": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"248046": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"248047": {
|
| 29 |
+
"content": "<|object_ref_start|>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"248048": {
|
| 37 |
+
"content": "<|object_ref_end|>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"248049": {
|
| 45 |
+
"content": "<|box_start|>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"248050": {
|
| 53 |
+
"content": "<|box_end|>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"248051": {
|
| 61 |
+
"content": "<|quad_start|>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"248052": {
|
| 69 |
+
"content": "<|quad_end|>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"248053": {
|
| 77 |
+
"content": "<|vision_start|>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"248054": {
|
| 85 |
+
"content": "<|vision_end|>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"248055": {
|
| 93 |
+
"content": "<|vision_pad|>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"248056": {
|
| 101 |
+
"content": "<|image_pad|>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"248057": {
|
| 109 |
+
"content": "<|video_pad|>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
},
|
| 116 |
+
"248058": {
|
| 117 |
+
"content": "<tool_call>",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": false,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": false
|
| 123 |
+
},
|
| 124 |
+
"248059": {
|
| 125 |
+
"content": "</tool_call>",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": false,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": false
|
| 131 |
+
},
|
| 132 |
+
"248060": {
|
| 133 |
+
"content": "<|fim_prefix|>",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": false,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": false
|
| 139 |
+
},
|
| 140 |
+
"248061": {
|
| 141 |
+
"content": "<|fim_middle|>",
|
| 142 |
+
"lstrip": false,
|
| 143 |
+
"normalized": false,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": false
|
| 147 |
+
},
|
| 148 |
+
"248062": {
|
| 149 |
+
"content": "<|fim_suffix|>",
|
| 150 |
+
"lstrip": false,
|
| 151 |
+
"normalized": false,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": false
|
| 155 |
+
},
|
| 156 |
+
"248063": {
|
| 157 |
+
"content": "<|fim_pad|>",
|
| 158 |
+
"lstrip": false,
|
| 159 |
+
"normalized": false,
|
| 160 |
+
"rstrip": false,
|
| 161 |
+
"single_word": false,
|
| 162 |
+
"special": false
|
| 163 |
+
},
|
| 164 |
+
"248064": {
|
| 165 |
+
"content": "<|repo_name|>",
|
| 166 |
+
"lstrip": false,
|
| 167 |
+
"normalized": false,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false,
|
| 170 |
+
"special": false
|
| 171 |
+
},
|
| 172 |
+
"248065": {
|
| 173 |
+
"content": "<|file_sep|>",
|
| 174 |
+
"lstrip": false,
|
| 175 |
+
"normalized": false,
|
| 176 |
+
"rstrip": false,
|
| 177 |
+
"single_word": false,
|
| 178 |
+
"special": false
|
| 179 |
+
},
|
| 180 |
+
"248066": {
|
| 181 |
+
"content": "<tool_response>",
|
| 182 |
+
"lstrip": false,
|
| 183 |
+
"normalized": false,
|
| 184 |
+
"rstrip": false,
|
| 185 |
+
"single_word": false,
|
| 186 |
+
"special": false
|
| 187 |
+
},
|
| 188 |
+
"248067": {
|
| 189 |
+
"content": "</tool_response>",
|
| 190 |
+
"lstrip": false,
|
| 191 |
+
"normalized": false,
|
| 192 |
+
"rstrip": false,
|
| 193 |
+
"single_word": false,
|
| 194 |
+
"special": false
|
| 195 |
+
},
|
| 196 |
+
"248068": {
|
| 197 |
+
"content": "<think>",
|
| 198 |
+
"lstrip": false,
|
| 199 |
+
"normalized": false,
|
| 200 |
+
"rstrip": false,
|
| 201 |
+
"single_word": false,
|
| 202 |
+
"special": false
|
| 203 |
+
},
|
| 204 |
+
"248069": {
|
| 205 |
+
"content": "</think>",
|
| 206 |
+
"lstrip": false,
|
| 207 |
+
"normalized": false,
|
| 208 |
+
"rstrip": false,
|
| 209 |
+
"single_word": false,
|
| 210 |
+
"special": false
|
| 211 |
+
},
|
| 212 |
+
"248070": {
|
| 213 |
+
"content": "<|audio_start|>",
|
| 214 |
+
"lstrip": false,
|
| 215 |
+
"normalized": false,
|
| 216 |
+
"rstrip": false,
|
| 217 |
+
"single_word": false,
|
| 218 |
+
"special": true
|
| 219 |
+
},
|
| 220 |
+
"248071": {
|
| 221 |
+
"content": "<|audio_end|>",
|
| 222 |
+
"lstrip": false,
|
| 223 |
+
"normalized": false,
|
| 224 |
+
"rstrip": false,
|
| 225 |
+
"single_word": false,
|
| 226 |
+
"special": true
|
| 227 |
+
},
|
| 228 |
+
"248072": {
|
| 229 |
+
"content": "<tts_pad>",
|
| 230 |
+
"lstrip": false,
|
| 231 |
+
"normalized": false,
|
| 232 |
+
"rstrip": false,
|
| 233 |
+
"single_word": false,
|
| 234 |
+
"special": true
|
| 235 |
+
},
|
| 236 |
+
"248073": {
|
| 237 |
+
"content": "<tts_text_bos>",
|
| 238 |
+
"lstrip": false,
|
| 239 |
+
"normalized": false,
|
| 240 |
+
"rstrip": false,
|
| 241 |
+
"single_word": false,
|
| 242 |
+
"special": true
|
| 243 |
+
},
|
| 244 |
+
"248074": {
|
| 245 |
+
"content": "<tts_text_eod>",
|
| 246 |
+
"lstrip": false,
|
| 247 |
+
"normalized": false,
|
| 248 |
+
"rstrip": false,
|
| 249 |
+
"single_word": false,
|
| 250 |
+
"special": true
|
| 251 |
+
},
|
| 252 |
+
"248075": {
|
| 253 |
+
"content": "<tts_text_bos_single>",
|
| 254 |
+
"lstrip": false,
|
| 255 |
+
"normalized": false,
|
| 256 |
+
"rstrip": false,
|
| 257 |
+
"single_word": false,
|
| 258 |
+
"special": true
|
| 259 |
+
},
|
| 260 |
+
"248076": {
|
| 261 |
+
"content": "<|audio_pad|>",
|
| 262 |
+
"lstrip": false,
|
| 263 |
+
"normalized": false,
|
| 264 |
+
"rstrip": false,
|
| 265 |
+
"single_word": false,
|
| 266 |
+
"special": true
|
| 267 |
+
}
|
| 268 |
+
},
|
| 269 |
+
"additional_special_tokens": [
|
| 270 |
+
"<|im_start|>",
|
| 271 |
+
"<|im_end|>",
|
| 272 |
+
"<|object_ref_start|>",
|
| 273 |
+
"<|object_ref_end|>",
|
| 274 |
+
"<|box_start|>",
|
| 275 |
+
"<|box_end|>",
|
| 276 |
+
"<|quad_start|>",
|
| 277 |
+
"<|quad_end|>",
|
| 278 |
+
"<|vision_start|>",
|
| 279 |
+
"<|vision_end|>",
|
| 280 |
+
"<|vision_pad|>",
|
| 281 |
+
"<|image_pad|>",
|
| 282 |
+
"<|video_pad|>"
|
| 283 |
+
],
|
| 284 |
+
"bos_token": null,
|
| 285 |
+
"chat_template": "{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- macro render_content(content, do_vision_count, is_system_content=false) %}\n {%- if content is string %}\n {{- content }}\n {%- elif content is iterable and content is not mapping %}\n {%- for item in content %}\n {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain images.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Picture ' ~ image_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|image_pad|><|vision_end|>' }}\n {%- elif 'video' in item or item.type == 'video' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain videos.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Video ' ~ video_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|video_pad|><|vision_end|>' }}\n {%- elif 'text' in item %}\n {{- item.text }}\n {%- else %}\n {{- raise_exception('Unexpected item type in content.') }}\n {%- endif %}\n {%- endfor %}\n {%- elif content is none or content is undefined %}\n {{- '' }}\n {%- else %}\n {{- raise_exception('Unexpected content type.') }}\n {%- endif %}\n{%- endmacro %}\n{%- if not messages %}\n {{- raise_exception('No messages provided.') }}\n{%- endif %}\n{%- if tools and tools is iterable and tools is not mapping %}\n {{- '<|im_start|>system\\n' }}\n {{- \"# Tools\\n\\nYou have access to the following functions:\\n\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\" }}\n {{- '\\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>' }}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {%- if content %}\n {{- '\\n\\n' + content }}\n {%- endif %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {{- '<|im_start|>system\\n' + content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" %}\n {%- set content = render_content(message.content, false)|trim %}\n {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if ns.multi_step_tool %}\n {{- raise_exception('No user query found in messages.') }}\n{%- endif %}\n{%- for message in messages %}\n {%- set content = render_content(message.content, true)|trim %}\n {%- if message.role == \"system\" %}\n {%- if not loop.first %}\n {{- raise_exception('System message must be at the beginning.') }}\n {%- endif %}\n {%- elif message.role == \"user\" %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in content %}\n {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- set reasoning_content = reasoning_content|trim %}\n {%- if loop.index0 > ns.last_query_index %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content + '\\n</think>\\n\\n' + content }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- if loop.first %}\n {%- if content|trim %}\n {{- '\\n\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- else %}\n {{- '<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- else %}\n {{- '\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- if tool_call.arguments is defined %}\n {%- for args_name, args_value in tool_call.arguments|items %}\n {{- '<parameter=' + args_name + '>\\n' }}\n {%- 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 %}\n {{- args_value }}\n {{- '\\n</parameter>\\n' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.previtem and loop.previtem.role != \"tool\" %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if not loop.last and loop.nextitem.role != \"tool\" %}\n {{- '<|im_end|>\\n' }}\n {%- elif loop.last %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- else %}\n {{- raise_exception('Unexpected message role.') }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is false %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- else %}\n {{- '<think>\\n' }}\n {%- endif %}\n{%- endif %}",
|
| 286 |
+
"clean_up_tokenization_spaces": false,
|
| 287 |
+
"eos_token": "<|im_end|>",
|
| 288 |
+
"errors": "replace",
|
| 289 |
+
"model_max_length": 262144,
|
| 290 |
+
"pad_token": "<|endoftext|>",
|
| 291 |
+
"split_special_tokens": false,
|
| 292 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 293 |
+
"unk_token": null,
|
| 294 |
+
"add_bos_token": false,
|
| 295 |
+
"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+",
|
| 296 |
+
"extra_special_tokens": {
|
| 297 |
+
"audio_bos_token": "<|audio_start|>",
|
| 298 |
+
"audio_eos_token": "<|audio_end|>",
|
| 299 |
+
"audio_token": "<|audio_pad|>",
|
| 300 |
+
"image_token": "<|image_pad|>",
|
| 301 |
+
"video_token": "<|video_pad|>",
|
| 302 |
+
"vision_bos_token": "<|vision_start|>",
|
| 303 |
+
"vision_eos_token": "<|vision_end|>"
|
| 304 |
+
}
|
| 305 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ztc_config.json
ADDED
|
@@ -0,0 +1,106 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"ztc_version": "1.0",
|
| 3 |
+
"base_model": "Qwen/Qwen3.5-27B",
|
| 4 |
+
"base_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
|
| 5 |
+
"probe_file": "ZTC-Judge-27B.npz",
|
| 6 |
+
"hidden_dim": 5120,
|
| 7 |
+
"layer": "last",
|
| 8 |
+
"token": "last_non_pad",
|
| 9 |
+
"max_length": 2048,
|
| 10 |
+
"template": "다음은 어떤 문제와 그에 대한 답변이다. 이 답변이 옳은지 판단하라.\n\n[문제]\n%s\n\n[답변]\n%s\n\n이 답변은 옳은가?",
|
| 11 |
+
"thresholds": {
|
| 12 |
+
"review": -0.392403355096007,
|
| 13 |
+
"clear": 0.11326689029648167
|
| 14 |
+
},
|
| 15 |
+
"measured": {
|
| 16 |
+
"cv_auc": 0.7999247390498649,
|
| 17 |
+
"permutation_z": 25.68107120241526,
|
| 18 |
+
"n": 2018,
|
| 19 |
+
"n_wrong": 508,
|
| 20 |
+
"per_domain_lodo": {
|
| 21 |
+
"전문시험(법·수학·생물)": 0.8334,
|
| 22 |
+
"상식·다단계추론": 0.6151,
|
| 23 |
+
"생물·의학": 0.7121,
|
| 24 |
+
"과학추론": 0.7531,
|
| 25 |
+
"재난·안전 지침": 0.6819
|
| 26 |
+
},
|
| 27 |
+
"per_domain_detail": {
|
| 28 |
+
"전문시험(법·수학·생물)": {
|
| 29 |
+
"auc": 0.8334,
|
| 30 |
+
"z": 9.07,
|
| 31 |
+
"n": 400,
|
| 32 |
+
"n_wrong": 80
|
| 33 |
+
},
|
| 34 |
+
"상식·다단계추론": {
|
| 35 |
+
"auc": 0.6151,
|
| 36 |
+
"z": 3.63,
|
| 37 |
+
"n": 278,
|
| 38 |
+
"n_wrong": 130
|
| 39 |
+
},
|
| 40 |
+
"생물·의학": {
|
| 41 |
+
"auc": 0.7121,
|
| 42 |
+
"z": 8.92,
|
| 43 |
+
"n": 917,
|
| 44 |
+
"n_wrong": 169
|
| 45 |
+
},
|
| 46 |
+
"과학추론": {
|
| 47 |
+
"auc": 0.7531,
|
| 48 |
+
"z": 3.82,
|
| 49 |
+
"n": 198,
|
| 50 |
+
"n_wrong": 20
|
| 51 |
+
},
|
| 52 |
+
"재난·안전 지침": {
|
| 53 |
+
"auc": 0.6819,
|
| 54 |
+
"z": 4.72,
|
| 55 |
+
"n": 225,
|
| 56 |
+
"n_wrong": 109
|
| 57 |
+
}
|
| 58 |
+
},
|
| 59 |
+
"review_budget": [
|
| 60 |
+
{
|
| 61 |
+
"frac": 0.05,
|
| 62 |
+
"k": 100,
|
| 63 |
+
"caught": 71,
|
| 64 |
+
"recall": 0.13976377952755906,
|
| 65 |
+
"precision": 0.71
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"frac": 0.1,
|
| 69 |
+
"k": 201,
|
| 70 |
+
"caught": 142,
|
| 71 |
+
"recall": 0.2795275590551181,
|
| 72 |
+
"precision": 0.7064676616915423
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"frac": 0.15,
|
| 76 |
+
"k": 302,
|
| 77 |
+
"caught": 195,
|
| 78 |
+
"recall": 0.3838582677165354,
|
| 79 |
+
"precision": 0.6456953642384106
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"frac": 0.2,
|
| 83 |
+
"k": 403,
|
| 84 |
+
"caught": 240,
|
| 85 |
+
"recall": 0.47244094488188976,
|
| 86 |
+
"precision": 0.5955334987593052
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"frac": 0.3,
|
| 90 |
+
"k": 605,
|
| 91 |
+
"caught": 315,
|
| 92 |
+
"recall": 0.6200787401574803,
|
| 93 |
+
"precision": 0.5206611570247934
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"frac": 0.4,
|
| 97 |
+
"k": 807,
|
| 98 |
+
"caught": 369,
|
| 99 |
+
"recall": 0.7263779527559056,
|
| 100 |
+
"precision": 0.45724907063197023
|
| 101 |
+
}
|
| 102 |
+
]
|
| 103 |
+
},
|
| 104 |
+
"note": "per_domain_lodo 는 leave-one-domain-out 값 = 처음 보는 분야의 정직한 기대치.",
|
| 105 |
+
"do_not_cite": "self_auc (in-sample). 성능은 cv_auc 로만 말한다."
|
| 106 |
+
}
|
ztc_judge.py
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# -*- coding: utf-8 -*-
|
| 2 |
+
"""ZTC-Judge-27B — 남의 모델이 쓴 답을 읽고 믿어도 되는지 판정한다.
|
| 3 |
+
|
| 4 |
+
글자를 한 개도 생성하지 않는다. 한 번 읽고, 곱셈 한 번 한다.
|
| 5 |
+
|
| 6 |
+
from ztc_judge import ZTCJudge
|
| 7 |
+
j = ZTCJudge.from_pretrained("FINAL-Bench/ZTC-Judge-27B")
|
| 8 |
+
j.score(question, answer, domain="과학추론")
|
| 9 |
+
"""
|
| 10 |
+
import json
|
| 11 |
+
import os
|
| 12 |
+
|
| 13 |
+
import numpy as np
|
| 14 |
+
import torch
|
| 15 |
+
from transformers import AutoModel, AutoTokenizer
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
class ZTCJudge:
|
| 19 |
+
def __init__(self, path, device="cuda", dtype=torch.bfloat16):
|
| 20 |
+
self.cfg = json.load(open(os.path.join(path, "ztc_config.json"), encoding="utf-8"))
|
| 21 |
+
p = dict(np.load(os.path.join(path, self.cfg["probe_file"]), allow_pickle=True))
|
| 22 |
+
self.w = p["w"].astype(np.float64)
|
| 23 |
+
self.mu = p["mu"].astype(np.float64)
|
| 24 |
+
self.sd = p["sd"].astype(np.float64)
|
| 25 |
+
self.per = self.cfg["measured"]["per_domain_lodo"]
|
| 26 |
+
self.tok = AutoTokenizer.from_pretrained(path)
|
| 27 |
+
if self.tok.pad_token is None:
|
| 28 |
+
self.tok.pad_token = self.tok.eos_token
|
| 29 |
+
self.model = AutoModel.from_pretrained(path, dtype=dtype,
|
| 30 |
+
low_cpu_mem_usage=True,
|
| 31 |
+
attn_implementation="sdpa").eval().to(device)
|
| 32 |
+
self.device = device
|
| 33 |
+
|
| 34 |
+
@classmethod
|
| 35 |
+
def from_pretrained(cls, repo, token=None, **kw):
|
| 36 |
+
from huggingface_hub import snapshot_download
|
| 37 |
+
return cls(snapshot_download(repo, token=token), **kw)
|
| 38 |
+
|
| 39 |
+
@torch.no_grad()
|
| 40 |
+
def _hidden(self, texts):
|
| 41 |
+
b = self.tok(texts, return_tensors="pt", padding=True, truncation=True,
|
| 42 |
+
max_length=self.cfg["max_length"])
|
| 43 |
+
ids = b["input_ids"].to(self.device)
|
| 44 |
+
am = b["attention_mask"].to(self.device)
|
| 45 |
+
h = self.model(input_ids=ids, attention_mask=am).last_hidden_state
|
| 46 |
+
idx = am.sum(1) - 1
|
| 47 |
+
return torch.stack([h[i, idx[i]] for i in range(len(texts))]).float().cpu().numpy()
|
| 48 |
+
|
| 49 |
+
def _verdict(self, s):
|
| 50 |
+
t = self.cfg["thresholds"]
|
| 51 |
+
return "검토 요망" if s <= t["review"] else ("이상 없음" if s >= t["clear"] else "판단 보류")
|
| 52 |
+
|
| 53 |
+
def score(self, question, answer, domain=None):
|
| 54 |
+
"""점수가 높을수록 '맞을 가능성이 높다'. 안 재본 분야에는 신뢰도를 주지 않는다."""
|
| 55 |
+
v = self._hidden([self.cfg["template"] % (question, answer)])[0].astype(np.float64)
|
| 56 |
+
s = float(((v - self.mu) / self.sd) @ self.w)
|
| 57 |
+
return {"score": round(s, 2), "verdict": self._verdict(s),
|
| 58 |
+
"domain_auc": self.per.get(domain), # None = 이 분야는 측정 안 됨
|
| 59 |
+
"overall_auc": self.cfg["measured"]["cv_auc"],
|
| 60 |
+
"generated_tokens": 0}
|
| 61 |
+
|
| 62 |
+
def score_batch(self, pairs, domain=None):
|
| 63 |
+
"""🔑 생성이 없고 읽기만이라 묶을수록 그대로 빨라진다 (VKIE)."""
|
| 64 |
+
V = self._hidden([self.cfg["template"] % (q, a) for q, a in pairs]).astype(np.float64)
|
| 65 |
+
S = ((V - self.mu) / self.sd) @ self.w
|
| 66 |
+
return [{"score": round(float(s), 2), "verdict": self._verdict(s),
|
| 67 |
+
"domain_auc": self.per.get(domain)} for s in S]
|