eval-team commited on
Commit
feb59eb
·
verified ·
1 Parent(s): f3bd1ab

Upload Gemma3ForCausalLM

Browse files
README.md CHANGED
@@ -1,276 +1,199 @@
1
- # EVAL-EStar-Base-v0.1
 
 
 
2
 
3
- - 소유자: 권용훈
4
- - 작성일: 2026-03-24
5
- - 상태: active
6
 
7
- ---
8
 
9
- ## 1. 모델 설명
10
 
11
- ### 아키텍처 / 파라미터
12
 
13
- | 항목 | 내용 |
14
- |------|------|
15
- | 베이스 모델 | Gemma-3-12B-IT |
16
- | 파라미터 수 | 12B |
17
- | 학습 방식 | Full Fine-Tuning (SFT) |
18
- | 출력 구조 | feedback → highlight → decision |
19
 
20
- ### Github 레포
21
 
22
- - **데이터 디베이트 레이블링**: [https://github.com/selectstar-ai/EVAL.Langgraph-Debate](https://github.com/selectstar-ai/EVAL.Langgraph-Debate)
23
 
24
- ### 버전 정보
25
 
26
- | 버전 | 설명 |
27
- |------|------|
28
- | v0.1 | 초기 Base 버전. K2-Feedback 기반 3단계 필터링 데이터(6,311개)로 학습 |
 
 
 
 
29
 
30
- ### 목적 (사용 사례)
31
 
32
- 한국어 루브릭 기반 평가를 안정적으로 수행하는 SLM 기반 evaluator. 금융·법률 도메인 RAG 파이프라인의 품질 평가에 특화되어 있으며, 다음 평가 축을 지원한다.
33
 
34
- - **Faithfulness**: 응답이 제공된 문서에 근거하는지 판단 (환각 진단)
35
- - **Context Relevancy**: 검색된 문서가 질의에 관련되는지 판단 (검색 품질)
36
- - **Response Relevancy**: 응답이 질의에 적절히 대응하는지 판단 (종합 응답 적합성)
37
 
38
- ### 모델 응용 가능성
39
 
40
- - 금융·법률 도메인의 RAG 평가로 확장 가능 (도메인별 루브릭 설계 필요)
41
- - 범용 루브릭 기반 LLM 출력 품질 평가 (Ko Feedback Bench에서 검증된 rubric following 능력)
42
- - 평가 파이프라인 자동화 시 frontier 모델 대비 비용 효율적 대안으로 활용
43
 
44
- ---
45
 
46
- ## 2. 모델 실행 방법
47
-
48
- ### 학습 코드 스니펫
49
-
50
- ```python
51
- from trl import SFTTrainer, SFTConfig
52
- from transformers import GemmaForCausalLM, AutoTokenizer
53
-
54
- model_name = "google/gemma-3-12b-it"
55
- model = GemmaForCausalLM.from_pretrained(model_name)
56
- tokenizer = AutoTokenizer.from_pretrained(model_name)
57
-
58
- sft_config = SFTConfig(
59
- output_dir="./eval-estar-base-v0.1",
60
- per_device_train_batch_size=1,
61
- gradient_accumulation_steps=8,
62
- learning_rate=1e-5,
63
- num_train_epochs=5,
64
- eval_strategy="epoch",
65
- save_strategy="epoch",
66
- load_best_model_at_end=True, # early stopping 기준: validation loss
67
- metric_for_best_model="eval_loss",
68
- greater_is_better=False,
69
- bf16=True,
70
- logging_steps=10,
71
- )
72
-
73
- trainer = SFTTrainer(
74
- model=model,
75
- tokenizer=tokenizer,
76
- train_dataset=train_dataset,
77
- eval_dataset=eval_dataset,
78
- args=sft_config,
79
- )
80
-
81
- trainer.train()
82
- ```
83
-
84
- > ※ 실제 학습 시 validation loss 기준 2 에폭에서 early stopping 적용
85
-
86
- ### 추론 코드 스니펫
87
-
88
- ```python
89
- from transformers import AutoModelForCausalLM, AutoTokenizer
90
-
91
- model_path = "datumo/E-Star-12B-v2-Base"
92
- model = AutoModelForCausalLM.from_pretrained(model_path, torch_dtype="bfloat16", device_map="auto")
93
- tokenizer = AutoTokenizer.from_pretrained(model_path)
94
-
95
- # ── System Prompt ──
96
- system_prompt = """You are a rubric evaluator.
97
- Your task is to evaluate a response strictly and only according to the provided pass criteria and scoring rubric.
98
- In your output, return the final evaluation (the three output tags: <feedback>, <highlight>, and <decision>).
99
-
100
- # Evaluation Procedure (must follow all steps):
101
-
102
- 1. First, carefully read the Data to Evaluate, the pass criteria, and the scoring rubric to fully understand the requirements.
103
- 2. Evaluate the response only against the given criteria: do not introduce external standards, do not reward style unless the rubric explicitly allows it, and judge by absolute rubric definitions rather than relative comparisons.
104
- 3. Re-check fine-grained details in the response and the rubric, ensuring any tags (if present) are correctly mapped to the pass criteria and that small deviations are not overlooked.
105
- 4. Write criterion-focused feedback that explicitly references the rubric, quoting exact words or phrases from the response when they are decisive, and clearly stating which criteria are satisfied and which are violated.
106
- 5. Finally, extract the key verbatim spans that most influenced your judgment and assign the final score according to the scoring rubric.
107
- """
108
-
109
- # ── User Prompt 예시 (Reasoning / Problem Solving) ──
110
- user_prompt = """You MUST write ALL output (<feedback>, <highlight>, <decision>) in the SAME language as the input question and response being evaluated. If the input is in Korean, your entire output MUST be in Korean.
111
- # Output Format:
112
- <feedback>
113
- Write detailed feedback (reasons) that strictly evaluates the quality of the response using only the given scoring rubric. Do not explicitly state the score in a sentence (e.g., "Therefore, the score is …").
114
- </feedback>
115
- <highlight>
116
- List of words or phrases that you believe are the most important in determining the score.
117
- </highlight>
118
- <decision>
119
- Provide the final integer score assigned based on the scoring rubric.
120
- </decision>
121
-
122
- # Data to Evaluate
123
-
124
- ### Problem
125
- 한 공장에서 하루에 120개의 제품을 생산한다. 불량률이 5%일 때, 일주일(7일) 동안 생산되는 정상 제품의 수는?
126
-
127
- ### Model Response
128
- 하루 생산량: 120개
129
- 불량률: 5% → 불량품: 120 × 0.05 = 6개
130
- 하루 정상 제품: 120 - 6 = 114개
131
- 일주일 정상 제품: 114 × 7 = 798개
132
-
133
- ### Optional Ground Truth
134
- 798개
135
-
136
- # Rubric
137
- Evaluate whether the model correctly solves the problem and provides reasoning that is logically consistent with the final answer. Prioritize correctness of the conclusion, then soundness of the reasoning.
138
-
139
- Score 1: The final answer is wrong and the reasoning is invalid, irrelevant, or missing.
140
- Score 2: The response shows limited progress but contains major reasoning flaws leading to an incorrect or unreliable answer.
141
- Score 3: The response demonstrates partial reasoning ability but is incomplete, contains mistakes, or reaches an uncertain result.
142
- Score 4: The response is mostly correct with generally sound reasoning, though minor errors or gaps may remain.
143
- Score 5: The response reaches the correct answer through clear, consistent, and logically valid reasoning appropriate to the problem."""
144
-
145
- # ── Inference ──
146
- messages = [
147
- {"role": "system", "content": system_prompt},
148
- {"role": "user", "content": user_prompt},
149
- ]
150
-
151
- input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
152
-
153
- outputs = model.generate(
154
- input_ids,
155
- max_new_tokens=2048,
156
- temperature=0.0,
157
- do_sample=False,
158
- )
159
-
160
- response = tokenizer.decode(outputs[0][input_ids.shape[-1]:], skip_special_tokens=True)
161
- print(response)
162
- ```
163
 
164
- ---
165
 
166
- ## 3. 학습 데이터셋
167
 
168
- ### HuggingFace 데이터셋
169
 
170
- 🤗 **[datumo/E-Star-Train-6K](https://huggingface.co/datasets/datumo/E-Star-Train-6K)**
171
 
172
- | 항목 | 내용 |
173
- |------|------|
174
- | 시드 데이터 | K2-Feedback (HAERAEHUB, 2024) — 99.7K |
175
- | 최종 학습 데이터 | 6,311개 (3단계 필터링 후) |
176
 
177
- ### 필터링 파이프라인 요약
178
 
179
- | 단계 | 규모 변화 | 방법 |
180
- |------|-----------|------|
181
- | Stage 1 | 99.7K → 26K | Qwen3-30B-A3B / Qwen3-Next-80B-A3B 간 초기 합의 |
182
- | Stage 2 | 26K → 8K | Gemma 베이스 모델 기준 일치/불일치 균형화 |
183
- | Stage 3 | 8K → 6K | GPT-5.2 단일 평가 + 소형 frontier debate 교차 검증 |
184
 
185
- ### 평가 벤치마크
186
 
187
- - 🤗 [datumo/Feedback-Bench](https://huggingface.co/datasets/datumo/Feedback-Bench) Rubric following (한국어 / 영어)
188
- - 🤗 [datumo/Rag-Quality-Bench](https://huggingface.co/datasets/datumo/Rag-Quality-Bench) — Domain adaptation (금융·법률)
189
 
190
- ---
191
 
192
- ## 4. 학습 설정
193
 
194
- ### 주요 학습 파라미터
195
 
196
- ```yaml
197
- # SFT Config
198
- learning_rate: 1e-5
199
- num_train_epochs: 5 (early stopping at epoch 2)
200
- per_device_train_batch_size: 1
201
- gradient_accumulation_steps: 8
202
- bf16: true
203
- eval_strategy: epoch
204
- metric_for_best_model: eval_loss
205
- load_best_model_at_end: true
206
 
207
- # Fine-tuning
208
- method: Full Fine-Tuning (no LoRA)
209
- framework: TRL SFTTrainer (von Werra et al., 2020)
210
- ```
211
 
212
- ### 사용 GPU / 학습 시간
213
 
214
- | 항목 | 내용 |
215
- |------|------|
216
- | GPU | OOO |
217
- | 학습 시간 | OOO |
218
 
219
- ---
220
 
221
- ## 5. 평가 결과
222
-
223
- ### 5.1 Feedback Bench (영어, Rubric Following)
224
-
225
- | Type | Models | Pearson | Kendall τ | Spearman |
226
- |------|--------|---------|-----------|----------|
227
- | Frontier | GPT-5.2 | 0.916 | 0.865 | 0.911 |
228
- | Frontier | Sonnet-4.6 | 0.840 | 0.776 | 0.847 |
229
- | Instruct SLM | Gemma-3-12B-IT | 0.810 | 0.725 | 0.794 |
230
- | Instruct SLM | oss-20b | 0.844 | 0.762 | 0.839 |
231
- | Evaluator LM | Prometheus-8x7B-v2.0 | 0.823 | 0.736 | 0.806 |
232
- | Evaluator LM | GLIDER 3.8B | 0.678 | 0.595 | 0.688 |
233
- | **Ours** | **E-Star-12B-Base** | **0.856** | **0.778** | **0.847** |
234
-
235
- ### 5.2 Ko Feedback Bench (한국어, Rubric Following)
236
-
237
- | Type | Models | Pearson | Kendall τ | Spearman |
238
- |------|--------|---------|-----------|----------|
239
- | Frontier | GPT-5.2 | 0.929 | 0.886 | 0.925 |
240
- | Frontier | Sonnet-4.6 | 0.820 | 0.758 | 0.833 |
241
- | Instruct SLM | Gemma-3-12B-IT | 0.653 | 0.593 | 0.661 |
242
- | Instruct SLM | oss-20b | 0.778 | 0.704 | 0.779 |
243
- | Evaluator LM | Prometheus-8x7B-v2.0 | 0.377 | 0.441 | 0.501 |
244
- | Evaluator LM | GLIDER 3.8B | 0.523 | 0.487 | 0.563 |
245
- | **Ours** | **E-Star-12B-Base** | **0.826** | **0.754** | **0.819** |
246
-
247
- ### 5.3 RAG Quality Bench (금융·법률, Domain Adaptation)
248
-
249
- | Models | LAW(CR) | LAW(FF) | LAW(RR) | FIN(CR) | FIN(FF) | FIN(RR) | Average |
250
- |--------|---------|---------|---------|---------|---------|---------|---------|
251
- | GPT-5.2 | 0.846 | 0.785 | 0.941 | 0.882 | 0.740 | 0.970 | 0.861 |
252
- | Sonnet-4.6 | 0.910 | 0.786 | 0.872 | 0.932 | 0.845 | 0.925 | 0.878 |
253
- | Gemma-3-12B-IT | 0.620 | 0.742 | 0.742 | 0.830 | 0.713 | 0.821 | 0.745 |
254
- | oss-20b | 0.846 | 0.722 | 0.870 | 0.793 | 0.752 | 0.900 | 0.813 |
255
- | Prometheus-8x7B-v2.0 | 0.392 | 0.477 | 0.772 | 0.386 | 0.240 | 0.806 | 0.512 |
256
- | GLIDER 3.8B | 0.657 | 0.670 | 0.680 | 0.432 | 0.415 | 0.548 | 0.567 |
257
- | **E-Star-12B-Base** | **0.853** | **0.730** | **0.816** | **0.835** | **0.720** | **0.880** | **0.806** |
258
-
259
- > CR = Context Relevancy, FF = Faithfulness, RR = Response Relevancy
260
 
261
- ---
262
 
263
- ## 6. 한계
264
 
265
- - Ko Feedback Bench는 기계번역 기반으로 구축되었으므로, 번역 노이즈가 평가 성능 측정에 영향을 줄 수 있음
266
- - 학습 데이터와 벤치마크 레이블이 동일한 debate 기반 절차로 구축되었으므로, 절대적 평가 품질보다는 합의 기반 레이블링 기준과의 정렬 정도를 반영할 수 있음 (도메인 전문가 human evaluation 미실시)
267
- - Reference-free 설정으로 학습 및 평가되었으므로, reference 포함 환경에서의 성능은 별도 검증 필요
268
- - 12B 규모 SLM 특성상 frontier 모델 대비 복잡한 루브릭 해석 능력에 한계가 있을 수 있음
269
- - RAG 평가 시 입력 문서 수 증가에 따른 성능 변화는 미검증
270
 
271
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
272
 
273
- ## 7. 라이선스
274
 
275
- - 베이스 모델: Gemma License (Google)
276
- - 학습 데이터: K2-Feedback (HAERAEHUB, 2024) 라이선스 정책에 따름
 
1
+ ---
2
+ library_name: transformers
3
+ tags: []
4
+ ---
5
 
6
+ # Model Card for Model ID
 
 
7
 
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
 
 
10
 
 
11
 
12
+ ## Model Details
 
 
 
 
 
13
 
14
+ ### Model Description
15
 
16
+ <!-- Provide a longer summary of what this model is. -->
17
 
18
+ This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
19
 
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
 
28
+ ### Model Sources [optional]
29
 
30
+ <!-- Provide the basic links for the model. -->
31
 
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
 
36
+ ## Uses
37
 
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
 
 
39
 
40
+ ### Direct Use
41
 
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
43
 
44
+ [More Information Needed]
45
 
46
+ ### Downstream Use [optional]
47
 
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
 
50
+ [More Information Needed]
51
 
52
+ ### Out-of-Scope Use
 
 
 
53
 
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
 
56
+ [More Information Needed]
 
 
 
 
57
 
58
+ ## Bias, Risks, and Limitations
59
 
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
 
61
 
62
+ [More Information Needed]
63
 
64
+ ### Recommendations
65
 
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
 
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
 
 
 
 
 
 
 
 
 
69
 
70
+ ## How to Get Started with the Model
 
 
 
71
 
72
+ Use the code below to get started with the model.
73
 
74
+ [More Information Needed]
 
 
 
75
 
76
+ ## Training Details
77
 
78
+ ### Training Data
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
79
 
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
 
82
+ [More Information Needed]
83
 
84
+ ### Training Procedure
 
 
 
 
85
 
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
 
197
+ ## Model Card Contact
198
 
199
+ [More Information Needed]
 
config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "dtype": "float32",
11
+ "eos_token_id": 1,
12
+ "final_logit_softcapping": null,
13
+ "head_dim": 256,
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 3840,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 15360,
18
+ "layer_types": [
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "full_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention"
67
+ ],
68
+ "max_position_embeddings": 131072,
69
+ "model_type": "gemma3_text",
70
+ "num_attention_heads": 16,
71
+ "num_hidden_layers": 48,
72
+ "num_key_value_heads": 8,
73
+ "pad_token_id": 0,
74
+ "query_pre_attn_scalar": 256,
75
+ "rms_norm_eps": 1e-06,
76
+ "rope_local_base_freq": 10000.0,
77
+ "rope_scaling": {
78
+ "factor": 8.0,
79
+ "rope_type": "linear"
80
+ },
81
+ "rope_theta": 1000000.0,
82
+ "sliding_window": 1024,
83
+ "transformers_version": "4.57.6",
84
+ "use_bidirectional_attention": false,
85
+ "use_cache": true,
86
+ "vocab_size": 262208
87
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 1,
8
+ 106
9
+ ],
10
+ "pad_token_id": 0,
11
+ "top_k": 64,
12
+ "top_p": 0.95,
13
+ "transformers_version": "4.57.6"
14
+ }
model-00001-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b9fc7b75bb0a9e50db2868561053d55a567a967a7028335f15e284321b4ff3b
3
+ size 4987027160
model-00002-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b8879dc1923b978340512056cfd3fd3bf68fbb89d02981a79207d109f8341bf
3
+ size 4844748760
model-00003-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79aacec5fd4003d07e3b12f3154b10d3eeb068a4e2bb20a1c04fa1f1fe7e5e34
3
+ size 4954908672
model-00004-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2b577a0c401dcd67a41d8d4012580a878ef34e07cda85e4211d88e34bb94929
3
+ size 4907663504
model-00005-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcba81d02f5e834cf9d6f7568ceb14a08345e2bb09f183de9c131fd101622a26
3
+ size 4954908728
model-00006-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f21ae16a635945bb542aa9489c4015cd7ae259ff40c1ddef16d222d953fe6794
3
+ size 4907663504
model-00007-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c69aa502df387c761c8ce1a9b4e77894a2f0626732e19e1875a3720ff417c615
3
+ size 4954908728
model-00008-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a15b6cacc95fe4c38e0706dd56f0c7ed738d28c1bea09af97fd79688e7287ef
3
+ size 4907663504
model-00009-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baa914441a191a65e6cc90d432c1b893a884b3a02eed441c10b68af2ff76936f
3
+ size 4954908728
model-00010-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c17610971c940d364087854652d267663ec51f7b92f2e5e72bc4a31d248fdb68
3
+ size 2689807872
model.safetensors.index.json ADDED
@@ -0,0 +1,634 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 11766034176,
4
+ "total_size": 47064136704
5
+ },
6
+ "weight_map": {
7
+ "model.embed_tokens.weight": "model-00001-of-00010.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00010.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00010.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00010.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00010.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00010.safetensors",
13
+ "model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00010.safetensors",
14
+ "model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00010.safetensors",
15
+ "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00010.safetensors",
16
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00010.safetensors",
17
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00010.safetensors",
18
+ "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00010.safetensors",
19
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00010.safetensors",
20
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00010.safetensors",
21
+ "model.layers.1.input_layernorm.weight": "model-00002-of-00010.safetensors",
22
+ "model.layers.1.mlp.down_proj.weight": "model-00002-of-00010.safetensors",
23
+ "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00010.safetensors",
24
+ "model.layers.1.mlp.up_proj.weight": "model-00002-of-00010.safetensors",
25
+ "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00010.safetensors",
26
+ "model.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
27
+ "model.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
28
+ "model.layers.1.self_attn.k_norm.weight": "model-00002-of-00010.safetensors",
29
+ "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00010.safetensors",
30
+ "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00010.safetensors",
31
+ "model.layers.1.self_attn.q_norm.weight": "model-00002-of-00010.safetensors",
32
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00010.safetensors",
33
+ "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00010.safetensors",
34
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00010.safetensors",
35
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00010.safetensors",
36
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00010.safetensors",
37
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00010.safetensors",
38
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00010.safetensors",
39
+ "model.layers.10.post_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
40
+ "model.layers.10.pre_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
41
+ "model.layers.10.self_attn.k_norm.weight": "model-00003-of-00010.safetensors",
42
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00010.safetensors",
43
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00010.safetensors",
44
+ "model.layers.10.self_attn.q_norm.weight": "model-00003-of-00010.safetensors",
45
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00010.safetensors",
46
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00010.safetensors",
47
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00010.safetensors",
48
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00010.safetensors",
49
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00010.safetensors",
50
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00010.safetensors",
51
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00010.safetensors",
52
+ "model.layers.11.post_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
53
+ "model.layers.11.pre_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
54
+ "model.layers.11.self_attn.k_norm.weight": "model-00003-of-00010.safetensors",
55
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00010.safetensors",
56
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00010.safetensors",
57
+ "model.layers.11.self_attn.q_norm.weight": "model-00003-of-00010.safetensors",
58
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00010.safetensors",
59
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00010.safetensors",
60
+ "model.layers.12.input_layernorm.weight": "model-00004-of-00010.safetensors",
61
+ "model.layers.12.mlp.down_proj.weight": "model-00004-of-00010.safetensors",
62
+ "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00010.safetensors",
63
+ "model.layers.12.mlp.up_proj.weight": "model-00004-of-00010.safetensors",
64
+ "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00010.safetensors",
65
+ "model.layers.12.post_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
66
+ "model.layers.12.pre_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
67
+ "model.layers.12.self_attn.k_norm.weight": "model-00004-of-00010.safetensors",
68
+ "model.layers.12.self_attn.k_proj.weight": "model-00004-of-00010.safetensors",
69
+ "model.layers.12.self_attn.o_proj.weight": "model-00004-of-00010.safetensors",
70
+ "model.layers.12.self_attn.q_norm.weight": "model-00004-of-00010.safetensors",
71
+ "model.layers.12.self_attn.q_proj.weight": "model-00004-of-00010.safetensors",
72
+ "model.layers.12.self_attn.v_proj.weight": "model-00004-of-00010.safetensors",
73
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00010.safetensors",
74
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00010.safetensors",
75
+ "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00010.safetensors",
76
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00010.safetensors",
77
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00010.safetensors",
78
+ "model.layers.13.post_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
79
+ "model.layers.13.pre_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
80
+ "model.layers.13.self_attn.k_norm.weight": "model-00004-of-00010.safetensors",
81
+ "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00010.safetensors",
82
+ "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00010.safetensors",
83
+ "model.layers.13.self_attn.q_norm.weight": "model-00004-of-00010.safetensors",
84
+ "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00010.safetensors",
85
+ "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00010.safetensors",
86
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00010.safetensors",
87
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00010.safetensors",
88
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00010.safetensors",
89
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00010.safetensors",
90
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00010.safetensors",
91
+ "model.layers.14.post_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
92
+ "model.layers.14.pre_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
93
+ "model.layers.14.self_attn.k_norm.weight": "model-00004-of-00010.safetensors",
94
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00010.safetensors",
95
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00010.safetensors",
96
+ "model.layers.14.self_attn.q_norm.weight": "model-00004-of-00010.safetensors",
97
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00010.safetensors",
98
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00010.safetensors",
99
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00010.safetensors",
100
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00010.safetensors",
101
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00010.safetensors",
102
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00010.safetensors",
103
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00010.safetensors",
104
+ "model.layers.15.post_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
105
+ "model.layers.15.pre_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
106
+ "model.layers.15.self_attn.k_norm.weight": "model-00004-of-00010.safetensors",
107
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00010.safetensors",
108
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00010.safetensors",
109
+ "model.layers.15.self_attn.q_norm.weight": "model-00004-of-00010.safetensors",
110
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00010.safetensors",
111
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00010.safetensors",
112
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00010.safetensors",
113
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00010.safetensors",
114
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00010.safetensors",
115
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00010.safetensors",
116
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00010.safetensors",
117
+ "model.layers.16.post_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
118
+ "model.layers.16.pre_feedforward_layernorm.weight": "model-00004-of-00010.safetensors",
119
+ "model.layers.16.self_attn.k_norm.weight": "model-00004-of-00010.safetensors",
120
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00010.safetensors",
121
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00010.safetensors",
122
+ "model.layers.16.self_attn.q_norm.weight": "model-00004-of-00010.safetensors",
123
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00010.safetensors",
124
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00010.safetensors",
125
+ "model.layers.17.input_layernorm.weight": "model-00005-of-00010.safetensors",
126
+ "model.layers.17.mlp.down_proj.weight": "model-00005-of-00010.safetensors",
127
+ "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00010.safetensors",
128
+ "model.layers.17.mlp.up_proj.weight": "model-00005-of-00010.safetensors",
129
+ "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00010.safetensors",
130
+ "model.layers.17.post_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
131
+ "model.layers.17.pre_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
132
+ "model.layers.17.self_attn.k_norm.weight": "model-00004-of-00010.safetensors",
133
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00010.safetensors",
134
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00010.safetensors",
135
+ "model.layers.17.self_attn.q_norm.weight": "model-00004-of-00010.safetensors",
136
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00010.safetensors",
137
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00010.safetensors",
138
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00010.safetensors",
139
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00010.safetensors",
140
+ "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00010.safetensors",
141
+ "model.layers.18.mlp.up_proj.weight": "model-00005-of-00010.safetensors",
142
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00010.safetensors",
143
+ "model.layers.18.post_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
144
+ "model.layers.18.pre_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
145
+ "model.layers.18.self_attn.k_norm.weight": "model-00005-of-00010.safetensors",
146
+ "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00010.safetensors",
147
+ "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00010.safetensors",
148
+ "model.layers.18.self_attn.q_norm.weight": "model-00005-of-00010.safetensors",
149
+ "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00010.safetensors",
150
+ "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00010.safetensors",
151
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00010.safetensors",
152
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00010.safetensors",
153
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00010.safetensors",
154
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00010.safetensors",
155
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00010.safetensors",
156
+ "model.layers.19.post_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
157
+ "model.layers.19.pre_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
158
+ "model.layers.19.self_attn.k_norm.weight": "model-00005-of-00010.safetensors",
159
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00010.safetensors",
160
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00010.safetensors",
161
+ "model.layers.19.self_attn.q_norm.weight": "model-00005-of-00010.safetensors",
162
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00010.safetensors",
163
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00010.safetensors",
164
+ "model.layers.2.input_layernorm.weight": "model-00002-of-00010.safetensors",
165
+ "model.layers.2.mlp.down_proj.weight": "model-00002-of-00010.safetensors",
166
+ "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00010.safetensors",
167
+ "model.layers.2.mlp.up_proj.weight": "model-00002-of-00010.safetensors",
168
+ "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00010.safetensors",
169
+ "model.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
170
+ "model.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
171
+ "model.layers.2.self_attn.k_norm.weight": "model-00002-of-00010.safetensors",
172
+ "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00010.safetensors",
173
+ "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00010.safetensors",
174
+ "model.layers.2.self_attn.q_norm.weight": "model-00002-of-00010.safetensors",
175
+ "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00010.safetensors",
176
+ "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00010.safetensors",
177
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00010.safetensors",
178
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00010.safetensors",
179
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00010.safetensors",
180
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00010.safetensors",
181
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00010.safetensors",
182
+ "model.layers.20.post_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
183
+ "model.layers.20.pre_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
184
+ "model.layers.20.self_attn.k_norm.weight": "model-00005-of-00010.safetensors",
185
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00010.safetensors",
186
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00010.safetensors",
187
+ "model.layers.20.self_attn.q_norm.weight": "model-00005-of-00010.safetensors",
188
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00010.safetensors",
189
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00010.safetensors",
190
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00010.safetensors",
191
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00010.safetensors",
192
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00010.safetensors",
193
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00010.safetensors",
194
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00010.safetensors",
195
+ "model.layers.21.post_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
196
+ "model.layers.21.pre_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
197
+ "model.layers.21.self_attn.k_norm.weight": "model-00005-of-00010.safetensors",
198
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00010.safetensors",
199
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00010.safetensors",
200
+ "model.layers.21.self_attn.q_norm.weight": "model-00005-of-00010.safetensors",
201
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00010.safetensors",
202
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00010.safetensors",
203
+ "model.layers.22.input_layernorm.weight": "model-00005-of-00010.safetensors",
204
+ "model.layers.22.mlp.down_proj.weight": "model-00005-of-00010.safetensors",
205
+ "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00010.safetensors",
206
+ "model.layers.22.mlp.up_proj.weight": "model-00005-of-00010.safetensors",
207
+ "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00010.safetensors",
208
+ "model.layers.22.post_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
209
+ "model.layers.22.pre_feedforward_layernorm.weight": "model-00005-of-00010.safetensors",
210
+ "model.layers.22.self_attn.k_norm.weight": "model-00005-of-00010.safetensors",
211
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00010.safetensors",
212
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00010.safetensors",
213
+ "model.layers.22.self_attn.q_norm.weight": "model-00005-of-00010.safetensors",
214
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00010.safetensors",
215
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00010.safetensors",
216
+ "model.layers.23.input_layernorm.weight": "model-00006-of-00010.safetensors",
217
+ "model.layers.23.mlp.down_proj.weight": "model-00006-of-00010.safetensors",
218
+ "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00010.safetensors",
219
+ "model.layers.23.mlp.up_proj.weight": "model-00006-of-00010.safetensors",
220
+ "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00010.safetensors",
221
+ "model.layers.23.post_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
222
+ "model.layers.23.pre_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
223
+ "model.layers.23.self_attn.k_norm.weight": "model-00006-of-00010.safetensors",
224
+ "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00010.safetensors",
225
+ "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00010.safetensors",
226
+ "model.layers.23.self_attn.q_norm.weight": "model-00006-of-00010.safetensors",
227
+ "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00010.safetensors",
228
+ "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00010.safetensors",
229
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00010.safetensors",
230
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00010.safetensors",
231
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00010.safetensors",
232
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00010.safetensors",
233
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00010.safetensors",
234
+ "model.layers.24.post_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
235
+ "model.layers.24.pre_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
236
+ "model.layers.24.self_attn.k_norm.weight": "model-00006-of-00010.safetensors",
237
+ "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00010.safetensors",
238
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00010.safetensors",
239
+ "model.layers.24.self_attn.q_norm.weight": "model-00006-of-00010.safetensors",
240
+ "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00010.safetensors",
241
+ "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00010.safetensors",
242
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00010.safetensors",
243
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00010.safetensors",
244
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00010.safetensors",
245
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00010.safetensors",
246
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00010.safetensors",
247
+ "model.layers.25.post_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
248
+ "model.layers.25.pre_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
249
+ "model.layers.25.self_attn.k_norm.weight": "model-00006-of-00010.safetensors",
250
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00010.safetensors",
251
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00010.safetensors",
252
+ "model.layers.25.self_attn.q_norm.weight": "model-00006-of-00010.safetensors",
253
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00010.safetensors",
254
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00010.safetensors",
255
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00010.safetensors",
256
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00010.safetensors",
257
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00010.safetensors",
258
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00010.safetensors",
259
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00010.safetensors",
260
+ "model.layers.26.post_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
261
+ "model.layers.26.pre_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
262
+ "model.layers.26.self_attn.k_norm.weight": "model-00006-of-00010.safetensors",
263
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00010.safetensors",
264
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00010.safetensors",
265
+ "model.layers.26.self_attn.q_norm.weight": "model-00006-of-00010.safetensors",
266
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00010.safetensors",
267
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00010.safetensors",
268
+ "model.layers.27.input_layernorm.weight": "model-00006-of-00010.safetensors",
269
+ "model.layers.27.mlp.down_proj.weight": "model-00006-of-00010.safetensors",
270
+ "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00010.safetensors",
271
+ "model.layers.27.mlp.up_proj.weight": "model-00006-of-00010.safetensors",
272
+ "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00010.safetensors",
273
+ "model.layers.27.post_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
274
+ "model.layers.27.pre_feedforward_layernorm.weight": "model-00006-of-00010.safetensors",
275
+ "model.layers.27.self_attn.k_norm.weight": "model-00006-of-00010.safetensors",
276
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00010.safetensors",
277
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00010.safetensors",
278
+ "model.layers.27.self_attn.q_norm.weight": "model-00006-of-00010.safetensors",
279
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00010.safetensors",
280
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00010.safetensors",
281
+ "model.layers.28.input_layernorm.weight": "model-00007-of-00010.safetensors",
282
+ "model.layers.28.mlp.down_proj.weight": "model-00007-of-00010.safetensors",
283
+ "model.layers.28.mlp.gate_proj.weight": "model-00006-of-00010.safetensors",
284
+ "model.layers.28.mlp.up_proj.weight": "model-00007-of-00010.safetensors",
285
+ "model.layers.28.post_attention_layernorm.weight": "model-00007-of-00010.safetensors",
286
+ "model.layers.28.post_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
287
+ "model.layers.28.pre_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
288
+ "model.layers.28.self_attn.k_norm.weight": "model-00006-of-00010.safetensors",
289
+ "model.layers.28.self_attn.k_proj.weight": "model-00006-of-00010.safetensors",
290
+ "model.layers.28.self_attn.o_proj.weight": "model-00006-of-00010.safetensors",
291
+ "model.layers.28.self_attn.q_norm.weight": "model-00006-of-00010.safetensors",
292
+ "model.layers.28.self_attn.q_proj.weight": "model-00006-of-00010.safetensors",
293
+ "model.layers.28.self_attn.v_proj.weight": "model-00006-of-00010.safetensors",
294
+ "model.layers.29.input_layernorm.weight": "model-00007-of-00010.safetensors",
295
+ "model.layers.29.mlp.down_proj.weight": "model-00007-of-00010.safetensors",
296
+ "model.layers.29.mlp.gate_proj.weight": "model-00007-of-00010.safetensors",
297
+ "model.layers.29.mlp.up_proj.weight": "model-00007-of-00010.safetensors",
298
+ "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00010.safetensors",
299
+ "model.layers.29.post_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
300
+ "model.layers.29.pre_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
301
+ "model.layers.29.self_attn.k_norm.weight": "model-00007-of-00010.safetensors",
302
+ "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00010.safetensors",
303
+ "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00010.safetensors",
304
+ "model.layers.29.self_attn.q_norm.weight": "model-00007-of-00010.safetensors",
305
+ "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00010.safetensors",
306
+ "model.layers.29.self_attn.v_proj.weight": "model-00007-of-00010.safetensors",
307
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00010.safetensors",
308
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00010.safetensors",
309
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00010.safetensors",
310
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00010.safetensors",
311
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00010.safetensors",
312
+ "model.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
313
+ "model.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
314
+ "model.layers.3.self_attn.k_norm.weight": "model-00002-of-00010.safetensors",
315
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00010.safetensors",
316
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00010.safetensors",
317
+ "model.layers.3.self_attn.q_norm.weight": "model-00002-of-00010.safetensors",
318
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00010.safetensors",
319
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00010.safetensors",
320
+ "model.layers.30.input_layernorm.weight": "model-00007-of-00010.safetensors",
321
+ "model.layers.30.mlp.down_proj.weight": "model-00007-of-00010.safetensors",
322
+ "model.layers.30.mlp.gate_proj.weight": "model-00007-of-00010.safetensors",
323
+ "model.layers.30.mlp.up_proj.weight": "model-00007-of-00010.safetensors",
324
+ "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00010.safetensors",
325
+ "model.layers.30.post_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
326
+ "model.layers.30.pre_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
327
+ "model.layers.30.self_attn.k_norm.weight": "model-00007-of-00010.safetensors",
328
+ "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00010.safetensors",
329
+ "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00010.safetensors",
330
+ "model.layers.30.self_attn.q_norm.weight": "model-00007-of-00010.safetensors",
331
+ "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00010.safetensors",
332
+ "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00010.safetensors",
333
+ "model.layers.31.input_layernorm.weight": "model-00007-of-00010.safetensors",
334
+ "model.layers.31.mlp.down_proj.weight": "model-00007-of-00010.safetensors",
335
+ "model.layers.31.mlp.gate_proj.weight": "model-00007-of-00010.safetensors",
336
+ "model.layers.31.mlp.up_proj.weight": "model-00007-of-00010.safetensors",
337
+ "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00010.safetensors",
338
+ "model.layers.31.post_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
339
+ "model.layers.31.pre_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
340
+ "model.layers.31.self_attn.k_norm.weight": "model-00007-of-00010.safetensors",
341
+ "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00010.safetensors",
342
+ "model.layers.31.self_attn.o_proj.weight": "model-00007-of-00010.safetensors",
343
+ "model.layers.31.self_attn.q_norm.weight": "model-00007-of-00010.safetensors",
344
+ "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00010.safetensors",
345
+ "model.layers.31.self_attn.v_proj.weight": "model-00007-of-00010.safetensors",
346
+ "model.layers.32.input_layernorm.weight": "model-00007-of-00010.safetensors",
347
+ "model.layers.32.mlp.down_proj.weight": "model-00007-of-00010.safetensors",
348
+ "model.layers.32.mlp.gate_proj.weight": "model-00007-of-00010.safetensors",
349
+ "model.layers.32.mlp.up_proj.weight": "model-00007-of-00010.safetensors",
350
+ "model.layers.32.post_attention_layernorm.weight": "model-00007-of-00010.safetensors",
351
+ "model.layers.32.post_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
352
+ "model.layers.32.pre_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
353
+ "model.layers.32.self_attn.k_norm.weight": "model-00007-of-00010.safetensors",
354
+ "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00010.safetensors",
355
+ "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00010.safetensors",
356
+ "model.layers.32.self_attn.q_norm.weight": "model-00007-of-00010.safetensors",
357
+ "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00010.safetensors",
358
+ "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00010.safetensors",
359
+ "model.layers.33.input_layernorm.weight": "model-00007-of-00010.safetensors",
360
+ "model.layers.33.mlp.down_proj.weight": "model-00007-of-00010.safetensors",
361
+ "model.layers.33.mlp.gate_proj.weight": "model-00007-of-00010.safetensors",
362
+ "model.layers.33.mlp.up_proj.weight": "model-00007-of-00010.safetensors",
363
+ "model.layers.33.post_attention_layernorm.weight": "model-00007-of-00010.safetensors",
364
+ "model.layers.33.post_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
365
+ "model.layers.33.pre_feedforward_layernorm.weight": "model-00007-of-00010.safetensors",
366
+ "model.layers.33.self_attn.k_norm.weight": "model-00007-of-00010.safetensors",
367
+ "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00010.safetensors",
368
+ "model.layers.33.self_attn.o_proj.weight": "model-00007-of-00010.safetensors",
369
+ "model.layers.33.self_attn.q_norm.weight": "model-00007-of-00010.safetensors",
370
+ "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00010.safetensors",
371
+ "model.layers.33.self_attn.v_proj.weight": "model-00007-of-00010.safetensors",
372
+ "model.layers.34.input_layernorm.weight": "model-00008-of-00010.safetensors",
373
+ "model.layers.34.mlp.down_proj.weight": "model-00008-of-00010.safetensors",
374
+ "model.layers.34.mlp.gate_proj.weight": "model-00008-of-00010.safetensors",
375
+ "model.layers.34.mlp.up_proj.weight": "model-00008-of-00010.safetensors",
376
+ "model.layers.34.post_attention_layernorm.weight": "model-00008-of-00010.safetensors",
377
+ "model.layers.34.post_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
378
+ "model.layers.34.pre_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
379
+ "model.layers.34.self_attn.k_norm.weight": "model-00008-of-00010.safetensors",
380
+ "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00010.safetensors",
381
+ "model.layers.34.self_attn.o_proj.weight": "model-00008-of-00010.safetensors",
382
+ "model.layers.34.self_attn.q_norm.weight": "model-00008-of-00010.safetensors",
383
+ "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00010.safetensors",
384
+ "model.layers.34.self_attn.v_proj.weight": "model-00008-of-00010.safetensors",
385
+ "model.layers.35.input_layernorm.weight": "model-00008-of-00010.safetensors",
386
+ "model.layers.35.mlp.down_proj.weight": "model-00008-of-00010.safetensors",
387
+ "model.layers.35.mlp.gate_proj.weight": "model-00008-of-00010.safetensors",
388
+ "model.layers.35.mlp.up_proj.weight": "model-00008-of-00010.safetensors",
389
+ "model.layers.35.post_attention_layernorm.weight": "model-00008-of-00010.safetensors",
390
+ "model.layers.35.post_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
391
+ "model.layers.35.pre_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
392
+ "model.layers.35.self_attn.k_norm.weight": "model-00008-of-00010.safetensors",
393
+ "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00010.safetensors",
394
+ "model.layers.35.self_attn.o_proj.weight": "model-00008-of-00010.safetensors",
395
+ "model.layers.35.self_attn.q_norm.weight": "model-00008-of-00010.safetensors",
396
+ "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00010.safetensors",
397
+ "model.layers.35.self_attn.v_proj.weight": "model-00008-of-00010.safetensors",
398
+ "model.layers.36.input_layernorm.weight": "model-00008-of-00010.safetensors",
399
+ "model.layers.36.mlp.down_proj.weight": "model-00008-of-00010.safetensors",
400
+ "model.layers.36.mlp.gate_proj.weight": "model-00008-of-00010.safetensors",
401
+ "model.layers.36.mlp.up_proj.weight": "model-00008-of-00010.safetensors",
402
+ "model.layers.36.post_attention_layernorm.weight": "model-00008-of-00010.safetensors",
403
+ "model.layers.36.post_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
404
+ "model.layers.36.pre_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
405
+ "model.layers.36.self_attn.k_norm.weight": "model-00008-of-00010.safetensors",
406
+ "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00010.safetensors",
407
+ "model.layers.36.self_attn.o_proj.weight": "model-00008-of-00010.safetensors",
408
+ "model.layers.36.self_attn.q_norm.weight": "model-00008-of-00010.safetensors",
409
+ "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00010.safetensors",
410
+ "model.layers.36.self_attn.v_proj.weight": "model-00008-of-00010.safetensors",
411
+ "model.layers.37.input_layernorm.weight": "model-00008-of-00010.safetensors",
412
+ "model.layers.37.mlp.down_proj.weight": "model-00008-of-00010.safetensors",
413
+ "model.layers.37.mlp.gate_proj.weight": "model-00008-of-00010.safetensors",
414
+ "model.layers.37.mlp.up_proj.weight": "model-00008-of-00010.safetensors",
415
+ "model.layers.37.post_attention_layernorm.weight": "model-00008-of-00010.safetensors",
416
+ "model.layers.37.post_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
417
+ "model.layers.37.pre_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
418
+ "model.layers.37.self_attn.k_norm.weight": "model-00008-of-00010.safetensors",
419
+ "model.layers.37.self_attn.k_proj.weight": "model-00008-of-00010.safetensors",
420
+ "model.layers.37.self_attn.o_proj.weight": "model-00008-of-00010.safetensors",
421
+ "model.layers.37.self_attn.q_norm.weight": "model-00008-of-00010.safetensors",
422
+ "model.layers.37.self_attn.q_proj.weight": "model-00008-of-00010.safetensors",
423
+ "model.layers.37.self_attn.v_proj.weight": "model-00008-of-00010.safetensors",
424
+ "model.layers.38.input_layernorm.weight": "model-00008-of-00010.safetensors",
425
+ "model.layers.38.mlp.down_proj.weight": "model-00008-of-00010.safetensors",
426
+ "model.layers.38.mlp.gate_proj.weight": "model-00008-of-00010.safetensors",
427
+ "model.layers.38.mlp.up_proj.weight": "model-00008-of-00010.safetensors",
428
+ "model.layers.38.post_attention_layernorm.weight": "model-00008-of-00010.safetensors",
429
+ "model.layers.38.post_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
430
+ "model.layers.38.pre_feedforward_layernorm.weight": "model-00008-of-00010.safetensors",
431
+ "model.layers.38.self_attn.k_norm.weight": "model-00008-of-00010.safetensors",
432
+ "model.layers.38.self_attn.k_proj.weight": "model-00008-of-00010.safetensors",
433
+ "model.layers.38.self_attn.o_proj.weight": "model-00008-of-00010.safetensors",
434
+ "model.layers.38.self_attn.q_norm.weight": "model-00008-of-00010.safetensors",
435
+ "model.layers.38.self_attn.q_proj.weight": "model-00008-of-00010.safetensors",
436
+ "model.layers.38.self_attn.v_proj.weight": "model-00008-of-00010.safetensors",
437
+ "model.layers.39.input_layernorm.weight": "model-00009-of-00010.safetensors",
438
+ "model.layers.39.mlp.down_proj.weight": "model-00009-of-00010.safetensors",
439
+ "model.layers.39.mlp.gate_proj.weight": "model-00008-of-00010.safetensors",
440
+ "model.layers.39.mlp.up_proj.weight": "model-00009-of-00010.safetensors",
441
+ "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00010.safetensors",
442
+ "model.layers.39.post_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
443
+ "model.layers.39.pre_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
444
+ "model.layers.39.self_attn.k_norm.weight": "model-00008-of-00010.safetensors",
445
+ "model.layers.39.self_attn.k_proj.weight": "model-00008-of-00010.safetensors",
446
+ "model.layers.39.self_attn.o_proj.weight": "model-00008-of-00010.safetensors",
447
+ "model.layers.39.self_attn.q_norm.weight": "model-00008-of-00010.safetensors",
448
+ "model.layers.39.self_attn.q_proj.weight": "model-00008-of-00010.safetensors",
449
+ "model.layers.39.self_attn.v_proj.weight": "model-00008-of-00010.safetensors",
450
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00010.safetensors",
451
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00010.safetensors",
452
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00010.safetensors",
453
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00010.safetensors",
454
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00010.safetensors",
455
+ "model.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
456
+ "model.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
457
+ "model.layers.4.self_attn.k_norm.weight": "model-00002-of-00010.safetensors",
458
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00010.safetensors",
459
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00010.safetensors",
460
+ "model.layers.4.self_attn.q_norm.weight": "model-00002-of-00010.safetensors",
461
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00010.safetensors",
462
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00010.safetensors",
463
+ "model.layers.40.input_layernorm.weight": "model-00009-of-00010.safetensors",
464
+ "model.layers.40.mlp.down_proj.weight": "model-00009-of-00010.safetensors",
465
+ "model.layers.40.mlp.gate_proj.weight": "model-00009-of-00010.safetensors",
466
+ "model.layers.40.mlp.up_proj.weight": "model-00009-of-00010.safetensors",
467
+ "model.layers.40.post_attention_layernorm.weight": "model-00009-of-00010.safetensors",
468
+ "model.layers.40.post_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
469
+ "model.layers.40.pre_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
470
+ "model.layers.40.self_attn.k_norm.weight": "model-00009-of-00010.safetensors",
471
+ "model.layers.40.self_attn.k_proj.weight": "model-00009-of-00010.safetensors",
472
+ "model.layers.40.self_attn.o_proj.weight": "model-00009-of-00010.safetensors",
473
+ "model.layers.40.self_attn.q_norm.weight": "model-00009-of-00010.safetensors",
474
+ "model.layers.40.self_attn.q_proj.weight": "model-00009-of-00010.safetensors",
475
+ "model.layers.40.self_attn.v_proj.weight": "model-00009-of-00010.safetensors",
476
+ "model.layers.41.input_layernorm.weight": "model-00009-of-00010.safetensors",
477
+ "model.layers.41.mlp.down_proj.weight": "model-00009-of-00010.safetensors",
478
+ "model.layers.41.mlp.gate_proj.weight": "model-00009-of-00010.safetensors",
479
+ "model.layers.41.mlp.up_proj.weight": "model-00009-of-00010.safetensors",
480
+ "model.layers.41.post_attention_layernorm.weight": "model-00009-of-00010.safetensors",
481
+ "model.layers.41.post_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
482
+ "model.layers.41.pre_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
483
+ "model.layers.41.self_attn.k_norm.weight": "model-00009-of-00010.safetensors",
484
+ "model.layers.41.self_attn.k_proj.weight": "model-00009-of-00010.safetensors",
485
+ "model.layers.41.self_attn.o_proj.weight": "model-00009-of-00010.safetensors",
486
+ "model.layers.41.self_attn.q_norm.weight": "model-00009-of-00010.safetensors",
487
+ "model.layers.41.self_attn.q_proj.weight": "model-00009-of-00010.safetensors",
488
+ "model.layers.41.self_attn.v_proj.weight": "model-00009-of-00010.safetensors",
489
+ "model.layers.42.input_layernorm.weight": "model-00009-of-00010.safetensors",
490
+ "model.layers.42.mlp.down_proj.weight": "model-00009-of-00010.safetensors",
491
+ "model.layers.42.mlp.gate_proj.weight": "model-00009-of-00010.safetensors",
492
+ "model.layers.42.mlp.up_proj.weight": "model-00009-of-00010.safetensors",
493
+ "model.layers.42.post_attention_layernorm.weight": "model-00009-of-00010.safetensors",
494
+ "model.layers.42.post_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
495
+ "model.layers.42.pre_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
496
+ "model.layers.42.self_attn.k_norm.weight": "model-00009-of-00010.safetensors",
497
+ "model.layers.42.self_attn.k_proj.weight": "model-00009-of-00010.safetensors",
498
+ "model.layers.42.self_attn.o_proj.weight": "model-00009-of-00010.safetensors",
499
+ "model.layers.42.self_attn.q_norm.weight": "model-00009-of-00010.safetensors",
500
+ "model.layers.42.self_attn.q_proj.weight": "model-00009-of-00010.safetensors",
501
+ "model.layers.42.self_attn.v_proj.weight": "model-00009-of-00010.safetensors",
502
+ "model.layers.43.input_layernorm.weight": "model-00009-of-00010.safetensors",
503
+ "model.layers.43.mlp.down_proj.weight": "model-00009-of-00010.safetensors",
504
+ "model.layers.43.mlp.gate_proj.weight": "model-00009-of-00010.safetensors",
505
+ "model.layers.43.mlp.up_proj.weight": "model-00009-of-00010.safetensors",
506
+ "model.layers.43.post_attention_layernorm.weight": "model-00009-of-00010.safetensors",
507
+ "model.layers.43.post_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
508
+ "model.layers.43.pre_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
509
+ "model.layers.43.self_attn.k_norm.weight": "model-00009-of-00010.safetensors",
510
+ "model.layers.43.self_attn.k_proj.weight": "model-00009-of-00010.safetensors",
511
+ "model.layers.43.self_attn.o_proj.weight": "model-00009-of-00010.safetensors",
512
+ "model.layers.43.self_attn.q_norm.weight": "model-00009-of-00010.safetensors",
513
+ "model.layers.43.self_attn.q_proj.weight": "model-00009-of-00010.safetensors",
514
+ "model.layers.43.self_attn.v_proj.weight": "model-00009-of-00010.safetensors",
515
+ "model.layers.44.input_layernorm.weight": "model-00009-of-00010.safetensors",
516
+ "model.layers.44.mlp.down_proj.weight": "model-00009-of-00010.safetensors",
517
+ "model.layers.44.mlp.gate_proj.weight": "model-00009-of-00010.safetensors",
518
+ "model.layers.44.mlp.up_proj.weight": "model-00009-of-00010.safetensors",
519
+ "model.layers.44.post_attention_layernorm.weight": "model-00009-of-00010.safetensors",
520
+ "model.layers.44.post_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
521
+ "model.layers.44.pre_feedforward_layernorm.weight": "model-00009-of-00010.safetensors",
522
+ "model.layers.44.self_attn.k_norm.weight": "model-00009-of-00010.safetensors",
523
+ "model.layers.44.self_attn.k_proj.weight": "model-00009-of-00010.safetensors",
524
+ "model.layers.44.self_attn.o_proj.weight": "model-00009-of-00010.safetensors",
525
+ "model.layers.44.self_attn.q_norm.weight": "model-00009-of-00010.safetensors",
526
+ "model.layers.44.self_attn.q_proj.weight": "model-00009-of-00010.safetensors",
527
+ "model.layers.44.self_attn.v_proj.weight": "model-00009-of-00010.safetensors",
528
+ "model.layers.45.input_layernorm.weight": "model-00010-of-00010.safetensors",
529
+ "model.layers.45.mlp.down_proj.weight": "model-00010-of-00010.safetensors",
530
+ "model.layers.45.mlp.gate_proj.weight": "model-00010-of-00010.safetensors",
531
+ "model.layers.45.mlp.up_proj.weight": "model-00010-of-00010.safetensors",
532
+ "model.layers.45.post_attention_layernorm.weight": "model-00010-of-00010.safetensors",
533
+ "model.layers.45.post_feedforward_layernorm.weight": "model-00010-of-00010.safetensors",
534
+ "model.layers.45.pre_feedforward_layernorm.weight": "model-00010-of-00010.safetensors",
535
+ "model.layers.45.self_attn.k_norm.weight": "model-00010-of-00010.safetensors",
536
+ "model.layers.45.self_attn.k_proj.weight": "model-00010-of-00010.safetensors",
537
+ "model.layers.45.self_attn.o_proj.weight": "model-00010-of-00010.safetensors",
538
+ "model.layers.45.self_attn.q_norm.weight": "model-00010-of-00010.safetensors",
539
+ "model.layers.45.self_attn.q_proj.weight": "model-00010-of-00010.safetensors",
540
+ "model.layers.45.self_attn.v_proj.weight": "model-00010-of-00010.safetensors",
541
+ "model.layers.46.input_layernorm.weight": "model-00010-of-00010.safetensors",
542
+ "model.layers.46.mlp.down_proj.weight": "model-00010-of-00010.safetensors",
543
+ "model.layers.46.mlp.gate_proj.weight": "model-00010-of-00010.safetensors",
544
+ "model.layers.46.mlp.up_proj.weight": "model-00010-of-00010.safetensors",
545
+ "model.layers.46.post_attention_layernorm.weight": "model-00010-of-00010.safetensors",
546
+ "model.layers.46.post_feedforward_layernorm.weight": "model-00010-of-00010.safetensors",
547
+ "model.layers.46.pre_feedforward_layernorm.weight": "model-00010-of-00010.safetensors",
548
+ "model.layers.46.self_attn.k_norm.weight": "model-00010-of-00010.safetensors",
549
+ "model.layers.46.self_attn.k_proj.weight": "model-00010-of-00010.safetensors",
550
+ "model.layers.46.self_attn.o_proj.weight": "model-00010-of-00010.safetensors",
551
+ "model.layers.46.self_attn.q_norm.weight": "model-00010-of-00010.safetensors",
552
+ "model.layers.46.self_attn.q_proj.weight": "model-00010-of-00010.safetensors",
553
+ "model.layers.46.self_attn.v_proj.weight": "model-00010-of-00010.safetensors",
554
+ "model.layers.47.input_layernorm.weight": "model-00010-of-00010.safetensors",
555
+ "model.layers.47.mlp.down_proj.weight": "model-00010-of-00010.safetensors",
556
+ "model.layers.47.mlp.gate_proj.weight": "model-00010-of-00010.safetensors",
557
+ "model.layers.47.mlp.up_proj.weight": "model-00010-of-00010.safetensors",
558
+ "model.layers.47.post_attention_layernorm.weight": "model-00010-of-00010.safetensors",
559
+ "model.layers.47.post_feedforward_layernorm.weight": "model-00010-of-00010.safetensors",
560
+ "model.layers.47.pre_feedforward_layernorm.weight": "model-00010-of-00010.safetensors",
561
+ "model.layers.47.self_attn.k_norm.weight": "model-00010-of-00010.safetensors",
562
+ "model.layers.47.self_attn.k_proj.weight": "model-00010-of-00010.safetensors",
563
+ "model.layers.47.self_attn.o_proj.weight": "model-00010-of-00010.safetensors",
564
+ "model.layers.47.self_attn.q_norm.weight": "model-00010-of-00010.safetensors",
565
+ "model.layers.47.self_attn.q_proj.weight": "model-00010-of-00010.safetensors",
566
+ "model.layers.47.self_attn.v_proj.weight": "model-00010-of-00010.safetensors",
567
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00010.safetensors",
568
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00010.safetensors",
569
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00010.safetensors",
570
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00010.safetensors",
571
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00010.safetensors",
572
+ "model.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
573
+ "model.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00010.safetensors",
574
+ "model.layers.5.self_attn.k_norm.weight": "model-00002-of-00010.safetensors",
575
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00010.safetensors",
576
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00010.safetensors",
577
+ "model.layers.5.self_attn.q_norm.weight": "model-00002-of-00010.safetensors",
578
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00010.safetensors",
579
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00010.safetensors",
580
+ "model.layers.6.input_layernorm.weight": "model-00003-of-00010.safetensors",
581
+ "model.layers.6.mlp.down_proj.weight": "model-00003-of-00010.safetensors",
582
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00010.safetensors",
583
+ "model.layers.6.mlp.up_proj.weight": "model-00003-of-00010.safetensors",
584
+ "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00010.safetensors",
585
+ "model.layers.6.post_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
586
+ "model.layers.6.pre_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
587
+ "model.layers.6.self_attn.k_norm.weight": "model-00002-of-00010.safetensors",
588
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00010.safetensors",
589
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00010.safetensors",
590
+ "model.layers.6.self_attn.q_norm.weight": "model-00002-of-00010.safetensors",
591
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00010.safetensors",
592
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00010.safetensors",
593
+ "model.layers.7.input_layernorm.weight": "model-00003-of-00010.safetensors",
594
+ "model.layers.7.mlp.down_proj.weight": "model-00003-of-00010.safetensors",
595
+ "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00010.safetensors",
596
+ "model.layers.7.mlp.up_proj.weight": "model-00003-of-00010.safetensors",
597
+ "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00010.safetensors",
598
+ "model.layers.7.post_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
599
+ "model.layers.7.pre_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
600
+ "model.layers.7.self_attn.k_norm.weight": "model-00003-of-00010.safetensors",
601
+ "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00010.safetensors",
602
+ "model.layers.7.self_attn.o_proj.weight": "model-00003-of-00010.safetensors",
603
+ "model.layers.7.self_attn.q_norm.weight": "model-00003-of-00010.safetensors",
604
+ "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00010.safetensors",
605
+ "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00010.safetensors",
606
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00010.safetensors",
607
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00010.safetensors",
608
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00010.safetensors",
609
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00010.safetensors",
610
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00010.safetensors",
611
+ "model.layers.8.post_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
612
+ "model.layers.8.pre_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
613
+ "model.layers.8.self_attn.k_norm.weight": "model-00003-of-00010.safetensors",
614
+ "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00010.safetensors",
615
+ "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00010.safetensors",
616
+ "model.layers.8.self_attn.q_norm.weight": "model-00003-of-00010.safetensors",
617
+ "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00010.safetensors",
618
+ "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00010.safetensors",
619
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00010.safetensors",
620
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00010.safetensors",
621
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00010.safetensors",
622
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00010.safetensors",
623
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00010.safetensors",
624
+ "model.layers.9.post_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
625
+ "model.layers.9.pre_feedforward_layernorm.weight": "model-00003-of-00010.safetensors",
626
+ "model.layers.9.self_attn.k_norm.weight": "model-00003-of-00010.safetensors",
627
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00010.safetensors",
628
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00010.safetensors",
629
+ "model.layers.9.self_attn.q_norm.weight": "model-00003-of-00010.safetensors",
630
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00010.safetensors",
631
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00010.safetensors",
632
+ "model.norm.weight": "model-00010-of-00010.safetensors"
633
+ }
634
+ }