Hugging Face's logo

HumanF-MarkrAI
/
Gukbap-Qwen2-34B-VL

Text Generation
Transformers
Safetensors
ovis
conversational
custom_code
Model card Files Community
kyujinpy commited on
Commit
9ec5a8d
·
verified ·
1 Parent(s): 3a4da1b

Upload Ovis

Browse files
Files changed (30) hide show
  1. .gitattributes +1 -0
  2. README.md +199 -0
  3. added_tokens.json +24 -0
  4. config.json +257 -0
  5. configuration_aimv2.py +63 -0
  6. configuration_ovis.py +204 -0
  7. generation_config.json +15 -0
  8. merges.txt +0 -0
  9. model-00001-of-00015.safetensors +3 -0
  10. model-00002-of-00015.safetensors +3 -0
  11. model-00003-of-00015.safetensors +3 -0
  12. model-00004-of-00015.safetensors +3 -0
  13. model-00005-of-00015.safetensors +3 -0
  14. model-00006-of-00015.safetensors +3 -0
  15. model-00007-of-00015.safetensors +3 -0
  16. model-00008-of-00015.safetensors +3 -0
  17. model-00009-of-00015.safetensors +3 -0
  18. model-00010-of-00015.safetensors +3 -0
  19. model-00011-of-00015.safetensors +3 -0
  20. model-00012-of-00015.safetensors +3 -0
  21. model-00013-of-00015.safetensors +3 -0
  22. model-00014-of-00015.safetensors +3 -0
  23. model-00015-of-00015.safetensors +3 -0
  24. model.safetensors.index.json +955 -0
  25. modeling_aimv2.py +198 -0
  26. preprocessor_config.json +27 -0
  27. special_tokens_map.json +31 -0
  28. tokenizer.json +3 -0
  29. tokenizer_config.json +207 -0
  30. vocab.json +0 -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 +1,199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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]
added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
config.json ADDED
@@ -0,0 +1,257 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/data/LMM/Gukbap-Gemma2-34B-ver2/checkpoint-42",
3
+ "architectures": [
4
+ "Ovis"
5
+ ],
6
+ "auto_map": {
7
+ "AutoConfig": "configuration_ovis.OvisConfig",
8
+ "AutoModelForCausalLM": "AIDC-AI/Ovis2-34B--modeling_ovis.Ovis"
9
+ },
10
+ "conversation_formatter_class": "QwenConversationFormatter",
11
+ "disable_tie_weight": false,
12
+ "hidden_size": 5120,
13
+ "llm_attn_implementation": "flash_attention_2",
14
+ "llm_config": {
15
+ "_attn_implementation_autoset": true,
16
+ "_name_or_path": "Qwen/Qwen2.5-32B-Instruct",
17
+ "add_cross_attention": false,
18
+ "architectures": [
19
+ "Qwen2ForCausalLM"
20
+ ],
21
+ "attention_dropout": 0.0,
22
+ "bad_words_ids": null,
23
+ "begin_suppress_tokens": null,
24
+ "bos_token_id": 151643,
25
+ "chunk_size_feed_forward": 0,
26
+ "cross_attention_hidden_size": null,
27
+ "decoder_start_token_id": null,
28
+ "diversity_penalty": 0.0,
29
+ "do_sample": false,
30
+ "early_stopping": false,
31
+ "encoder_no_repeat_ngram_size": 0,
32
+ "eos_token_id": 151645,
33
+ "exponential_decay_length_penalty": null,
34
+ "finetuning_task": null,
35
+ "forced_bos_token_id": null,
36
+ "forced_eos_token_id": null,
37
+ "hidden_act": "silu",
38
+ "hidden_size": 5120,
39
+ "id2label": {
40
+ "0": "LABEL_0",
41
+ "1": "LABEL_1"
42
+ },
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 27648,
45
+ "is_decoder": false,
46
+ "is_encoder_decoder": false,
47
+ "label2id": {
48
+ "LABEL_0": 0,
49
+ "LABEL_1": 1
50
+ },
51
+ "length_penalty": 1.0,
52
+ "max_length": 20,
53
+ "max_position_embeddings": 32768,
54
+ "max_window_layers": 70,
55
+ "min_length": 0,
56
+ "model_type": "qwen2",
57
+ "no_repeat_ngram_size": 0,
58
+ "num_attention_heads": 40,
59
+ "num_beam_groups": 1,
60
+ "num_beams": 1,
61
+ "num_hidden_layers": 64,
62
+ "num_key_value_heads": 8,
63
+ "num_return_sequences": 1,
64
+ "output_attentions": false,
65
+ "output_hidden_states": false,
66
+ "output_scores": false,
67
+ "pad_token_id": null,
68
+ "prefix": null,
69
+ "problem_type": null,
70
+ "pruned_heads": {},
71
+ "remove_invalid_values": false,
72
+ "repetition_penalty": 1.0,
73
+ "return_dict": true,
74
+ "return_dict_in_generate": false,
75
+ "rms_norm_eps": 1e-06,
76
+ "rope_scaling": null,
77
+ "rope_theta": 1000000.0,
78
+ "sep_token_id": null,
79
+ "sliding_window": null,
80
+ "suppress_tokens": null,
81
+ "task_specific_params": null,
82
+ "temperature": 1.0,
83
+ "tf_legacy_loss": false,
84
+ "tie_encoder_decoder": false,
85
+ "tie_word_embeddings": false,
86
+ "tokenizer_class": null,
87
+ "top_k": 50,
88
+ "top_p": 1.0,
89
+ "torch_dtype": "bfloat16",
90
+ "torchscript": false,
91
+ "typical_p": 1.0,
92
+ "use_bfloat16": false,
93
+ "use_cache": true,
94
+ "use_sliding_window": false,
95
+ "vocab_size": 152064
96
+ },
97
+ "model_type": "ovis",
98
+ "multimodal_max_length": 2048,
99
+ "torch_dtype": "float16",
100
+ "transformers_version": "4.46.2",
101
+ "use_cache": false,
102
+ "visual_tokenizer_config": {
103
+ "_attn_implementation_autoset": true,
104
+ "_name_or_path": "",
105
+ "add_cross_attention": false,
106
+ "architectures": null,
107
+ "backbone_config": {
108
+ "_attn_implementation_autoset": true,
109
+ "_name_or_path": "apple/aimv2-1B-patch14-448",
110
+ "add_cross_attention": false,
111
+ "architectures": [
112
+ "AIMv2Model"
113
+ ],
114
+ "attention_dropout": 0.0,
115
+ "auto_map": {
116
+ "AutoConfig": "configuration_aimv2.AIMv2Config",
117
+ "AutoModel": "modeling_aimv2.AIMv2Model",
118
+ "FlaxAutoModel": "modeling_flax_aimv2.FlaxAIMv2Model"
119
+ },
120
+ "bad_words_ids": null,
121
+ "begin_suppress_tokens": null,
122
+ "bos_token_id": null,
123
+ "chunk_size_feed_forward": 0,
124
+ "cross_attention_hidden_size": null,
125
+ "decoder_start_token_id": null,
126
+ "diversity_penalty": 0.0,
127
+ "do_sample": false,
128
+ "early_stopping": false,
129
+ "encoder_no_repeat_ngram_size": 0,
130
+ "eos_token_id": null,
131
+ "exponential_decay_length_penalty": null,
132
+ "finetuning_task": null,
133
+ "forced_bos_token_id": null,
134
+ "forced_eos_token_id": null,
135
+ "hidden_size": 2048,
136
+ "id2label": {
137
+ "0": "LABEL_0",
138
+ "1": "LABEL_1"
139
+ },
140
+ "image_size": 448,
141
+ "intermediate_size": 5632,
142
+ "is_decoder": false,
143
+ "is_encoder_decoder": false,
144
+ "label2id": {
145
+ "LABEL_0": 0,
146
+ "LABEL_1": 1
147
+ },
148
+ "length_penalty": 1.0,
149
+ "max_length": 20,
150
+ "min_length": 0,
151
+ "model_type": "aimv2",
152
+ "no_repeat_ngram_size": 0,
153
+ "num_attention_heads": 16,
154
+ "num_beam_groups": 1,
155
+ "num_beams": 1,
156
+ "num_channels": 3,
157
+ "num_hidden_layers": 24,
158
+ "num_return_sequences": 1,
159
+ "output_attentions": false,
160
+ "output_hidden_states": false,
161
+ "output_scores": false,
162
+ "pad_token_id": null,
163
+ "patch_size": 14,
164
+ "prefix": null,
165
+ "problem_type": null,
166
+ "projection_dropout": 0.0,
167
+ "pruned_heads": {},
168
+ "qkv_bias": false,
169
+ "remove_invalid_values": false,
170
+ "repetition_penalty": 1.0,
171
+ "return_dict": true,
172
+ "return_dict_in_generate": false,
173
+ "rms_norm_eps": 1e-05,
174
+ "sep_token_id": null,
175
+ "suppress_tokens": null,
176
+ "task_specific_params": null,
177
+ "temperature": 1.0,
178
+ "tf_legacy_loss": false,
179
+ "tie_encoder_decoder": false,
180
+ "tie_word_embeddings": true,
181
+ "tokenizer_class": null,
182
+ "top_k": 50,
183
+ "top_p": 1.0,
184
+ "torch_dtype": "float32",
185
+ "torchscript": false,
186
+ "typical_p": 1.0,
187
+ "use_bfloat16": false,
188
+ "use_bias": false
189
+ },
190
+ "backbone_kwargs": {},
191
+ "bad_words_ids": null,
192
+ "begin_suppress_tokens": null,
193
+ "bos_token_id": null,
194
+ "chunk_size_feed_forward": 0,
195
+ "cross_attention_hidden_size": null,
196
+ "decoder_start_token_id": null,
197
+ "depths": null,
198
+ "diversity_penalty": 0.0,
199
+ "do_sample": false,
200
+ "drop_cls_token": false,
201
+ "early_stopping": false,
202
+ "encoder_no_repeat_ngram_size": 0,
203
+ "eos_token_id": null,
204
+ "exponential_decay_length_penalty": null,
205
+ "finetuning_task": null,
206
+ "forced_bos_token_id": null,
207
+ "forced_eos_token_id": null,
208
+ "hidden_stride": 2,
209
+ "id2label": {
210
+ "0": "LABEL_0",
211
+ "1": "LABEL_1"
212
+ },
213
+ "is_decoder": false,
214
+ "is_encoder_decoder": false,
215
+ "label2id": {
216
+ "LABEL_0": 0,
217
+ "LABEL_1": 1
218
+ },
219
+ "length_penalty": 1.0,
220
+ "max_length": 20,
221
+ "min_length": 0,
222
+ "model_type": "aimv2_visual_tokenizer",
223
+ "no_repeat_ngram_size": 0,
224
+ "num_beam_groups": 1,
225
+ "num_beams": 1,
226
+ "num_return_sequences": 1,
227
+ "output_attentions": false,
228
+ "output_hidden_states": false,
229
+ "output_scores": false,
230
+ "pad_token_id": null,
231
+ "prefix": null,
232
+ "problem_type": null,
233
+ "pruned_heads": {},
234
+ "remove_invalid_values": false,
235
+ "repetition_penalty": 1.0,
236
+ "return_dict": true,
237
+ "return_dict_in_generate": false,
238
+ "sep_token_id": null,
239
+ "suppress_tokens": null,
240
+ "task_specific_params": null,
241
+ "tau": 1.0,
242
+ "temperature": 1.0,
243
+ "tf_legacy_loss": false,
244
+ "tie_encoder_decoder": false,
245
+ "tie_word_embeddings": true,
246
+ "tokenize_function": "softmax",
247
+ "tokenizer_class": null,
248
+ "top_k": 50,
249
+ "top_p": 1.0,
250
+ "torch_dtype": null,
251
+ "torchscript": false,
252
+ "typical_p": 1.0,
253
+ "use_bfloat16": false,
254
+ "use_indicators": false,
255
+ "vocab_size": 65536
256
+ }
257
+ }
configuration_aimv2.py ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # copied from https://huggingface.co/apple/aimv2-huge-patch14-448
2
+ from typing import Any
3
+
4
+ from transformers.configuration_utils import PretrainedConfig
5
+
6
+ __all__ = ["AIMv2Config"]
7
+
8
+
9
+ class AIMv2Config(PretrainedConfig):
10
+ """This is the configuration class to store the configuration of an [`AIMv2Model`].
11
+
12
+ Instantiating a configuration with the defaults will yield a similar configuration
13
+ to that of the [apple/aimv2-large-patch14-224](https://huggingface.co/apple/aimv2-large-patch14-224).
14
+
15
+ Args:
16
+ hidden_size: Dimension of the hidden representations.
17
+ intermediate_size: Dimension of the SwiGLU representations.
18
+ num_hidden_layers: Number of hidden layers in the Transformer.
19
+ num_attention_heads: Number of attention heads for each attention layer
20
+ in the Transformer.
21
+ num_channels: Number of input channels.
22
+ image_size: Image size.
23
+ patch_size: Patch size.
24
+ rms_norm_eps: Epsilon value used for the RMS normalization layer.
25
+ attention_dropout: Dropout ratio for attention probabilities.
26
+ projection_dropout: Dropout ratio for the projection layer after the attention.
27
+ qkv_bias: Whether to add a bias to the queries, keys and values.
28
+ use_bias: Whether to add a bias in the feed-forward and projection layers.
29
+ kwargs: Keyword arguments for the [`PretrainedConfig`].
30
+ """
31
+
32
+ model_type: str = "aimv2"
33
+
34
+ def __init__(
35
+ self,
36
+ hidden_size: int = 1024,
37
+ intermediate_size: int = 2816,
38
+ num_hidden_layers: int = 24,
39
+ num_attention_heads: int = 8,
40
+ num_channels: int = 3,
41
+ image_size: int = 224,
42
+ patch_size: int = 14,
43
+ rms_norm_eps: float = 1e-5,
44
+ attention_dropout: float = 0.0,
45
+ projection_dropout: float = 0.0,
46
+ qkv_bias: bool = False,
47
+ use_bias: bool = False,
48
+ **kwargs: Any,
49
+ ):
50
+ super().__init__(**kwargs)
51
+ self.hidden_size = hidden_size
52
+ self.intermediate_size = intermediate_size
53
+ self.num_hidden_layers = num_hidden_layers
54
+ self.num_attention_heads = num_attention_heads
55
+ self.num_channels = num_channels
56
+ self.patch_size = patch_size
57
+ self.image_size = image_size
58
+ self.attention_dropout = attention_dropout
59
+ self.rms_norm_eps = rms_norm_eps
60
+
61
+ self.projection_dropout = projection_dropout
62
+ self.qkv_bias = qkv_bias
63
+ self.use_bias = use_bias
configuration_ovis.py ADDED
@@ -0,0 +1,204 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from abc import ABC, abstractmethod
2
+ from typing import List, Dict, Union, Optional
3
+
4
+ from transformers import PretrainedConfig, AutoConfig, AutoModel
5
+ from .configuration_aimv2 import AIMv2Config
6
+ from .modeling_aimv2 import AIMv2Model
7
+
8
+ IGNORE_ID = -100
9
+ IMAGE_TOKEN_ID = -200
10
+ IMAGE_TOKEN = "<image>"
11
+ IMAGE_ATOM_ID = -300
12
+ IMAGE_INDICATOR_IDS = [-301, -302, -303, -304, -305]
13
+
14
+ AutoConfig.register("aimv2", AIMv2Config)
15
+ AutoModel.register(AIMv2Config, AIMv2Model)
16
+
17
+ # ----------------------------------------------------------------------
18
+ # Visual Tokenizer Configuration
19
+ # ----------------------------------------------------------------------
20
+ class BaseVisualTokenizerConfig(PretrainedConfig):
21
+ def __init__(
22
+ self,
23
+ vocab_size=16384,
24
+ tokenize_function="softmax",
25
+ tau=1.0,
26
+ depths=None,
27
+ drop_cls_token=False,
28
+ backbone_config: Optional[Union[PretrainedConfig, dict]] = None,
29
+ hidden_stride: int = 1,
30
+ **kwargs
31
+ ):
32
+ super().__init__(**kwargs)
33
+ self.vocab_size = vocab_size
34
+ self.tokenize_function = tokenize_function
35
+ self.tau = tau
36
+ if isinstance(depths, str):
37
+ depths = [int(x) for x in depths.split('|')]
38
+ self.depths = depths
39
+ self.backbone_kwargs = {}
40
+ self.drop_cls_token = drop_cls_token
41
+ if backbone_config is not None:
42
+ assert isinstance(backbone_config, (PretrainedConfig, dict)), \
43
+ f"expect `backbone_config` to be instance of PretrainedConfig or dict, but got {type(backbone_config)} type"
44
+ if not isinstance(backbone_config, PretrainedConfig):
45
+ model_type = backbone_config['model_type']
46
+ backbone_config.pop('model_type')
47
+ backbone_config = AutoConfig.for_model(model_type, **backbone_config)
48
+ self.backbone_config = backbone_config
49
+ self.hidden_stride = hidden_stride
50
+
51
+
52
+ class Aimv2VisualTokenizerConfig(BaseVisualTokenizerConfig):
53
+ model_type = "aimv2_visual_tokenizer"
54
+
55
+ def __init__(self, **kwargs):
56
+ super().__init__(**kwargs)
57
+ if self.drop_cls_token:
58
+ self.drop_cls_token = False
59
+ if self.depths:
60
+ assert len(self.depths) == 1
61
+ self.backbone_kwargs['num_hidden_layers'] = self.depths[0]
62
+
63
+
64
+ AutoConfig.register("aimv2_visual_tokenizer", Aimv2VisualTokenizerConfig)
65
+
66
+
67
+ # ----------------------------------------------------------------------
68
+ # Ovis Configuration
69
+ # ----------------------------------------------------------------------
70
+ class OvisConfig(PretrainedConfig):
71
+ model_type = "ovis"
72
+
73
+ def __init__(
74
+ self,
75
+ llm_config: Optional[Union[PretrainedConfig, dict]] = None,
76
+ visual_tokenizer_config: Optional[Union[PretrainedConfig, dict]] = None,
77
+ multimodal_max_length=8192,
78
+ hidden_size=None,
79
+ conversation_formatter_class=None,
80
+ llm_attn_implementation=None,
81
+ disable_tie_weight=False,
82
+ **kwargs
83
+ ):
84
+ super().__init__(**kwargs)
85
+ if llm_config is not None:
86
+ assert isinstance(llm_config, (PretrainedConfig, dict)), \
87
+ f"expect `llm_config` to be instance of PretrainedConfig or dict, but got {type(llm_config)} type"
88
+ if not isinstance(llm_config, PretrainedConfig):
89
+ model_type = llm_config['model_type']
90
+ llm_config.pop('model_type')
91
+ llm_config = AutoConfig.for_model(model_type, **llm_config)
92
+ self.llm_config = llm_config
93
+ if visual_tokenizer_config is not None:
94
+ assert isinstance(visual_tokenizer_config, (PretrainedConfig, dict)), \
95
+ f"expect `visual_tokenizer_config` to be instance of PretrainedConfig or dict, but got {type(visual_tokenizer_config)} type"
96
+ if not isinstance(visual_tokenizer_config, PretrainedConfig):
97
+ model_type = visual_tokenizer_config['model_type']
98
+ visual_tokenizer_config.pop('model_type')
99
+ visual_tokenizer_config = AutoConfig.for_model(model_type, **visual_tokenizer_config)
100
+ self.visual_tokenizer_config = visual_tokenizer_config
101
+ self.multimodal_max_length = multimodal_max_length
102
+ self.hidden_size = hidden_size
103
+ self.conversation_formatter_class = conversation_formatter_class
104
+ self.llm_attn_implementation = llm_attn_implementation
105
+ self.disable_tie_weight = disable_tie_weight
106
+
107
+
108
+ # ----------------------------------------------------------------------
109
+ # Conversation Formatter
110
+ # ----------------------------------------------------------------------
111
+ class ConversationFormatter(ABC):
112
+ support_tokenizer_types = None
113
+
114
+ def __init__(self, tokenizer):
115
+ tokenizer_type = type(tokenizer).__name__
116
+ assert tokenizer_type in self.support_tokenizer_types, \
117
+ f'Invalid tokenizer type, expected one from `{self.support_tokenizer_types}`, but got `{tokenizer_type}`'
118
+ self.tokenizer = tokenizer
119
+ self.image_token = IMAGE_TOKEN
120
+ self.image_token_id = IMAGE_TOKEN_ID
121
+ self.ignore_id = IGNORE_ID
122
+
123
+ def _tokenize_with_image_symbol(self, text):
124
+ text_chunks = [self.tokenizer(chunk, add_special_tokens=False).input_ids for chunk in
125
+ text.split(self.image_token)]
126
+ token_ids = []
127
+ num_chuck = len(text_chunks)
128
+ for i, chunk in enumerate(text_chunks):
129
+ token_ids.extend(chunk)
130
+ if i < num_chuck - 1:
131
+ token_ids.append(self.image_token_id)
132
+ return token_ids
133
+
134
+ @abstractmethod
135
+ def format(self, conversations: List[Dict], generation_preface=None):
136
+ pass
137
+
138
+ @abstractmethod
139
+ def format_query(self, query, generation_preface=""):
140
+ pass
141
+
142
+
143
+ class QwenConversationFormatter(ConversationFormatter):
144
+ support_tokenizer_types = ['QWenTokenizer', 'Qwen2TokenizerFast']
145
+
146
+ def __init__(self, tokenizer):
147
+ super().__init__(tokenizer)
148
+ self.from2role = {
149
+ "system": "<|im_start|>system\n",
150
+ "human": "<|im_start|>user\n",
151
+ "gpt": "<|im_start|>assistant\n",
152
+ }
153
+ self.gpt_token_num = None
154
+ self.im_end = "<|im_end|>\n"
155
+ self.default_system_prompt = "You are a helpful assistant."
156
+
157
+ def format(self, conversations: List[Dict], generation_preface=None):
158
+ if self.gpt_token_num is None:
159
+ self.gpt_token_num = len(self.tokenizer(self.from2role["gpt"], add_special_tokens=False).input_ids)
160
+
161
+ if conversations[0]["from"] != "system":
162
+ conversations.insert(0, {
163
+ "from": "system",
164
+ "value": self.default_system_prompt
165
+ })
166
+
167
+ if generation_preface is not None:
168
+ conversations.append({
169
+ "from": "gpt",
170
+ "value": generation_preface
171
+ })
172
+
173
+ prompt = ""
174
+ input_ids = []
175
+ labels = []
176
+ num_conversation = len(conversations)
177
+ for i, conversation in enumerate(conversations):
178
+ frm = conversation["from"]
179
+ role = self.from2role[frm]
180
+ message = conversation["value"]
181
+ text = role + message
182
+ if i < num_conversation - 1 or generation_preface is None:
183
+ text += self.im_end
184
+ prompt += text
185
+ token_ids = self._tokenize_with_image_symbol(text)
186
+ input_ids.extend(token_ids)
187
+ label_ids = [self.ignore_id] * len(token_ids)
188
+ if frm == "gpt" and generation_preface is None:
189
+ # learning `\n` following `im_end` is meaningless, so the last `\n` token is ignored in label
190
+ label_ids[self.gpt_token_num:-1] = token_ids[self.gpt_token_num:-1]
191
+ labels.extend(label_ids)
192
+
193
+ assert self._tokenize_with_image_symbol(prompt) == input_ids
194
+ assert len(input_ids) == len(labels)
195
+
196
+ return prompt, input_ids, labels
197
+
198
+ def format_query(self, query, generation_preface=""):
199
+ prompt, input_ids, _ = self.format([{
200
+ "from": "human",
201
+ "value": query
202
+ }], generation_preface=generation_preface)
203
+
204
+ return prompt, input_ids
generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "multimodal_max_length": 32768,
9
+ "pad_token_id": 151643,
10
+ "repetition_penalty": 1.05,
11
+ "temperature": 0.7,
12
+ "top_k": 20,
13
+ "top_p": 0.8,
14
+ "transformers_version": "4.46.2"
15
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:532a33dd1821f22458f030bff7ebd1e5a34d7a6372312ca975fbb84e7fdc4e97
3
+ size 4891731128
model-00002-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5de73bff4a2173d6cdbe4087e2f30c622c3761750a7711dd397deeb69b70861c
3
+ size 4876059536
model-00003-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e199724971e60bf8903c7d7c6a1e28883f1d8eb281bf72019f5ccf16f644743
3
+ size 4876059568
model-00004-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be99bbb2818d7dd297c0b4223346c5c3ea61ce263e59307f444a63301e4900f4
3
+ size 4876059592
model-00005-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d352e7c16cf56242081c7efbe1c5b08a013df02811ce271727e18b02b6fa6df2
3
+ size 4876059592
model-00006-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7bc4643a2f10540dc283e5ce7944872a2aa0a5fd46b3676d8597630a88caaa8
3
+ size 4876059592
model-00007-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a5823124493c352fb3a094d42993b2d9e2f960ad672778569cb4eae241f6f5d
3
+ size 4876059592
model-00008-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7202b7d362372c074f7bc222b7a79b3e95009eae5f400a26583d3b96c769a2fd
3
+ size 4876059592
model-00009-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba850168957211e9640600606cad93490752e3998432e0c557c86d1a9777cc53
3
+ size 4876059592
model-00010-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:054a8461ab24f7db87c516165005f0f4ece33707821eb1366ea8b703f4bb2c93
3
+ size 4876059592
model-00011-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fd011879484badd772813c93b573410d9af085c3844aa0a85feb326c84a2b97
3
+ size 4876059592
model-00012-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30db476f7cfd0f653c69c9bd5a8bfd2ec86d895a93ed503ce9a0fdaab3355605
3
+ size 4876059592
model-00013-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:993612dc1de43868b14ee7b43740f896afbeccc87c4ce61da755bea0d46f18d7
3
+ size 4876059592
model-00014-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50915597c36a2807eb491bf92abc67d1c7d6b7e8bd221918238fcbb790f32715
3
+ size 4596483112
model-00015-of-00015.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56d99aac122e6125983d45cd47033ec1e2cd6b81c7a8ed1674db4f5fa928324a
3
+ size 1745011108
model.safetensors.index.json ADDED
@@ -0,0 +1,955 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 69745825772
4
+ },
5
+ "weight_map": {
6
+ "llm.lm_head.weight": "model-00014-of-00015.safetensors",
7
+ "llm.model.embed_tokens.weight": "model-00001-of-00015.safetensors",
8
+ "llm.model.layers.0.input_layernorm.weight": "model-00001-of-00015.safetensors",
9
+ "llm.model.layers.0.mlp.down_proj.weight": "model-00001-of-00015.safetensors",
10
+ "llm.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00015.safetensors",
11
+ "llm.model.layers.0.mlp.up_proj.weight": "model-00001-of-00015.safetensors",
12
+ "llm.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00015.safetensors",
13
+ "llm.model.layers.0.self_attn.k_proj.bias": "model-00001-of-00015.safetensors",
14
+ "llm.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00015.safetensors",
15
+ "llm.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00015.safetensors",
16
+ "llm.model.layers.0.self_attn.q_proj.bias": "model-00001-of-00015.safetensors",
17
+ "llm.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00015.safetensors",
18
+ "llm.model.layers.0.self_attn.v_proj.bias": "model-00001-of-00015.safetensors",
19
+ "llm.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00015.safetensors",
20
+ "llm.model.layers.1.input_layernorm.weight": "model-00001-of-00015.safetensors",
21
+ "llm.model.layers.1.mlp.down_proj.weight": "model-00001-of-00015.safetensors",
22
+ "llm.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00015.safetensors",
23
+ "llm.model.layers.1.mlp.up_proj.weight": "model-00001-of-00015.safetensors",
24
+ "llm.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00015.safetensors",
25
+ "llm.model.layers.1.self_attn.k_proj.bias": "model-00001-of-00015.safetensors",
26
+ "llm.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00015.safetensors",
27
+ "llm.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00015.safetensors",
28
+ "llm.model.layers.1.self_attn.q_proj.bias": "model-00001-of-00015.safetensors",
29
+ "llm.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00015.safetensors",
30
+ "llm.model.layers.1.self_attn.v_proj.bias": "model-00001-of-00015.safetensors",
31
+ "llm.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00015.safetensors",
32
+ "llm.model.layers.10.input_layernorm.weight": "model-00003-of-00015.safetensors",
33
+ "llm.model.layers.10.mlp.down_proj.weight": "model-00003-of-00015.safetensors",
34
+ "llm.model.layers.10.mlp.gate_proj.weight": "model-00003-of-00015.safetensors",
35
+ "llm.model.layers.10.mlp.up_proj.weight": "model-00003-of-00015.safetensors",
36
+ "llm.model.layers.10.post_attention_layernorm.weight": "model-00003-of-00015.safetensors",
37
+ "llm.model.layers.10.self_attn.k_proj.bias": "model-00003-of-00015.safetensors",
38
+ "llm.model.layers.10.self_attn.k_proj.weight": "model-00003-of-00015.safetensors",
39
+ "llm.model.layers.10.self_attn.o_proj.weight": "model-00003-of-00015.safetensors",
40
+ "llm.model.layers.10.self_attn.q_proj.bias": "model-00003-of-00015.safetensors",
41
+ "llm.model.layers.10.self_attn.q_proj.weight": "model-00003-of-00015.safetensors",
42
+ "llm.model.layers.10.self_attn.v_proj.bias": "model-00003-of-00015.safetensors",
43
+ "llm.model.layers.10.self_attn.v_proj.weight": "model-00003-of-00015.safetensors",
44
+ "llm.model.layers.11.input_layernorm.weight": "model-00003-of-00015.safetensors",
45
+ "llm.model.layers.11.mlp.down_proj.weight": "model-00003-of-00015.safetensors",
46
+ "llm.model.layers.11.mlp.gate_proj.weight": "model-00003-of-00015.safetensors",
47
+ "llm.model.layers.11.mlp.up_proj.weight": "model-00003-of-00015.safetensors",
48
+ "llm.model.layers.11.post_attention_layernorm.weight": "model-00003-of-00015.safetensors",
49
+ "llm.model.layers.11.self_attn.k_proj.bias": "model-00003-of-00015.safetensors",
50
+ "llm.model.layers.11.self_attn.k_proj.weight": "model-00003-of-00015.safetensors",
51
+ "llm.model.layers.11.self_attn.o_proj.weight": "model-00003-of-00015.safetensors",
52
+ "llm.model.layers.11.self_attn.q_proj.bias": "model-00003-of-00015.safetensors",
53
+ "llm.model.layers.11.self_attn.q_proj.weight": "model-00003-of-00015.safetensors",
54
+ "llm.model.layers.11.self_attn.v_proj.bias": "model-00003-of-00015.safetensors",
55
+ "llm.model.layers.11.self_attn.v_proj.weight": "model-00003-of-00015.safetensors",
56
+ "llm.model.layers.12.input_layernorm.weight": "model-00003-of-00015.safetensors",
57
+ "llm.model.layers.12.mlp.down_proj.weight": "model-00003-of-00015.safetensors",
58
+ "llm.model.layers.12.mlp.gate_proj.weight": "model-00003-of-00015.safetensors",
59
+ "llm.model.layers.12.mlp.up_proj.weight": "model-00003-of-00015.safetensors",
60
+ "llm.model.layers.12.post_attention_layernorm.weight": "model-00003-of-00015.safetensors",
61
+ "llm.model.layers.12.self_attn.k_proj.bias": "model-00003-of-00015.safetensors",
62
+ "llm.model.layers.12.self_attn.k_proj.weight": "model-00003-of-00015.safetensors",
63
+ "llm.model.layers.12.self_attn.o_proj.weight": "model-00003-of-00015.safetensors",
64
+ "llm.model.layers.12.self_attn.q_proj.bias": "model-00003-of-00015.safetensors",
65
+ "llm.model.layers.12.self_attn.q_proj.weight": "model-00003-of-00015.safetensors",
66
+ "llm.model.layers.12.self_attn.v_proj.bias": "model-00003-of-00015.safetensors",
67
+ "llm.model.layers.12.self_attn.v_proj.weight": "model-00003-of-00015.safetensors",
68
+ "llm.model.layers.13.input_layernorm.weight": "model-00004-of-00015.safetensors",
69
+ "llm.model.layers.13.mlp.down_proj.weight": "model-00004-of-00015.safetensors",
70
+ "llm.model.layers.13.mlp.gate_proj.weight": "model-00003-of-00015.safetensors",
71
+ "llm.model.layers.13.mlp.up_proj.weight": "model-00004-of-00015.safetensors",
72
+ "llm.model.layers.13.post_attention_layernorm.weight": "model-00004-of-00015.safetensors",
73
+ "llm.model.layers.13.self_attn.k_proj.bias": "model-00003-of-00015.safetensors",
74
+ "llm.model.layers.13.self_attn.k_proj.weight": "model-00003-of-00015.safetensors",
75
+ "llm.model.layers.13.self_attn.o_proj.weight": "model-00003-of-00015.safetensors",
76
+ "llm.model.layers.13.self_attn.q_proj.bias": "model-00003-of-00015.safetensors",
77
+ "llm.model.layers.13.self_attn.q_proj.weight": "model-00003-of-00015.safetensors",
78
+ "llm.model.layers.13.self_attn.v_proj.bias": "model-00003-of-00015.safetensors",
79
+ "llm.model.layers.13.self_attn.v_proj.weight": "model-00003-of-00015.safetensors",
80
+ "llm.model.layers.14.input_layernorm.weight": "model-00004-of-00015.safetensors",
81
+ "llm.model.layers.14.mlp.down_proj.weight": "model-00004-of-00015.safetensors",
82
+ "llm.model.layers.14.mlp.gate_proj.weight": "model-00004-of-00015.safetensors",
83
+ "llm.model.layers.14.mlp.up_proj.weight": "model-00004-of-00015.safetensors",
84
+ "llm.model.layers.14.post_attention_layernorm.weight": "model-00004-of-00015.safetensors",
85
+ "llm.model.layers.14.self_attn.k_proj.bias": "model-00004-of-00015.safetensors",
86
+ "llm.model.layers.14.self_attn.k_proj.weight": "model-00004-of-00015.safetensors",
87
+ "llm.model.layers.14.self_attn.o_proj.weight": "model-00004-of-00015.safetensors",
88
+ "llm.model.layers.14.self_attn.q_proj.bias": "model-00004-of-00015.safetensors",
89
+ "llm.model.layers.14.self_attn.q_proj.weight": "model-00004-of-00015.safetensors",
90
+ "llm.model.layers.14.self_attn.v_proj.bias": "model-00004-of-00015.safetensors",
91
+ "llm.model.layers.14.self_attn.v_proj.weight": "model-00004-of-00015.safetensors",
92
+ "llm.model.layers.15.input_layernorm.weight": "model-00004-of-00015.safetensors",
93
+ "llm.model.layers.15.mlp.down_proj.weight": "model-00004-of-00015.safetensors",
94
+ "llm.model.layers.15.mlp.gate_proj.weight": "model-00004-of-00015.safetensors",
95
+ "llm.model.layers.15.mlp.up_proj.weight": "model-00004-of-00015.safetensors",
96
+ "llm.model.layers.15.post_attention_layernorm.weight": "model-00004-of-00015.safetensors",
97
+ "llm.model.layers.15.self_attn.k_proj.bias": "model-00004-of-00015.safetensors",
98
+ "llm.model.layers.15.self_attn.k_proj.weight": "model-00004-of-00015.safetensors",
99
+ "llm.model.layers.15.self_attn.o_proj.weight": "model-00004-of-00015.safetensors",
100
+ "llm.model.layers.15.self_attn.q_proj.bias": "model-00004-of-00015.safetensors",
101
+ "llm.model.layers.15.self_attn.q_proj.weight": "model-00004-of-00015.safetensors",
102
+ "llm.model.layers.15.self_attn.v_proj.bias": "model-00004-of-00015.safetensors",
103
+ "llm.model.layers.15.self_attn.v_proj.weight": "model-00004-of-00015.safetensors",
104
+ "llm.model.layers.16.input_layernorm.weight": "model-00004-of-00015.safetensors",
105
+ "llm.model.layers.16.mlp.down_proj.weight": "model-00004-of-00015.safetensors",
106
+ "llm.model.layers.16.mlp.gate_proj.weight": "model-00004-of-00015.safetensors",
107
+ "llm.model.layers.16.mlp.up_proj.weight": "model-00004-of-00015.safetensors",
108
+ "llm.model.layers.16.post_attention_layernorm.weight": "model-00004-of-00015.safetensors",
109
+ "llm.model.layers.16.self_attn.k_proj.bias": "model-00004-of-00015.safetensors",
110
+ "llm.model.layers.16.self_attn.k_proj.weight": "model-00004-of-00015.safetensors",
111
+ "llm.model.layers.16.self_attn.o_proj.weight": "model-00004-of-00015.safetensors",
112
+ "llm.model.layers.16.self_attn.q_proj.bias": "model-00004-of-00015.safetensors",
113
+ "llm.model.layers.16.self_attn.q_proj.weight": "model-00004-of-00015.safetensors",
114
+ "llm.model.layers.16.self_attn.v_proj.bias": "model-00004-of-00015.safetensors",
115
+ "llm.model.layers.16.self_attn.v_proj.weight": "model-00004-of-00015.safetensors",
116
+ "llm.model.layers.17.input_layernorm.weight": "model-00004-of-00015.safetensors",
117
+ "llm.model.layers.17.mlp.down_proj.weight": "model-00004-of-00015.safetensors",
118
+ "llm.model.layers.17.mlp.gate_proj.weight": "model-00004-of-00015.safetensors",
119
+ "llm.model.layers.17.mlp.up_proj.weight": "model-00004-of-00015.safetensors",
120
+ "llm.model.layers.17.post_attention_layernorm.weight": "model-00004-of-00015.safetensors",
121
+ "llm.model.layers.17.self_attn.k_proj.bias": "model-00004-of-00015.safetensors",
122
+ "llm.model.layers.17.self_attn.k_proj.weight": "model-00004-of-00015.safetensors",
123
+ "llm.model.layers.17.self_attn.o_proj.weight": "model-00004-of-00015.safetensors",
124
+ "llm.model.layers.17.self_attn.q_proj.bias": "model-00004-of-00015.safetensors",
125
+ "llm.model.layers.17.self_attn.q_proj.weight": "model-00004-of-00015.safetensors",
126
+ "llm.model.layers.17.self_attn.v_proj.bias": "model-00004-of-00015.safetensors",
127
+ "llm.model.layers.17.self_attn.v_proj.weight": "model-00004-of-00015.safetensors",
128
+ "llm.model.layers.18.input_layernorm.weight": "model-00005-of-00015.safetensors",
129
+ "llm.model.layers.18.mlp.down_proj.weight": "model-00005-of-00015.safetensors",
130
+ "llm.model.layers.18.mlp.gate_proj.weight": "model-00004-of-00015.safetensors",
131
+ "llm.model.layers.18.mlp.up_proj.weight": "model-00005-of-00015.safetensors",
132
+ "llm.model.layers.18.post_attention_layernorm.weight": "model-00005-of-00015.safetensors",
133
+ "llm.model.layers.18.self_attn.k_proj.bias": "model-00004-of-00015.safetensors",
134
+ "llm.model.layers.18.self_attn.k_proj.weight": "model-00004-of-00015.safetensors",
135
+ "llm.model.layers.18.self_attn.o_proj.weight": "model-00004-of-00015.safetensors",
136
+ "llm.model.layers.18.self_attn.q_proj.bias": "model-00004-of-00015.safetensors",
137
+ "llm.model.layers.18.self_attn.q_proj.weight": "model-00004-of-00015.safetensors",
138
+ "llm.model.layers.18.self_attn.v_proj.bias": "model-00004-of-00015.safetensors",
139
+ "llm.model.layers.18.self_attn.v_proj.weight": "model-00004-of-00015.safetensors",
140
+ "llm.model.layers.19.input_layernorm.weight": "model-00005-of-00015.safetensors",
141
+ "llm.model.layers.19.mlp.down_proj.weight": "model-00005-of-00015.safetensors",
142
+ "llm.model.layers.19.mlp.gate_proj.weight": "model-00005-of-00015.safetensors",
143
+ "llm.model.layers.19.mlp.up_proj.weight": "model-00005-of-00015.safetensors",
144
+ "llm.model.layers.19.post_attention_layernorm.weight": "model-00005-of-00015.safetensors",
145
+ "llm.model.layers.19.self_attn.k_proj.bias": "model-00005-of-00015.safetensors",
146
+ "llm.model.layers.19.self_attn.k_proj.weight": "model-00005-of-00015.safetensors",
147
+ "llm.model.layers.19.self_attn.o_proj.weight": "model-00005-of-00015.safetensors",
148
+ "llm.model.layers.19.self_attn.q_proj.bias": "model-00005-of-00015.safetensors",
149
+ "llm.model.layers.19.self_attn.q_proj.weight": "model-00005-of-00015.safetensors",
150
+ "llm.model.layers.19.self_attn.v_proj.bias": "model-00005-of-00015.safetensors",
151
+ "llm.model.layers.19.self_attn.v_proj.weight": "model-00005-of-00015.safetensors",
152
+ "llm.model.layers.2.input_layernorm.weight": "model-00001-of-00015.safetensors",
153
+ "llm.model.layers.2.mlp.down_proj.weight": "model-00001-of-00015.safetensors",
154
+ "llm.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00015.safetensors",
155
+ "llm.model.layers.2.mlp.up_proj.weight": "model-00001-of-00015.safetensors",
156
+ "llm.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00015.safetensors",
157
+ "llm.model.layers.2.self_attn.k_proj.bias": "model-00001-of-00015.safetensors",
158
+ "llm.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00015.safetensors",
159
+ "llm.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00015.safetensors",
160
+ "llm.model.layers.2.self_attn.q_proj.bias": "model-00001-of-00015.safetensors",
161
+ "llm.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00015.safetensors",
162
+ "llm.model.layers.2.self_attn.v_proj.bias": "model-00001-of-00015.safetensors",
163
+ "llm.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00015.safetensors",
164
+ "llm.model.layers.20.input_layernorm.weight": "model-00005-of-00015.safetensors",
165
+ "llm.model.layers.20.mlp.down_proj.weight": "model-00005-of-00015.safetensors",
166
+ "llm.model.layers.20.mlp.gate_proj.weight": "model-00005-of-00015.safetensors",
167
+ "llm.model.layers.20.mlp.up_proj.weight": "model-00005-of-00015.safetensors",
168
+ "llm.model.layers.20.post_attention_layernorm.weight": "model-00005-of-00015.safetensors",
169
+ "llm.model.layers.20.self_attn.k_proj.bias": "model-00005-of-00015.safetensors",
170
+ "llm.model.layers.20.self_attn.k_proj.weight": "model-00005-of-00015.safetensors",
171
+ "llm.model.layers.20.self_attn.o_proj.weight": "model-00005-of-00015.safetensors",
172
+ "llm.model.layers.20.self_attn.q_proj.bias": "model-00005-of-00015.safetensors",
173
+ "llm.model.layers.20.self_attn.q_proj.weight": "model-00005-of-00015.safetensors",
174
+ "llm.model.layers.20.self_attn.v_proj.bias": "model-00005-of-00015.safetensors",
175
+ "llm.model.layers.20.self_attn.v_proj.weight": "model-00005-of-00015.safetensors",
176
+ "llm.model.layers.21.input_layernorm.weight": "model-00005-of-00015.safetensors",
177
+ "llm.model.layers.21.mlp.down_proj.weight": "model-00005-of-00015.safetensors",
178
+ "llm.model.layers.21.mlp.gate_proj.weight": "model-00005-of-00015.safetensors",
179
+ "llm.model.layers.21.mlp.up_proj.weight": "model-00005-of-00015.safetensors",
180
+ "llm.model.layers.21.post_attention_layernorm.weight": "model-00005-of-00015.safetensors",
181
+ "llm.model.layers.21.self_attn.k_proj.bias": "model-00005-of-00015.safetensors",
182
+ "llm.model.layers.21.self_attn.k_proj.weight": "model-00005-of-00015.safetensors",
183
+ "llm.model.layers.21.self_attn.o_proj.weight": "model-00005-of-00015.safetensors",
184
+ "llm.model.layers.21.self_attn.q_proj.bias": "model-00005-of-00015.safetensors",
185
+ "llm.model.layers.21.self_attn.q_proj.weight": "model-00005-of-00015.safetensors",
186
+ "llm.model.layers.21.self_attn.v_proj.bias": "model-00005-of-00015.safetensors",
187
+ "llm.model.layers.21.self_attn.v_proj.weight": "model-00005-of-00015.safetensors",
188
+ "llm.model.layers.22.input_layernorm.weight": "model-00005-of-00015.safetensors",
189
+ "llm.model.layers.22.mlp.down_proj.weight": "model-00005-of-00015.safetensors",
190
+ "llm.model.layers.22.mlp.gate_proj.weight": "model-00005-of-00015.safetensors",
191
+ "llm.model.layers.22.mlp.up_proj.weight": "model-00005-of-00015.safetensors",
192
+ "llm.model.layers.22.post_attention_layernorm.weight": "model-00005-of-00015.safetensors",
193
+ "llm.model.layers.22.self_attn.k_proj.bias": "model-00005-of-00015.safetensors",
194
+ "llm.model.layers.22.self_attn.k_proj.weight": "model-00005-of-00015.safetensors",
195
+ "llm.model.layers.22.self_attn.o_proj.weight": "model-00005-of-00015.safetensors",
196
+ "llm.model.layers.22.self_attn.q_proj.bias": "model-00005-of-00015.safetensors",
197
+ "llm.model.layers.22.self_attn.q_proj.weight": "model-00005-of-00015.safetensors",
198
+ "llm.model.layers.22.self_attn.v_proj.bias": "model-00005-of-00015.safetensors",
199
+ "llm.model.layers.22.self_attn.v_proj.weight": "model-00005-of-00015.safetensors",
200
+ "llm.model.layers.23.input_layernorm.weight": "model-00006-of-00015.safetensors",
201
+ "llm.model.layers.23.mlp.down_proj.weight": "model-00006-of-00015.safetensors",
202
+ "llm.model.layers.23.mlp.gate_proj.weight": "model-00005-of-00015.safetensors",
203
+ "llm.model.layers.23.mlp.up_proj.weight": "model-00006-of-00015.safetensors",
204
+ "llm.model.layers.23.post_attention_layernorm.weight": "model-00006-of-00015.safetensors",
205
+ "llm.model.layers.23.self_attn.k_proj.bias": "model-00005-of-00015.safetensors",
206
+ "llm.model.layers.23.self_attn.k_proj.weight": "model-00005-of-00015.safetensors",
207
+ "llm.model.layers.23.self_attn.o_proj.weight": "model-00005-of-00015.safetensors",
208
+ "llm.model.layers.23.self_attn.q_proj.bias": "model-00005-of-00015.safetensors",
209
+ "llm.model.layers.23.self_attn.q_proj.weight": "model-00005-of-00015.safetensors",
210
+ "llm.model.layers.23.self_attn.v_proj.bias": "model-00005-of-00015.safetensors",
211
+ "llm.model.layers.23.self_attn.v_proj.weight": "model-00005-of-00015.safetensors",
212
+ "llm.model.layers.24.input_layernorm.weight": "model-00006-of-00015.safetensors",
213
+ "llm.model.layers.24.mlp.down_proj.weight": "model-00006-of-00015.safetensors",
214
+ "llm.model.layers.24.mlp.gate_proj.weight": "model-00006-of-00015.safetensors",
215
+ "llm.model.layers.24.mlp.up_proj.weight": "model-00006-of-00015.safetensors",
216
+ "llm.model.layers.24.post_attention_layernorm.weight": "model-00006-of-00015.safetensors",
217
+ "llm.model.layers.24.self_attn.k_proj.bias": "model-00006-of-00015.safetensors",
218
+ "llm.model.layers.24.self_attn.k_proj.weight": "model-00006-of-00015.safetensors",
219
+ "llm.model.layers.24.self_attn.o_proj.weight": "model-00006-of-00015.safetensors",
220
+ "llm.model.layers.24.self_attn.q_proj.bias": "model-00006-of-00015.safetensors",
221
+ "llm.model.layers.24.self_attn.q_proj.weight": "model-00006-of-00015.safetensors",
222
+ "llm.model.layers.24.self_attn.v_proj.bias": "model-00006-of-00015.safetensors",
223
+ "llm.model.layers.24.self_attn.v_proj.weight": "model-00006-of-00015.safetensors",
224
+ "llm.model.layers.25.input_layernorm.weight": "model-00006-of-00015.safetensors",
225
+ "llm.model.layers.25.mlp.down_proj.weight": "model-00006-of-00015.safetensors",
226
+ "llm.model.layers.25.mlp.gate_proj.weight": "model-00006-of-00015.safetensors",
227
+ "llm.model.layers.25.mlp.up_proj.weight": "model-00006-of-00015.safetensors",
228
+ "llm.model.layers.25.post_attention_layernorm.weight": "model-00006-of-00015.safetensors",
229
+ "llm.model.layers.25.self_attn.k_proj.bias": "model-00006-of-00015.safetensors",
230
+ "llm.model.layers.25.self_attn.k_proj.weight": "model-00006-of-00015.safetensors",
231
+ "llm.model.layers.25.self_attn.o_proj.weight": "model-00006-of-00015.safetensors",
232
+ "llm.model.layers.25.self_attn.q_proj.bias": "model-00006-of-00015.safetensors",
233
+ "llm.model.layers.25.self_attn.q_proj.weight": "model-00006-of-00015.safetensors",
234
+ "llm.model.layers.25.self_attn.v_proj.bias": "model-00006-of-00015.safetensors",
235
+ "llm.model.layers.25.self_attn.v_proj.weight": "model-00006-of-00015.safetensors",
236
+ "llm.model.layers.26.input_layernorm.weight": "model-00006-of-00015.safetensors",
237
+ "llm.model.layers.26.mlp.down_proj.weight": "model-00006-of-00015.safetensors",
238
+ "llm.model.layers.26.mlp.gate_proj.weight": "model-00006-of-00015.safetensors",
239
+ "llm.model.layers.26.mlp.up_proj.weight": "model-00006-of-00015.safetensors",
240
+ "llm.model.layers.26.post_attention_layernorm.weight": "model-00006-of-00015.safetensors",
241
+ "llm.model.layers.26.self_attn.k_proj.bias": "model-00006-of-00015.safetensors",
242
+ "llm.model.layers.26.self_attn.k_proj.weight": "model-00006-of-00015.safetensors",
243
+ "llm.model.layers.26.self_attn.o_proj.weight": "model-00006-of-00015.safetensors",
244
+ "llm.model.layers.26.self_attn.q_proj.bias": "model-00006-of-00015.safetensors",
245
+ "llm.model.layers.26.self_attn.q_proj.weight": "model-00006-of-00015.safetensors",
246
+ "llm.model.layers.26.self_attn.v_proj.bias": "model-00006-of-00015.safetensors",
247
+ "llm.model.layers.26.self_attn.v_proj.weight": "model-00006-of-00015.safetensors",
248
+ "llm.model.layers.27.input_layernorm.weight": "model-00006-of-00015.safetensors",
249
+ "llm.model.layers.27.mlp.down_proj.weight": "model-00006-of-00015.safetensors",
250
+ "llm.model.layers.27.mlp.gate_proj.weight": "model-00006-of-00015.safetensors",
251
+ "llm.model.layers.27.mlp.up_proj.weight": "model-00006-of-00015.safetensors",
252
+ "llm.model.layers.27.post_attention_layernorm.weight": "model-00006-of-00015.safetensors",
253
+ "llm.model.layers.27.self_attn.k_proj.bias": "model-00006-of-00015.safetensors",
254
+ "llm.model.layers.27.self_attn.k_proj.weight": "model-00006-of-00015.safetensors",
255
+ "llm.model.layers.27.self_attn.o_proj.weight": "model-00006-of-00015.safetensors",
256
+ "llm.model.layers.27.self_attn.q_proj.bias": "model-00006-of-00015.safetensors",
257
+ "llm.model.layers.27.self_attn.q_proj.weight": "model-00006-of-00015.safetensors",
258
+ "llm.model.layers.27.self_attn.v_proj.bias": "model-00006-of-00015.safetensors",
259
+ "llm.model.layers.27.self_attn.v_proj.weight": "model-00006-of-00015.safetensors",
260
+ "llm.model.layers.28.input_layernorm.weight": "model-00007-of-00015.safetensors",
261
+ "llm.model.layers.28.mlp.down_proj.weight": "model-00007-of-00015.safetensors",
262
+ "llm.model.layers.28.mlp.gate_proj.weight": "model-00006-of-00015.safetensors",
263
+ "llm.model.layers.28.mlp.up_proj.weight": "model-00007-of-00015.safetensors",
264
+ "llm.model.layers.28.post_attention_layernorm.weight": "model-00007-of-00015.safetensors",
265
+ "llm.model.layers.28.self_attn.k_proj.bias": "model-00006-of-00015.safetensors",
266
+ "llm.model.layers.28.self_attn.k_proj.weight": "model-00006-of-00015.safetensors",
267
+ "llm.model.layers.28.self_attn.o_proj.weight": "model-00006-of-00015.safetensors",
268
+ "llm.model.layers.28.self_attn.q_proj.bias": "model-00006-of-00015.safetensors",
269
+ "llm.model.layers.28.self_attn.q_proj.weight": "model-00006-of-00015.safetensors",
270
+ "llm.model.layers.28.self_attn.v_proj.bias": "model-00006-of-00015.safetensors",
271
+ "llm.model.layers.28.self_attn.v_proj.weight": "model-00006-of-00015.safetensors",
272
+ "llm.model.layers.29.input_layernorm.weight": "model-00007-of-00015.safetensors",
273
+ "llm.model.layers.29.mlp.down_proj.weight": "model-00007-of-00015.safetensors",
274
+ "llm.model.layers.29.mlp.gate_proj.weight": "model-00007-of-00015.safetensors",
275
+ "llm.model.layers.29.mlp.up_proj.weight": "model-00007-of-00015.safetensors",
276
+ "llm.model.layers.29.post_attention_layernorm.weight": "model-00007-of-00015.safetensors",
277
+ "llm.model.layers.29.self_attn.k_proj.bias": "model-00007-of-00015.safetensors",
278
+ "llm.model.layers.29.self_attn.k_proj.weight": "model-00007-of-00015.safetensors",
279
+ "llm.model.layers.29.self_attn.o_proj.weight": "model-00007-of-00015.safetensors",
280
+ "llm.model.layers.29.self_attn.q_proj.bias": "model-00007-of-00015.safetensors",
281
+ "llm.model.layers.29.self_attn.q_proj.weight": "model-00007-of-00015.safetensors",
282
+ "llm.model.layers.29.self_attn.v_proj.bias": "model-00007-of-00015.safetensors",
283
+ "llm.model.layers.29.self_attn.v_proj.weight": "model-00007-of-00015.safetensors",
284
+ "llm.model.layers.3.input_layernorm.weight": "model-00002-of-00015.safetensors",
285
+ "llm.model.layers.3.mlp.down_proj.weight": "model-00002-of-00015.safetensors",
286
+ "llm.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00015.safetensors",
287
+ "llm.model.layers.3.mlp.up_proj.weight": "model-00002-of-00015.safetensors",
288
+ "llm.model.layers.3.post_attention_layernorm.weight": "model-00002-of-00015.safetensors",
289
+ "llm.model.layers.3.self_attn.k_proj.bias": "model-00001-of-00015.safetensors",
290
+ "llm.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00015.safetensors",
291
+ "llm.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00015.safetensors",
292
+ "llm.model.layers.3.self_attn.q_proj.bias": "model-00001-of-00015.safetensors",
293
+ "llm.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00015.safetensors",
294
+ "llm.model.layers.3.self_attn.v_proj.bias": "model-00001-of-00015.safetensors",
295
+ "llm.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00015.safetensors",
296
+ "llm.model.layers.30.input_layernorm.weight": "model-00007-of-00015.safetensors",
297
+ "llm.model.layers.30.mlp.down_proj.weight": "model-00007-of-00015.safetensors",
298
+ "llm.model.layers.30.mlp.gate_proj.weight": "model-00007-of-00015.safetensors",
299
+ "llm.model.layers.30.mlp.up_proj.weight": "model-00007-of-00015.safetensors",
300
+ "llm.model.layers.30.post_attention_layernorm.weight": "model-00007-of-00015.safetensors",
301
+ "llm.model.layers.30.self_attn.k_proj.bias": "model-00007-of-00015.safetensors",
302
+ "llm.model.layers.30.self_attn.k_proj.weight": "model-00007-of-00015.safetensors",
303
+ "llm.model.layers.30.self_attn.o_proj.weight": "model-00007-of-00015.safetensors",
304
+ "llm.model.layers.30.self_attn.q_proj.bias": "model-00007-of-00015.safetensors",
305
+ "llm.model.layers.30.self_attn.q_proj.weight": "model-00007-of-00015.safetensors",
306
+ "llm.model.layers.30.self_attn.v_proj.bias": "model-00007-of-00015.safetensors",
307
+ "llm.model.layers.30.self_attn.v_proj.weight": "model-00007-of-00015.safetensors",
308
+ "llm.model.layers.31.input_layernorm.weight": "model-00007-of-00015.safetensors",
309
+ "llm.model.layers.31.mlp.down_proj.weight": "model-00007-of-00015.safetensors",
310
+ "llm.model.layers.31.mlp.gate_proj.weight": "model-00007-of-00015.safetensors",
311
+ "llm.model.layers.31.mlp.up_proj.weight": "model-00007-of-00015.safetensors",
312
+ "llm.model.layers.31.post_attention_layernorm.weight": "model-00007-of-00015.safetensors",
313
+ "llm.model.layers.31.self_attn.k_proj.bias": "model-00007-of-00015.safetensors",
314
+ "llm.model.layers.31.self_attn.k_proj.weight": "model-00007-of-00015.safetensors",
315
+ "llm.model.layers.31.self_attn.o_proj.weight": "model-00007-of-00015.safetensors",
316
+ "llm.model.layers.31.self_attn.q_proj.bias": "model-00007-of-00015.safetensors",
317
+ "llm.model.layers.31.self_attn.q_proj.weight": "model-00007-of-00015.safetensors",
318
+ "llm.model.layers.31.self_attn.v_proj.bias": "model-00007-of-00015.safetensors",
319
+ "llm.model.layers.31.self_attn.v_proj.weight": "model-00007-of-00015.safetensors",
320
+ "llm.model.layers.32.input_layernorm.weight": "model-00007-of-00015.safetensors",
321
+ "llm.model.layers.32.mlp.down_proj.weight": "model-00007-of-00015.safetensors",
322
+ "llm.model.layers.32.mlp.gate_proj.weight": "model-00007-of-00015.safetensors",
323
+ "llm.model.layers.32.mlp.up_proj.weight": "model-00007-of-00015.safetensors",
324
+ "llm.model.layers.32.post_attention_layernorm.weight": "model-00007-of-00015.safetensors",
325
+ "llm.model.layers.32.self_attn.k_proj.bias": "model-00007-of-00015.safetensors",
326
+ "llm.model.layers.32.self_attn.k_proj.weight": "model-00007-of-00015.safetensors",
327
+ "llm.model.layers.32.self_attn.o_proj.weight": "model-00007-of-00015.safetensors",
328
+ "llm.model.layers.32.self_attn.q_proj.bias": "model-00007-of-00015.safetensors",
329
+ "llm.model.layers.32.self_attn.q_proj.weight": "model-00007-of-00015.safetensors",
330
+ "llm.model.layers.32.self_attn.v_proj.bias": "model-00007-of-00015.safetensors",
331
+ "llm.model.layers.32.self_attn.v_proj.weight": "model-00007-of-00015.safetensors",
332
+ "llm.model.layers.33.input_layernorm.weight": "model-00008-of-00015.safetensors",
333
+ "llm.model.layers.33.mlp.down_proj.weight": "model-00008-of-00015.safetensors",
334
+ "llm.model.layers.33.mlp.gate_proj.weight": "model-00007-of-00015.safetensors",
335
+ "llm.model.layers.33.mlp.up_proj.weight": "model-00008-of-00015.safetensors",
336
+ "llm.model.layers.33.post_attention_layernorm.weight": "model-00008-of-00015.safetensors",
337
+ "llm.model.layers.33.self_attn.k_proj.bias": "model-00007-of-00015.safetensors",
338
+ "llm.model.layers.33.self_attn.k_proj.weight": "model-00007-of-00015.safetensors",
339
+ "llm.model.layers.33.self_attn.o_proj.weight": "model-00007-of-00015.safetensors",
340
+ "llm.model.layers.33.self_attn.q_proj.bias": "model-00007-of-00015.safetensors",
341
+ "llm.model.layers.33.self_attn.q_proj.weight": "model-00007-of-00015.safetensors",
342
+ "llm.model.layers.33.self_attn.v_proj.bias": "model-00007-of-00015.safetensors",
343
+ "llm.model.layers.33.self_attn.v_proj.weight": "model-00007-of-00015.safetensors",
344
+ "llm.model.layers.34.input_layernorm.weight": "model-00008-of-00015.safetensors",
345
+ "llm.model.layers.34.mlp.down_proj.weight": "model-00008-of-00015.safetensors",
346
+ "llm.model.layers.34.mlp.gate_proj.weight": "model-00008-of-00015.safetensors",
347
+ "llm.model.layers.34.mlp.up_proj.weight": "model-00008-of-00015.safetensors",
348
+ "llm.model.layers.34.post_attention_layernorm.weight": "model-00008-of-00015.safetensors",
349
+ "llm.model.layers.34.self_attn.k_proj.bias": "model-00008-of-00015.safetensors",
350
+ "llm.model.layers.34.self_attn.k_proj.weight": "model-00008-of-00015.safetensors",
351
+ "llm.model.layers.34.self_attn.o_proj.weight": "model-00008-of-00015.safetensors",
352
+ "llm.model.layers.34.self_attn.q_proj.bias": "model-00008-of-00015.safetensors",
353
+ "llm.model.layers.34.self_attn.q_proj.weight": "model-00008-of-00015.safetensors",
354
+ "llm.model.layers.34.self_attn.v_proj.bias": "model-00008-of-00015.safetensors",
355
+ "llm.model.layers.34.self_attn.v_proj.weight": "model-00008-of-00015.safetensors",
356
+ "llm.model.layers.35.input_layernorm.weight": "model-00008-of-00015.safetensors",
357
+ "llm.model.layers.35.mlp.down_proj.weight": "model-00008-of-00015.safetensors",
358
+ "llm.model.layers.35.mlp.gate_proj.weight": "model-00008-of-00015.safetensors",
359
+ "llm.model.layers.35.mlp.up_proj.weight": "model-00008-of-00015.safetensors",
360
+ "llm.model.layers.35.post_attention_layernorm.weight": "model-00008-of-00015.safetensors",
361
+ "llm.model.layers.35.self_attn.k_proj.bias": "model-00008-of-00015.safetensors",
362
+ "llm.model.layers.35.self_attn.k_proj.weight": "model-00008-of-00015.safetensors",
363
+ "llm.model.layers.35.self_attn.o_proj.weight": "model-00008-of-00015.safetensors",
364
+ "llm.model.layers.35.self_attn.q_proj.bias": "model-00008-of-00015.safetensors",
365
+ "llm.model.layers.35.self_attn.q_proj.weight": "model-00008-of-00015.safetensors",
366
+ "llm.model.layers.35.self_attn.v_proj.bias": "model-00008-of-00015.safetensors",
367
+ "llm.model.layers.35.self_attn.v_proj.weight": "model-00008-of-00015.safetensors",
368
+ "llm.model.layers.36.input_layernorm.weight": "model-00008-of-00015.safetensors",
369
+ "llm.model.layers.36.mlp.down_proj.weight": "model-00008-of-00015.safetensors",
370
+ "llm.model.layers.36.mlp.gate_proj.weight": "model-00008-of-00015.safetensors",
371
+ "llm.model.layers.36.mlp.up_proj.weight": "model-00008-of-00015.safetensors",
372
+ "llm.model.layers.36.post_attention_layernorm.weight": "model-00008-of-00015.safetensors",
373
+ "llm.model.layers.36.self_attn.k_proj.bias": "model-00008-of-00015.safetensors",
374
+ "llm.model.layers.36.self_attn.k_proj.weight": "model-00008-of-00015.safetensors",
375
+ "llm.model.layers.36.self_attn.o_proj.weight": "model-00008-of-00015.safetensors",
376
+ "llm.model.layers.36.self_attn.q_proj.bias": "model-00008-of-00015.safetensors",
377
+ "llm.model.layers.36.self_attn.q_proj.weight": "model-00008-of-00015.safetensors",
378
+ "llm.model.layers.36.self_attn.v_proj.bias": "model-00008-of-00015.safetensors",
379
+ "llm.model.layers.36.self_attn.v_proj.weight": "model-00008-of-00015.safetensors",
380
+ "llm.model.layers.37.input_layernorm.weight": "model-00008-of-00015.safetensors",
381
+ "llm.model.layers.37.mlp.down_proj.weight": "model-00008-of-00015.safetensors",
382
+ "llm.model.layers.37.mlp.gate_proj.weight": "model-00008-of-00015.safetensors",
383
+ "llm.model.layers.37.mlp.up_proj.weight": "model-00008-of-00015.safetensors",
384
+ "llm.model.layers.37.post_attention_layernorm.weight": "model-00008-of-00015.safetensors",
385
+ "llm.model.layers.37.self_attn.k_proj.bias": "model-00008-of-00015.safetensors",
386
+ "llm.model.layers.37.self_attn.k_proj.weight": "model-00008-of-00015.safetensors",
387
+ "llm.model.layers.37.self_attn.o_proj.weight": "model-00008-of-00015.safetensors",
388
+ "llm.model.layers.37.self_attn.q_proj.bias": "model-00008-of-00015.safetensors",
389
+ "llm.model.layers.37.self_attn.q_proj.weight": "model-00008-of-00015.safetensors",
390
+ "llm.model.layers.37.self_attn.v_proj.bias": "model-00008-of-00015.safetensors",
391
+ "llm.model.layers.37.self_attn.v_proj.weight": "model-00008-of-00015.safetensors",
392
+ "llm.model.layers.38.input_layernorm.weight": "model-00009-of-00015.safetensors",
393
+ "llm.model.layers.38.mlp.down_proj.weight": "model-00009-of-00015.safetensors",
394
+ "llm.model.layers.38.mlp.gate_proj.weight": "model-00008-of-00015.safetensors",
395
+ "llm.model.layers.38.mlp.up_proj.weight": "model-00009-of-00015.safetensors",
396
+ "llm.model.layers.38.post_attention_layernorm.weight": "model-00009-of-00015.safetensors",
397
+ "llm.model.layers.38.self_attn.k_proj.bias": "model-00008-of-00015.safetensors",
398
+ "llm.model.layers.38.self_attn.k_proj.weight": "model-00008-of-00015.safetensors",
399
+ "llm.model.layers.38.self_attn.o_proj.weight": "model-00008-of-00015.safetensors",
400
+ "llm.model.layers.38.self_attn.q_proj.bias": "model-00008-of-00015.safetensors",
401
+ "llm.model.layers.38.self_attn.q_proj.weight": "model-00008-of-00015.safetensors",
402
+ "llm.model.layers.38.self_attn.v_proj.bias": "model-00008-of-00015.safetensors",
403
+ "llm.model.layers.38.self_attn.v_proj.weight": "model-00008-of-00015.safetensors",
404
+ "llm.model.layers.39.input_layernorm.weight": "model-00009-of-00015.safetensors",
405
+ "llm.model.layers.39.mlp.down_proj.weight": "model-00009-of-00015.safetensors",
406
+ "llm.model.layers.39.mlp.gate_proj.weight": "model-00009-of-00015.safetensors",
407
+ "llm.model.layers.39.mlp.up_proj.weight": "model-00009-of-00015.safetensors",
408
+ "llm.model.layers.39.post_attention_layernorm.weight": "model-00009-of-00015.safetensors",
409
+ "llm.model.layers.39.self_attn.k_proj.bias": "model-00009-of-00015.safetensors",
410
+ "llm.model.layers.39.self_attn.k_proj.weight": "model-00009-of-00015.safetensors",
411
+ "llm.model.layers.39.self_attn.o_proj.weight": "model-00009-of-00015.safetensors",
412
+ "llm.model.layers.39.self_attn.q_proj.bias": "model-00009-of-00015.safetensors",
413
+ "llm.model.layers.39.self_attn.q_proj.weight": "model-00009-of-00015.safetensors",
414
+ "llm.model.layers.39.self_attn.v_proj.bias": "model-00009-of-00015.safetensors",
415
+ "llm.model.layers.39.self_attn.v_proj.weight": "model-00009-of-00015.safetensors",
416
+ "llm.model.layers.4.input_layernorm.weight": "model-00002-of-00015.safetensors",
417
+ "llm.model.layers.4.mlp.down_proj.weight": "model-00002-of-00015.safetensors",
418
+ "llm.model.layers.4.mlp.gate_proj.weight": "model-00002-of-00015.safetensors",
419
+ "llm.model.layers.4.mlp.up_proj.weight": "model-00002-of-00015.safetensors",
420
+ "llm.model.layers.4.post_attention_layernorm.weight": "model-00002-of-00015.safetensors",
421
+ "llm.model.layers.4.self_attn.k_proj.bias": "model-00002-of-00015.safetensors",
422
+ "llm.model.layers.4.self_attn.k_proj.weight": "model-00002-of-00015.safetensors",
423
+ "llm.model.layers.4.self_attn.o_proj.weight": "model-00002-of-00015.safetensors",
424
+ "llm.model.layers.4.self_attn.q_proj.bias": "model-00002-of-00015.safetensors",
425
+ "llm.model.layers.4.self_attn.q_proj.weight": "model-00002-of-00015.safetensors",
426
+ "llm.model.layers.4.self_attn.v_proj.bias": "model-00002-of-00015.safetensors",
427
+ "llm.model.layers.4.self_attn.v_proj.weight": "model-00002-of-00015.safetensors",
428
+ "llm.model.layers.40.input_layernorm.weight": "model-00009-of-00015.safetensors",
429
+ "llm.model.layers.40.mlp.down_proj.weight": "model-00009-of-00015.safetensors",
430
+ "llm.model.layers.40.mlp.gate_proj.weight": "model-00009-of-00015.safetensors",
431
+ "llm.model.layers.40.mlp.up_proj.weight": "model-00009-of-00015.safetensors",
432
+ "llm.model.layers.40.post_attention_layernorm.weight": "model-00009-of-00015.safetensors",
433
+ "llm.model.layers.40.self_attn.k_proj.bias": "model-00009-of-00015.safetensors",
434
+ "llm.model.layers.40.self_attn.k_proj.weight": "model-00009-of-00015.safetensors",
435
+ "llm.model.layers.40.self_attn.o_proj.weight": "model-00009-of-00015.safetensors",
436
+ "llm.model.layers.40.self_attn.q_proj.bias": "model-00009-of-00015.safetensors",
437
+ "llm.model.layers.40.self_attn.q_proj.weight": "model-00009-of-00015.safetensors",
438
+ "llm.model.layers.40.self_attn.v_proj.bias": "model-00009-of-00015.safetensors",
439
+ "llm.model.layers.40.self_attn.v_proj.weight": "model-00009-of-00015.safetensors",
440
+ "llm.model.layers.41.input_layernorm.weight": "model-00009-of-00015.safetensors",
441
+ "llm.model.layers.41.mlp.down_proj.weight": "model-00009-of-00015.safetensors",
442
+ "llm.model.layers.41.mlp.gate_proj.weight": "model-00009-of-00015.safetensors",
443
+ "llm.model.layers.41.mlp.up_proj.weight": "model-00009-of-00015.safetensors",
444
+ "llm.model.layers.41.post_attention_layernorm.weight": "model-00009-of-00015.safetensors",
445
+ "llm.model.layers.41.self_attn.k_proj.bias": "model-00009-of-00015.safetensors",
446
+ "llm.model.layers.41.self_attn.k_proj.weight": "model-00009-of-00015.safetensors",
447
+ "llm.model.layers.41.self_attn.o_proj.weight": "model-00009-of-00015.safetensors",
448
+ "llm.model.layers.41.self_attn.q_proj.bias": "model-00009-of-00015.safetensors",
449
+ "llm.model.layers.41.self_attn.q_proj.weight": "model-00009-of-00015.safetensors",
450
+ "llm.model.layers.41.self_attn.v_proj.bias": "model-00009-of-00015.safetensors",
451
+ "llm.model.layers.41.self_attn.v_proj.weight": "model-00009-of-00015.safetensors",
452
+ "llm.model.layers.42.input_layernorm.weight": "model-00009-of-00015.safetensors",
453
+ "llm.model.layers.42.mlp.down_proj.weight": "model-00009-of-00015.safetensors",
454
+ "llm.model.layers.42.mlp.gate_proj.weight": "model-00009-of-00015.safetensors",
455
+ "llm.model.layers.42.mlp.up_proj.weight": "model-00009-of-00015.safetensors",
456
+ "llm.model.layers.42.post_attention_layernorm.weight": "model-00009-of-00015.safetensors",
457
+ "llm.model.layers.42.self_attn.k_proj.bias": "model-00009-of-00015.safetensors",
458
+ "llm.model.layers.42.self_attn.k_proj.weight": "model-00009-of-00015.safetensors",
459
+ "llm.model.layers.42.self_attn.o_proj.weight": "model-00009-of-00015.safetensors",
460
+ "llm.model.layers.42.self_attn.q_proj.bias": "model-00009-of-00015.safetensors",
461
+ "llm.model.layers.42.self_attn.q_proj.weight": "model-00009-of-00015.safetensors",
462
+ "llm.model.layers.42.self_attn.v_proj.bias": "model-00009-of-00015.safetensors",
463
+ "llm.model.layers.42.self_attn.v_proj.weight": "model-00009-of-00015.safetensors",
464
+ "llm.model.layers.43.input_layernorm.weight": "model-00010-of-00015.safetensors",
465
+ "llm.model.layers.43.mlp.down_proj.weight": "model-00010-of-00015.safetensors",
466
+ "llm.model.layers.43.mlp.gate_proj.weight": "model-00009-of-00015.safetensors",
467
+ "llm.model.layers.43.mlp.up_proj.weight": "model-00010-of-00015.safetensors",
468
+ "llm.model.layers.43.post_attention_layernorm.weight": "model-00010-of-00015.safetensors",
469
+ "llm.model.layers.43.self_attn.k_proj.bias": "model-00009-of-00015.safetensors",
470
+ "llm.model.layers.43.self_attn.k_proj.weight": "model-00009-of-00015.safetensors",
471
+ "llm.model.layers.43.self_attn.o_proj.weight": "model-00009-of-00015.safetensors",
472
+ "llm.model.layers.43.self_attn.q_proj.bias": "model-00009-of-00015.safetensors",
473
+ "llm.model.layers.43.self_attn.q_proj.weight": "model-00009-of-00015.safetensors",
474
+ "llm.model.layers.43.self_attn.v_proj.bias": "model-00009-of-00015.safetensors",
475
+ "llm.model.layers.43.self_attn.v_proj.weight": "model-00009-of-00015.safetensors",
476
+ "llm.model.layers.44.input_layernorm.weight": "model-00010-of-00015.safetensors",
477
+ "llm.model.layers.44.mlp.down_proj.weight": "model-00010-of-00015.safetensors",
478
+ "llm.model.layers.44.mlp.gate_proj.weight": "model-00010-of-00015.safetensors",
479
+ "llm.model.layers.44.mlp.up_proj.weight": "model-00010-of-00015.safetensors",
480
+ "llm.model.layers.44.post_attention_layernorm.weight": "model-00010-of-00015.safetensors",
481
+ "llm.model.layers.44.self_attn.k_proj.bias": "model-00010-of-00015.safetensors",
482
+ "llm.model.layers.44.self_attn.k_proj.weight": "model-00010-of-00015.safetensors",
483
+ "llm.model.layers.44.self_attn.o_proj.weight": "model-00010-of-00015.safetensors",
484
+ "llm.model.layers.44.self_attn.q_proj.bias": "model-00010-of-00015.safetensors",
485
+ "llm.model.layers.44.self_attn.q_proj.weight": "model-00010-of-00015.safetensors",
486
+ "llm.model.layers.44.self_attn.v_proj.bias": "model-00010-of-00015.safetensors",
487
+ "llm.model.layers.44.self_attn.v_proj.weight": "model-00010-of-00015.safetensors",
488
+ "llm.model.layers.45.input_layernorm.weight": "model-00010-of-00015.safetensors",
489
+ "llm.model.layers.45.mlp.down_proj.weight": "model-00010-of-00015.safetensors",
490
+ "llm.model.layers.45.mlp.gate_proj.weight": "model-00010-of-00015.safetensors",
491
+ "llm.model.layers.45.mlp.up_proj.weight": "model-00010-of-00015.safetensors",
492
+ "llm.model.layers.45.post_attention_layernorm.weight": "model-00010-of-00015.safetensors",
493
+ "llm.model.layers.45.self_attn.k_proj.bias": "model-00010-of-00015.safetensors",
494
+ "llm.model.layers.45.self_attn.k_proj.weight": "model-00010-of-00015.safetensors",
495
+ "llm.model.layers.45.self_attn.o_proj.weight": "model-00010-of-00015.safetensors",
496
+ "llm.model.layers.45.self_attn.q_proj.bias": "model-00010-of-00015.safetensors",
497
+ "llm.model.layers.45.self_attn.q_proj.weight": "model-00010-of-00015.safetensors",
498
+ "llm.model.layers.45.self_attn.v_proj.bias": "model-00010-of-00015.safetensors",
499
+ "llm.model.layers.45.self_attn.v_proj.weight": "model-00010-of-00015.safetensors",
500
+ "llm.model.layers.46.input_layernorm.weight": "model-00010-of-00015.safetensors",
501
+ "llm.model.layers.46.mlp.down_proj.weight": "model-00010-of-00015.safetensors",
502
+ "llm.model.layers.46.mlp.gate_proj.weight": "model-00010-of-00015.safetensors",
503
+ "llm.model.layers.46.mlp.up_proj.weight": "model-00010-of-00015.safetensors",
504
+ "llm.model.layers.46.post_attention_layernorm.weight": "model-00010-of-00015.safetensors",
505
+ "llm.model.layers.46.self_attn.k_proj.bias": "model-00010-of-00015.safetensors",
506
+ "llm.model.layers.46.self_attn.k_proj.weight": "model-00010-of-00015.safetensors",
507
+ "llm.model.layers.46.self_attn.o_proj.weight": "model-00010-of-00015.safetensors",
508
+ "llm.model.layers.46.self_attn.q_proj.bias": "model-00010-of-00015.safetensors",
509
+ "llm.model.layers.46.self_attn.q_proj.weight": "model-00010-of-00015.safetensors",
510
+ "llm.model.layers.46.self_attn.v_proj.bias": "model-00010-of-00015.safetensors",
511
+ "llm.model.layers.46.self_attn.v_proj.weight": "model-00010-of-00015.safetensors",
512
+ "llm.model.layers.47.input_layernorm.weight": "model-00010-of-00015.safetensors",
513
+ "llm.model.layers.47.mlp.down_proj.weight": "model-00010-of-00015.safetensors",
514
+ "llm.model.layers.47.mlp.gate_proj.weight": "model-00010-of-00015.safetensors",
515
+ "llm.model.layers.47.mlp.up_proj.weight": "model-00010-of-00015.safetensors",
516
+ "llm.model.layers.47.post_attention_layernorm.weight": "model-00010-of-00015.safetensors",
517
+ "llm.model.layers.47.self_attn.k_proj.bias": "model-00010-of-00015.safetensors",
518
+ "llm.model.layers.47.self_attn.k_proj.weight": "model-00010-of-00015.safetensors",
519
+ "llm.model.layers.47.self_attn.o_proj.weight": "model-00010-of-00015.safetensors",
520
+ "llm.model.layers.47.self_attn.q_proj.bias": "model-00010-of-00015.safetensors",
521
+ "llm.model.layers.47.self_attn.q_proj.weight": "model-00010-of-00015.safetensors",
522
+ "llm.model.layers.47.self_attn.v_proj.bias": "model-00010-of-00015.safetensors",
523
+ "llm.model.layers.47.self_attn.v_proj.weight": "model-00010-of-00015.safetensors",
524
+ "llm.model.layers.48.input_layernorm.weight": "model-00011-of-00015.safetensors",
525
+ "llm.model.layers.48.mlp.down_proj.weight": "model-00011-of-00015.safetensors",
526
+ "llm.model.layers.48.mlp.gate_proj.weight": "model-00010-of-00015.safetensors",
527
+ "llm.model.layers.48.mlp.up_proj.weight": "model-00011-of-00015.safetensors",
528
+ "llm.model.layers.48.post_attention_layernorm.weight": "model-00011-of-00015.safetensors",
529
+ "llm.model.layers.48.self_attn.k_proj.bias": "model-00010-of-00015.safetensors",
530
+ "llm.model.layers.48.self_attn.k_proj.weight": "model-00010-of-00015.safetensors",
531
+ "llm.model.layers.48.self_attn.o_proj.weight": "model-00010-of-00015.safetensors",
532
+ "llm.model.layers.48.self_attn.q_proj.bias": "model-00010-of-00015.safetensors",
533
+ "llm.model.layers.48.self_attn.q_proj.weight": "model-00010-of-00015.safetensors",
534
+ "llm.model.layers.48.self_attn.v_proj.bias": "model-00010-of-00015.safetensors",
535
+ "llm.model.layers.48.self_attn.v_proj.weight": "model-00010-of-00015.safetensors",
536
+ "llm.model.layers.49.input_layernorm.weight": "model-00011-of-00015.safetensors",
537
+ "llm.model.layers.49.mlp.down_proj.weight": "model-00011-of-00015.safetensors",
538
+ "llm.model.layers.49.mlp.gate_proj.weight": "model-00011-of-00015.safetensors",
539
+ "llm.model.layers.49.mlp.up_proj.weight": "model-00011-of-00015.safetensors",
540
+ "llm.model.layers.49.post_attention_layernorm.weight": "model-00011-of-00015.safetensors",
541
+ "llm.model.layers.49.self_attn.k_proj.bias": "model-00011-of-00015.safetensors",
542
+ "llm.model.layers.49.self_attn.k_proj.weight": "model-00011-of-00015.safetensors",
543
+ "llm.model.layers.49.self_attn.o_proj.weight": "model-00011-of-00015.safetensors",
544
+ "llm.model.layers.49.self_attn.q_proj.bias": "model-00011-of-00015.safetensors",
545
+ "llm.model.layers.49.self_attn.q_proj.weight": "model-00011-of-00015.safetensors",
546
+ "llm.model.layers.49.self_attn.v_proj.bias": "model-00011-of-00015.safetensors",
547
+ "llm.model.layers.49.self_attn.v_proj.weight": "model-00011-of-00015.safetensors",
548
+ "llm.model.layers.5.input_layernorm.weight": "model-00002-of-00015.safetensors",
549
+ "llm.model.layers.5.mlp.down_proj.weight": "model-00002-of-00015.safetensors",
550
+ "llm.model.layers.5.mlp.gate_proj.weight": "model-00002-of-00015.safetensors",
551
+ "llm.model.layers.5.mlp.up_proj.weight": "model-00002-of-00015.safetensors",
552
+ "llm.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00015.safetensors",
553
+ "llm.model.layers.5.self_attn.k_proj.bias": "model-00002-of-00015.safetensors",
554
+ "llm.model.layers.5.self_attn.k_proj.weight": "model-00002-of-00015.safetensors",
555
+ "llm.model.layers.5.self_attn.o_proj.weight": "model-00002-of-00015.safetensors",
556
+ "llm.model.layers.5.self_attn.q_proj.bias": "model-00002-of-00015.safetensors",
557
+ "llm.model.layers.5.self_attn.q_proj.weight": "model-00002-of-00015.safetensors",
558
+ "llm.model.layers.5.self_attn.v_proj.bias": "model-00002-of-00015.safetensors",
559
+ "llm.model.layers.5.self_attn.v_proj.weight": "model-00002-of-00015.safetensors",
560
+ "llm.model.layers.50.input_layernorm.weight": "model-00011-of-00015.safetensors",
561
+ "llm.model.layers.50.mlp.down_proj.weight": "model-00011-of-00015.safetensors",
562
+ "llm.model.layers.50.mlp.gate_proj.weight": "model-00011-of-00015.safetensors",
563
+ "llm.model.layers.50.mlp.up_proj.weight": "model-00011-of-00015.safetensors",
564
+ "llm.model.layers.50.post_attention_layernorm.weight": "model-00011-of-00015.safetensors",
565
+ "llm.model.layers.50.self_attn.k_proj.bias": "model-00011-of-00015.safetensors",
566
+ "llm.model.layers.50.self_attn.k_proj.weight": "model-00011-of-00015.safetensors",
567
+ "llm.model.layers.50.self_attn.o_proj.weight": "model-00011-of-00015.safetensors",
568
+ "llm.model.layers.50.self_attn.q_proj.bias": "model-00011-of-00015.safetensors",
569
+ "llm.model.layers.50.self_attn.q_proj.weight": "model-00011-of-00015.safetensors",
570
+ "llm.model.layers.50.self_attn.v_proj.bias": "model-00011-of-00015.safetensors",
571
+ "llm.model.layers.50.self_attn.v_proj.weight": "model-00011-of-00015.safetensors",
572
+ "llm.model.layers.51.input_layernorm.weight": "model-00011-of-00015.safetensors",
573
+ "llm.model.layers.51.mlp.down_proj.weight": "model-00011-of-00015.safetensors",
574
+ "llm.model.layers.51.mlp.gate_proj.weight": "model-00011-of-00015.safetensors",
575
+ "llm.model.layers.51.mlp.up_proj.weight": "model-00011-of-00015.safetensors",
576
+ "llm.model.layers.51.post_attention_layernorm.weight": "model-00011-of-00015.safetensors",
577
+ "llm.model.layers.51.self_attn.k_proj.bias": "model-00011-of-00015.safetensors",
578
+ "llm.model.layers.51.self_attn.k_proj.weight": "model-00011-of-00015.safetensors",
579
+ "llm.model.layers.51.self_attn.o_proj.weight": "model-00011-of-00015.safetensors",
580
+ "llm.model.layers.51.self_attn.q_proj.bias": "model-00011-of-00015.safetensors",
581
+ "llm.model.layers.51.self_attn.q_proj.weight": "model-00011-of-00015.safetensors",
582
+ "llm.model.layers.51.self_attn.v_proj.bias": "model-00011-of-00015.safetensors",
583
+ "llm.model.layers.51.self_attn.v_proj.weight": "model-00011-of-00015.safetensors",
584
+ "llm.model.layers.52.input_layernorm.weight": "model-00011-of-00015.safetensors",
585
+ "llm.model.layers.52.mlp.down_proj.weight": "model-00011-of-00015.safetensors",
586
+ "llm.model.layers.52.mlp.gate_proj.weight": "model-00011-of-00015.safetensors",
587
+ "llm.model.layers.52.mlp.up_proj.weight": "model-00011-of-00015.safetensors",
588
+ "llm.model.layers.52.post_attention_layernorm.weight": "model-00011-of-00015.safetensors",
589
+ "llm.model.layers.52.self_attn.k_proj.bias": "model-00011-of-00015.safetensors",
590
+ "llm.model.layers.52.self_attn.k_proj.weight": "model-00011-of-00015.safetensors",
591
+ "llm.model.layers.52.self_attn.o_proj.weight": "model-00011-of-00015.safetensors",
592
+ "llm.model.layers.52.self_attn.q_proj.bias": "model-00011-of-00015.safetensors",
593
+ "llm.model.layers.52.self_attn.q_proj.weight": "model-00011-of-00015.safetensors",
594
+ "llm.model.layers.52.self_attn.v_proj.bias": "model-00011-of-00015.safetensors",
595
+ "llm.model.layers.52.self_attn.v_proj.weight": "model-00011-of-00015.safetensors",
596
+ "llm.model.layers.53.input_layernorm.weight": "model-00012-of-00015.safetensors",
597
+ "llm.model.layers.53.mlp.down_proj.weight": "model-00012-of-00015.safetensors",
598
+ "llm.model.layers.53.mlp.gate_proj.weight": "model-00011-of-00015.safetensors",
599
+ "llm.model.layers.53.mlp.up_proj.weight": "model-00012-of-00015.safetensors",
600
+ "llm.model.layers.53.post_attention_layernorm.weight": "model-00012-of-00015.safetensors",
601
+ "llm.model.layers.53.self_attn.k_proj.bias": "model-00011-of-00015.safetensors",
602
+ "llm.model.layers.53.self_attn.k_proj.weight": "model-00011-of-00015.safetensors",
603
+ "llm.model.layers.53.self_attn.o_proj.weight": "model-00011-of-00015.safetensors",
604
+ "llm.model.layers.53.self_attn.q_proj.bias": "model-00011-of-00015.safetensors",
605
+ "llm.model.layers.53.self_attn.q_proj.weight": "model-00011-of-00015.safetensors",
606
+ "llm.model.layers.53.self_attn.v_proj.bias": "model-00011-of-00015.safetensors",
607
+ "llm.model.layers.53.self_attn.v_proj.weight": "model-00011-of-00015.safetensors",
608
+ "llm.model.layers.54.input_layernorm.weight": "model-00012-of-00015.safetensors",
609
+ "llm.model.layers.54.mlp.down_proj.weight": "model-00012-of-00015.safetensors",
610
+ "llm.model.layers.54.mlp.gate_proj.weight": "model-00012-of-00015.safetensors",
611
+ "llm.model.layers.54.mlp.up_proj.weight": "model-00012-of-00015.safetensors",
612
+ "llm.model.layers.54.post_attention_layernorm.weight": "model-00012-of-00015.safetensors",
613
+ "llm.model.layers.54.self_attn.k_proj.bias": "model-00012-of-00015.safetensors",
614
+ "llm.model.layers.54.self_attn.k_proj.weight": "model-00012-of-00015.safetensors",
615
+ "llm.model.layers.54.self_attn.o_proj.weight": "model-00012-of-00015.safetensors",
616
+ "llm.model.layers.54.self_attn.q_proj.bias": "model-00012-of-00015.safetensors",
617
+ "llm.model.layers.54.self_attn.q_proj.weight": "model-00012-of-00015.safetensors",
618
+ "llm.model.layers.54.self_attn.v_proj.bias": "model-00012-of-00015.safetensors",
619
+ "llm.model.layers.54.self_attn.v_proj.weight": "model-00012-of-00015.safetensors",
620
+ "llm.model.layers.55.input_layernorm.weight": "model-00012-of-00015.safetensors",
621
+ "llm.model.layers.55.mlp.down_proj.weight": "model-00012-of-00015.safetensors",
622
+ "llm.model.layers.55.mlp.gate_proj.weight": "model-00012-of-00015.safetensors",
623
+ "llm.model.layers.55.mlp.up_proj.weight": "model-00012-of-00015.safetensors",
624
+ "llm.model.layers.55.post_attention_layernorm.weight": "model-00012-of-00015.safetensors",
625
+ "llm.model.layers.55.self_attn.k_proj.bias": "model-00012-of-00015.safetensors",
626
+ "llm.model.layers.55.self_attn.k_proj.weight": "model-00012-of-00015.safetensors",
627
+ "llm.model.layers.55.self_attn.o_proj.weight": "model-00012-of-00015.safetensors",
628
+ "llm.model.layers.55.self_attn.q_proj.bias": "model-00012-of-00015.safetensors",
629
+ "llm.model.layers.55.self_attn.q_proj.weight": "model-00012-of-00015.safetensors",
630
+ "llm.model.layers.55.self_attn.v_proj.bias": "model-00012-of-00015.safetensors",
631
+ "llm.model.layers.55.self_attn.v_proj.weight": "model-00012-of-00015.safetensors",
632
+ "llm.model.layers.56.input_layernorm.weight": "model-00012-of-00015.safetensors",
633
+ "llm.model.layers.56.mlp.down_proj.weight": "model-00012-of-00015.safetensors",
634
+ "llm.model.layers.56.mlp.gate_proj.weight": "model-00012-of-00015.safetensors",
635
+ "llm.model.layers.56.mlp.up_proj.weight": "model-00012-of-00015.safetensors",
636
+ "llm.model.layers.56.post_attention_layernorm.weight": "model-00012-of-00015.safetensors",
637
+ "llm.model.layers.56.self_attn.k_proj.bias": "model-00012-of-00015.safetensors",
638
+ "llm.model.layers.56.self_attn.k_proj.weight": "model-00012-of-00015.safetensors",
639
+ "llm.model.layers.56.self_attn.o_proj.weight": "model-00012-of-00015.safetensors",
640
+ "llm.model.layers.56.self_attn.q_proj.bias": "model-00012-of-00015.safetensors",
641
+ "llm.model.layers.56.self_attn.q_proj.weight": "model-00012-of-00015.safetensors",
642
+ "llm.model.layers.56.self_attn.v_proj.bias": "model-00012-of-00015.safetensors",
643
+ "llm.model.layers.56.self_attn.v_proj.weight": "model-00012-of-00015.safetensors",
644
+ "llm.model.layers.57.input_layernorm.weight": "model-00012-of-00015.safetensors",
645
+ "llm.model.layers.57.mlp.down_proj.weight": "model-00012-of-00015.safetensors",
646
+ "llm.model.layers.57.mlp.gate_proj.weight": "model-00012-of-00015.safetensors",
647
+ "llm.model.layers.57.mlp.up_proj.weight": "model-00012-of-00015.safetensors",
648
+ "llm.model.layers.57.post_attention_layernorm.weight": "model-00012-of-00015.safetensors",
649
+ "llm.model.layers.57.self_attn.k_proj.bias": "model-00012-of-00015.safetensors",
650
+ "llm.model.layers.57.self_attn.k_proj.weight": "model-00012-of-00015.safetensors",
651
+ "llm.model.layers.57.self_attn.o_proj.weight": "model-00012-of-00015.safetensors",
652
+ "llm.model.layers.57.self_attn.q_proj.bias": "model-00012-of-00015.safetensors",
653
+ "llm.model.layers.57.self_attn.q_proj.weight": "model-00012-of-00015.safetensors",
654
+ "llm.model.layers.57.self_attn.v_proj.bias": "model-00012-of-00015.safetensors",
655
+ "llm.model.layers.57.self_attn.v_proj.weight": "model-00012-of-00015.safetensors",
656
+ "llm.model.layers.58.input_layernorm.weight": "model-00013-of-00015.safetensors",
657
+ "llm.model.layers.58.mlp.down_proj.weight": "model-00013-of-00015.safetensors",
658
+ "llm.model.layers.58.mlp.gate_proj.weight": "model-00012-of-00015.safetensors",
659
+ "llm.model.layers.58.mlp.up_proj.weight": "model-00013-of-00015.safetensors",
660
+ "llm.model.layers.58.post_attention_layernorm.weight": "model-00013-of-00015.safetensors",
661
+ "llm.model.layers.58.self_attn.k_proj.bias": "model-00012-of-00015.safetensors",
662
+ "llm.model.layers.58.self_attn.k_proj.weight": "model-00012-of-00015.safetensors",
663
+ "llm.model.layers.58.self_attn.o_proj.weight": "model-00012-of-00015.safetensors",
664
+ "llm.model.layers.58.self_attn.q_proj.bias": "model-00012-of-00015.safetensors",
665
+ "llm.model.layers.58.self_attn.q_proj.weight": "model-00012-of-00015.safetensors",
666
+ "llm.model.layers.58.self_attn.v_proj.bias": "model-00012-of-00015.safetensors",
667
+ "llm.model.layers.58.self_attn.v_proj.weight": "model-00012-of-00015.safetensors",
668
+ "llm.model.layers.59.input_layernorm.weight": "model-00013-of-00015.safetensors",
669
+ "llm.model.layers.59.mlp.down_proj.weight": "model-00013-of-00015.safetensors",
670
+ "llm.model.layers.59.mlp.gate_proj.weight": "model-00013-of-00015.safetensors",
671
+ "llm.model.layers.59.mlp.up_proj.weight": "model-00013-of-00015.safetensors",
672
+ "llm.model.layers.59.post_attention_layernorm.weight": "model-00013-of-00015.safetensors",
673
+ "llm.model.layers.59.self_attn.k_proj.bias": "model-00013-of-00015.safetensors",
674
+ "llm.model.layers.59.self_attn.k_proj.weight": "model-00013-of-00015.safetensors",
675
+ "llm.model.layers.59.self_attn.o_proj.weight": "model-00013-of-00015.safetensors",
676
+ "llm.model.layers.59.self_attn.q_proj.bias": "model-00013-of-00015.safetensors",
677
+ "llm.model.layers.59.self_attn.q_proj.weight": "model-00013-of-00015.safetensors",
678
+ "llm.model.layers.59.self_attn.v_proj.bias": "model-00013-of-00015.safetensors",
679
+ "llm.model.layers.59.self_attn.v_proj.weight": "model-00013-of-00015.safetensors",
680
+ "llm.model.layers.6.input_layernorm.weight": "model-00002-of-00015.safetensors",
681
+ "llm.model.layers.6.mlp.down_proj.weight": "model-00002-of-00015.safetensors",
682
+ "llm.model.layers.6.mlp.gate_proj.weight": "model-00002-of-00015.safetensors",
683
+ "llm.model.layers.6.mlp.up_proj.weight": "model-00002-of-00015.safetensors",
684
+ "llm.model.layers.6.post_attention_layernorm.weight": "model-00002-of-00015.safetensors",
685
+ "llm.model.layers.6.self_attn.k_proj.bias": "model-00002-of-00015.safetensors",
686
+ "llm.model.layers.6.self_attn.k_proj.weight": "model-00002-of-00015.safetensors",
687
+ "llm.model.layers.6.self_attn.o_proj.weight": "model-00002-of-00015.safetensors",
688
+ "llm.model.layers.6.self_attn.q_proj.bias": "model-00002-of-00015.safetensors",
689
+ "llm.model.layers.6.self_attn.q_proj.weight": "model-00002-of-00015.safetensors",
690
+ "llm.model.layers.6.self_attn.v_proj.bias": "model-00002-of-00015.safetensors",
691
+ "llm.model.layers.6.self_attn.v_proj.weight": "model-00002-of-00015.safetensors",
692
+ "llm.model.layers.60.input_layernorm.weight": "model-00013-of-00015.safetensors",
693
+ "llm.model.layers.60.mlp.down_proj.weight": "model-00013-of-00015.safetensors",
694
+ "llm.model.layers.60.mlp.gate_proj.weight": "model-00013-of-00015.safetensors",
695
+ "llm.model.layers.60.mlp.up_proj.weight": "model-00013-of-00015.safetensors",
696
+ "llm.model.layers.60.post_attention_layernorm.weight": "model-00013-of-00015.safetensors",
697
+ "llm.model.layers.60.self_attn.k_proj.bias": "model-00013-of-00015.safetensors",
698
+ "llm.model.layers.60.self_attn.k_proj.weight": "model-00013-of-00015.safetensors",
699
+ "llm.model.layers.60.self_attn.o_proj.weight": "model-00013-of-00015.safetensors",
700
+ "llm.model.layers.60.self_attn.q_proj.bias": "model-00013-of-00015.safetensors",
701
+ "llm.model.layers.60.self_attn.q_proj.weight": "model-00013-of-00015.safetensors",
702
+ "llm.model.layers.60.self_attn.v_proj.bias": "model-00013-of-00015.safetensors",
703
+ "llm.model.layers.60.self_attn.v_proj.weight": "model-00013-of-00015.safetensors",
704
+ "llm.model.layers.61.input_layernorm.weight": "model-00013-of-00015.safetensors",
705
+ "llm.model.layers.61.mlp.down_proj.weight": "model-00013-of-00015.safetensors",
706
+ "llm.model.layers.61.mlp.gate_proj.weight": "model-00013-of-00015.safetensors",
707
+ "llm.model.layers.61.mlp.up_proj.weight": "model-00013-of-00015.safetensors",
708
+ "llm.model.layers.61.post_attention_layernorm.weight": "model-00013-of-00015.safetensors",
709
+ "llm.model.layers.61.self_attn.k_proj.bias": "model-00013-of-00015.safetensors",
710
+ "llm.model.layers.61.self_attn.k_proj.weight": "model-00013-of-00015.safetensors",
711
+ "llm.model.layers.61.self_attn.o_proj.weight": "model-00013-of-00015.safetensors",
712
+ "llm.model.layers.61.self_attn.q_proj.bias": "model-00013-of-00015.safetensors",
713
+ "llm.model.layers.61.self_attn.q_proj.weight": "model-00013-of-00015.safetensors",
714
+ "llm.model.layers.61.self_attn.v_proj.bias": "model-00013-of-00015.safetensors",
715
+ "llm.model.layers.61.self_attn.v_proj.weight": "model-00013-of-00015.safetensors",
716
+ "llm.model.layers.62.input_layernorm.weight": "model-00013-of-00015.safetensors",
717
+ "llm.model.layers.62.mlp.down_proj.weight": "model-00013-of-00015.safetensors",
718
+ "llm.model.layers.62.mlp.gate_proj.weight": "model-00013-of-00015.safetensors",
719
+ "llm.model.layers.62.mlp.up_proj.weight": "model-00013-of-00015.safetensors",
720
+ "llm.model.layers.62.post_attention_layernorm.weight": "model-00013-of-00015.safetensors",
721
+ "llm.model.layers.62.self_attn.k_proj.bias": "model-00013-of-00015.safetensors",
722
+ "llm.model.layers.62.self_attn.k_proj.weight": "model-00013-of-00015.safetensors",
723
+ "llm.model.layers.62.self_attn.o_proj.weight": "model-00013-of-00015.safetensors",
724
+ "llm.model.layers.62.self_attn.q_proj.bias": "model-00013-of-00015.safetensors",
725
+ "llm.model.layers.62.self_attn.q_proj.weight": "model-00013-of-00015.safetensors",
726
+ "llm.model.layers.62.self_attn.v_proj.bias": "model-00013-of-00015.safetensors",
727
+ "llm.model.layers.62.self_attn.v_proj.weight": "model-00013-of-00015.safetensors",
728
+ "llm.model.layers.63.input_layernorm.weight": "model-00014-of-00015.safetensors",
729
+ "llm.model.layers.63.mlp.down_proj.weight": "model-00014-of-00015.safetensors",
730
+ "llm.model.layers.63.mlp.gate_proj.weight": "model-00013-of-00015.safetensors",
731
+ "llm.model.layers.63.mlp.up_proj.weight": "model-00014-of-00015.safetensors",
732
+ "llm.model.layers.63.post_attention_layernorm.weight": "model-00014-of-00015.safetensors",
733
+ "llm.model.layers.63.self_attn.k_proj.bias": "model-00013-of-00015.safetensors",
734
+ "llm.model.layers.63.self_attn.k_proj.weight": "model-00013-of-00015.safetensors",
735
+ "llm.model.layers.63.self_attn.o_proj.weight": "model-00013-of-00015.safetensors",
736
+ "llm.model.layers.63.self_attn.q_proj.bias": "model-00013-of-00015.safetensors",
737
+ "llm.model.layers.63.self_attn.q_proj.weight": "model-00013-of-00015.safetensors",
738
+ "llm.model.layers.63.self_attn.v_proj.bias": "model-00013-of-00015.safetensors",
739
+ "llm.model.layers.63.self_attn.v_proj.weight": "model-00013-of-00015.safetensors",
740
+ "llm.model.layers.7.input_layernorm.weight": "model-00002-of-00015.safetensors",
741
+ "llm.model.layers.7.mlp.down_proj.weight": "model-00002-of-00015.safetensors",
742
+ "llm.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00015.safetensors",
743
+ "llm.model.layers.7.mlp.up_proj.weight": "model-00002-of-00015.safetensors",
744
+ "llm.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00015.safetensors",
745
+ "llm.model.layers.7.self_attn.k_proj.bias": "model-00002-of-00015.safetensors",
746
+ "llm.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00015.safetensors",
747
+ "llm.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00015.safetensors",
748
+ "llm.model.layers.7.self_attn.q_proj.bias": "model-00002-of-00015.safetensors",
749
+ "llm.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00015.safetensors",
750
+ "llm.model.layers.7.self_attn.v_proj.bias": "model-00002-of-00015.safetensors",
751
+ "llm.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00015.safetensors",
752
+ "llm.model.layers.8.input_layernorm.weight": "model-00003-of-00015.safetensors",
753
+ "llm.model.layers.8.mlp.down_proj.weight": "model-00003-of-00015.safetensors",
754
+ "llm.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00015.safetensors",
755
+ "llm.model.layers.8.mlp.up_proj.weight": "model-00003-of-00015.safetensors",
756
+ "llm.model.layers.8.post_attention_layernorm.weight": "model-00003-of-00015.safetensors",
757
+ "llm.model.layers.8.self_attn.k_proj.bias": "model-00002-of-00015.safetensors",
758
+ "llm.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00015.safetensors",
759
+ "llm.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00015.safetensors",
760
+ "llm.model.layers.8.self_attn.q_proj.bias": "model-00002-of-00015.safetensors",
761
+ "llm.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00015.safetensors",
762
+ "llm.model.layers.8.self_attn.v_proj.bias": "model-00002-of-00015.safetensors",
763
+ "llm.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00015.safetensors",
764
+ "llm.model.layers.9.input_layernorm.weight": "model-00003-of-00015.safetensors",
765
+ "llm.model.layers.9.mlp.down_proj.weight": "model-00003-of-00015.safetensors",
766
+ "llm.model.layers.9.mlp.gate_proj.weight": "model-00003-of-00015.safetensors",
767
+ "llm.model.layers.9.mlp.up_proj.weight": "model-00003-of-00015.safetensors",
768
+ "llm.model.layers.9.post_attention_layernorm.weight": "model-00003-of-00015.safetensors",
769
+ "llm.model.layers.9.self_attn.k_proj.bias": "model-00003-of-00015.safetensors",
770
+ "llm.model.layers.9.self_attn.k_proj.weight": "model-00003-of-00015.safetensors",
771
+ "llm.model.layers.9.self_attn.o_proj.weight": "model-00003-of-00015.safetensors",
772
+ "llm.model.layers.9.self_attn.q_proj.bias": "model-00003-of-00015.safetensors",
773
+ "llm.model.layers.9.self_attn.q_proj.weight": "model-00003-of-00015.safetensors",
774
+ "llm.model.layers.9.self_attn.v_proj.bias": "model-00003-of-00015.safetensors",
775
+ "llm.model.layers.9.self_attn.v_proj.weight": "model-00003-of-00015.safetensors",
776
+ "llm.model.norm.weight": "model-00014-of-00015.safetensors",
777
+ "visual_tokenizer.backbone.preprocessor.patchifier.norm.weight": "model-00014-of-00015.safetensors",
778
+ "visual_tokenizer.backbone.preprocessor.patchifier.proj.bias": "model-00014-of-00015.safetensors",
779
+ "visual_tokenizer.backbone.preprocessor.patchifier.proj.weight": "model-00014-of-00015.safetensors",
780
+ "visual_tokenizer.backbone.preprocessor.pos_embed": "model-00014-of-00015.safetensors",
781
+ "visual_tokenizer.backbone.trunk.blocks.0.attn.proj.weight": "model-00014-of-00015.safetensors",
782
+ "visual_tokenizer.backbone.trunk.blocks.0.attn.qkv.weight": "model-00014-of-00015.safetensors",
783
+ "visual_tokenizer.backbone.trunk.blocks.0.mlp.fc1.weight": "model-00014-of-00015.safetensors",
784
+ "visual_tokenizer.backbone.trunk.blocks.0.mlp.fc2.weight": "model-00014-of-00015.safetensors",
785
+ "visual_tokenizer.backbone.trunk.blocks.0.mlp.fc3.weight": "model-00014-of-00015.safetensors",
786
+ "visual_tokenizer.backbone.trunk.blocks.0.norm_1.weight": "model-00014-of-00015.safetensors",
787
+ "visual_tokenizer.backbone.trunk.blocks.0.norm_2.weight": "model-00014-of-00015.safetensors",
788
+ "visual_tokenizer.backbone.trunk.blocks.1.attn.proj.weight": "model-00014-of-00015.safetensors",
789
+ "visual_tokenizer.backbone.trunk.blocks.1.attn.qkv.weight": "model-00014-of-00015.safetensors",
790
+ "visual_tokenizer.backbone.trunk.blocks.1.mlp.fc1.weight": "model-00014-of-00015.safetensors",
791
+ "visual_tokenizer.backbone.trunk.blocks.1.mlp.fc2.weight": "model-00014-of-00015.safetensors",
792
+ "visual_tokenizer.backbone.trunk.blocks.1.mlp.fc3.weight": "model-00014-of-00015.safetensors",
793
+ "visual_tokenizer.backbone.trunk.blocks.1.norm_1.weight": "model-00014-of-00015.safetensors",
794
+ "visual_tokenizer.backbone.trunk.blocks.1.norm_2.weight": "model-00014-of-00015.safetensors",
795
+ "visual_tokenizer.backbone.trunk.blocks.10.attn.proj.weight": "model-00014-of-00015.safetensors",
796
+ "visual_tokenizer.backbone.trunk.blocks.10.attn.qkv.weight": "model-00014-of-00015.safetensors",
797
+ "visual_tokenizer.backbone.trunk.blocks.10.mlp.fc1.weight": "model-00014-of-00015.safetensors",
798
+ "visual_tokenizer.backbone.trunk.blocks.10.mlp.fc2.weight": "model-00014-of-00015.safetensors",
799
+ "visual_tokenizer.backbone.trunk.blocks.10.mlp.fc3.weight": "model-00014-of-00015.safetensors",
800
+ "visual_tokenizer.backbone.trunk.blocks.10.norm_1.weight": "model-00014-of-00015.safetensors",
801
+ "visual_tokenizer.backbone.trunk.blocks.10.norm_2.weight": "model-00014-of-00015.safetensors",
802
+ "visual_tokenizer.backbone.trunk.blocks.11.attn.proj.weight": "model-00014-of-00015.safetensors",
803
+ "visual_tokenizer.backbone.trunk.blocks.11.attn.qkv.weight": "model-00014-of-00015.safetensors",
804
+ "visual_tokenizer.backbone.trunk.blocks.11.mlp.fc1.weight": "model-00014-of-00015.safetensors",
805
+ "visual_tokenizer.backbone.trunk.blocks.11.mlp.fc2.weight": "model-00014-of-00015.safetensors",
806
+ "visual_tokenizer.backbone.trunk.blocks.11.mlp.fc3.weight": "model-00014-of-00015.safetensors",
807
+ "visual_tokenizer.backbone.trunk.blocks.11.norm_1.weight": "model-00014-of-00015.safetensors",
808
+ "visual_tokenizer.backbone.trunk.blocks.11.norm_2.weight": "model-00014-of-00015.safetensors",
809
+ "visual_tokenizer.backbone.trunk.blocks.12.attn.proj.weight": "model-00014-of-00015.safetensors",
810
+ "visual_tokenizer.backbone.trunk.blocks.12.attn.qkv.weight": "model-00014-of-00015.safetensors",
811
+ "visual_tokenizer.backbone.trunk.blocks.12.mlp.fc1.weight": "model-00014-of-00015.safetensors",
812
+ "visual_tokenizer.backbone.trunk.blocks.12.mlp.fc2.weight": "model-00014-of-00015.safetensors",
813
+ "visual_tokenizer.backbone.trunk.blocks.12.mlp.fc3.weight": "model-00014-of-00015.safetensors",
814
+ "visual_tokenizer.backbone.trunk.blocks.12.norm_1.weight": "model-00014-of-00015.safetensors",
815
+ "visual_tokenizer.backbone.trunk.blocks.12.norm_2.weight": "model-00014-of-00015.safetensors",
816
+ "visual_tokenizer.backbone.trunk.blocks.13.attn.proj.weight": "model-00014-of-00015.safetensors",
817
+ "visual_tokenizer.backbone.trunk.blocks.13.attn.qkv.weight": "model-00014-of-00015.safetensors",
818
+ "visual_tokenizer.backbone.trunk.blocks.13.mlp.fc1.weight": "model-00014-of-00015.safetensors",
819
+ "visual_tokenizer.backbone.trunk.blocks.13.mlp.fc2.weight": "model-00014-of-00015.safetensors",
820
+ "visual_tokenizer.backbone.trunk.blocks.13.mlp.fc3.weight": "model-00014-of-00015.safetensors",
821
+ "visual_tokenizer.backbone.trunk.blocks.13.norm_1.weight": "model-00014-of-00015.safetensors",
822
+ "visual_tokenizer.backbone.trunk.blocks.13.norm_2.weight": "model-00014-of-00015.safetensors",
823
+ "visual_tokenizer.backbone.trunk.blocks.14.attn.proj.weight": "model-00014-of-00015.safetensors",
824
+ "visual_tokenizer.backbone.trunk.blocks.14.attn.qkv.weight": "model-00014-of-00015.safetensors",
825
+ "visual_tokenizer.backbone.trunk.blocks.14.mlp.fc1.weight": "model-00014-of-00015.safetensors",
826
+ "visual_tokenizer.backbone.trunk.blocks.14.mlp.fc2.weight": "model-00014-of-00015.safetensors",
827
+ "visual_tokenizer.backbone.trunk.blocks.14.mlp.fc3.weight": "model-00014-of-00015.safetensors",
828
+ "visual_tokenizer.backbone.trunk.blocks.14.norm_1.weight": "model-00014-of-00015.safetensors",
829
+ "visual_tokenizer.backbone.trunk.blocks.14.norm_2.weight": "model-00014-of-00015.safetensors",
830
+ "visual_tokenizer.backbone.trunk.blocks.15.attn.proj.weight": "model-00014-of-00015.safetensors",
831
+ "visual_tokenizer.backbone.trunk.blocks.15.attn.qkv.weight": "model-00014-of-00015.safetensors",
832
+ "visual_tokenizer.backbone.trunk.blocks.15.mlp.fc1.weight": "model-00014-of-00015.safetensors",
833
+ "visual_tokenizer.backbone.trunk.blocks.15.mlp.fc2.weight": "model-00014-of-00015.safetensors",
834
+ "visual_tokenizer.backbone.trunk.blocks.15.mlp.fc3.weight": "model-00014-of-00015.safetensors",
835
+ "visual_tokenizer.backbone.trunk.blocks.15.norm_1.weight": "model-00014-of-00015.safetensors",
836
+ "visual_tokenizer.backbone.trunk.blocks.15.norm_2.weight": "model-00014-of-00015.safetensors",
837
+ "visual_tokenizer.backbone.trunk.blocks.16.attn.proj.weight": "model-00014-of-00015.safetensors",
838
+ "visual_tokenizer.backbone.trunk.blocks.16.attn.qkv.weight": "model-00014-of-00015.safetensors",
839
+ "visual_tokenizer.backbone.trunk.blocks.16.mlp.fc1.weight": "model-00014-of-00015.safetensors",
840
+ "visual_tokenizer.backbone.trunk.blocks.16.mlp.fc2.weight": "model-00014-of-00015.safetensors",
841
+ "visual_tokenizer.backbone.trunk.blocks.16.mlp.fc3.weight": "model-00014-of-00015.safetensors",
842
+ "visual_tokenizer.backbone.trunk.blocks.16.norm_1.weight": "model-00014-of-00015.safetensors",
843
+ "visual_tokenizer.backbone.trunk.blocks.16.norm_2.weight": "model-00014-of-00015.safetensors",
844
+ "visual_tokenizer.backbone.trunk.blocks.17.attn.proj.weight": "model-00014-of-00015.safetensors",
845
+ "visual_tokenizer.backbone.trunk.blocks.17.attn.qkv.weight": "model-00014-of-00015.safetensors",
846
+ "visual_tokenizer.backbone.trunk.blocks.17.mlp.fc1.weight": "model-00014-of-00015.safetensors",
847
+ "visual_tokenizer.backbone.trunk.blocks.17.mlp.fc2.weight": "model-00014-of-00015.safetensors",
848
+ "visual_tokenizer.backbone.trunk.blocks.17.mlp.fc3.weight": "model-00014-of-00015.safetensors",
849
+ "visual_tokenizer.backbone.trunk.blocks.17.norm_1.weight": "model-00014-of-00015.safetensors",
850
+ "visual_tokenizer.backbone.trunk.blocks.17.norm_2.weight": "model-00014-of-00015.safetensors",
851
+ "visual_tokenizer.backbone.trunk.blocks.18.attn.proj.weight": "model-00014-of-00015.safetensors",
852
+ "visual_tokenizer.backbone.trunk.blocks.18.attn.qkv.weight": "model-00014-of-00015.safetensors",
853
+ "visual_tokenizer.backbone.trunk.blocks.18.mlp.fc1.weight": "model-00014-of-00015.safetensors",
854
+ "visual_tokenizer.backbone.trunk.blocks.18.mlp.fc2.weight": "model-00014-of-00015.safetensors",
855
+ "visual_tokenizer.backbone.trunk.blocks.18.mlp.fc3.weight": "model-00014-of-00015.safetensors",
856
+ "visual_tokenizer.backbone.trunk.blocks.18.norm_1.weight": "model-00014-of-00015.safetensors",
857
+ "visual_tokenizer.backbone.trunk.blocks.18.norm_2.weight": "model-00014-of-00015.safetensors",
858
+ "visual_tokenizer.backbone.trunk.blocks.19.attn.proj.weight": "model-00014-of-00015.safetensors",
859
+ "visual_tokenizer.backbone.trunk.blocks.19.attn.qkv.weight": "model-00014-of-00015.safetensors",
860
+ "visual_tokenizer.backbone.trunk.blocks.19.mlp.fc1.weight": "model-00014-of-00015.safetensors",
861
+ "visual_tokenizer.backbone.trunk.blocks.19.mlp.fc2.weight": "model-00014-of-00015.safetensors",
862
+ "visual_tokenizer.backbone.trunk.blocks.19.mlp.fc3.weight": "model-00014-of-00015.safetensors",
863
+ "visual_tokenizer.backbone.trunk.blocks.19.norm_1.weight": "model-00014-of-00015.safetensors",
864
+ "visual_tokenizer.backbone.trunk.blocks.19.norm_2.weight": "model-00014-of-00015.safetensors",
865
+ "visual_tokenizer.backbone.trunk.blocks.2.attn.proj.weight": "model-00014-of-00015.safetensors",
866
+ "visual_tokenizer.backbone.trunk.blocks.2.attn.qkv.weight": "model-00014-of-00015.safetensors",
867
+ "visual_tokenizer.backbone.trunk.blocks.2.mlp.fc1.weight": "model-00014-of-00015.safetensors",
868
+ "visual_tokenizer.backbone.trunk.blocks.2.mlp.fc2.weight": "model-00014-of-00015.safetensors",
869
+ "visual_tokenizer.backbone.trunk.blocks.2.mlp.fc3.weight": "model-00014-of-00015.safetensors",
870
+ "visual_tokenizer.backbone.trunk.blocks.2.norm_1.weight": "model-00014-of-00015.safetensors",
871
+ "visual_tokenizer.backbone.trunk.blocks.2.norm_2.weight": "model-00014-of-00015.safetensors",
872
+ "visual_tokenizer.backbone.trunk.blocks.20.attn.proj.weight": "model-00014-of-00015.safetensors",
873
+ "visual_tokenizer.backbone.trunk.blocks.20.attn.qkv.weight": "model-00014-of-00015.safetensors",
874
+ "visual_tokenizer.backbone.trunk.blocks.20.mlp.fc1.weight": "model-00014-of-00015.safetensors",
875
+ "visual_tokenizer.backbone.trunk.blocks.20.mlp.fc2.weight": "model-00014-of-00015.safetensors",
876
+ "visual_tokenizer.backbone.trunk.blocks.20.mlp.fc3.weight": "model-00014-of-00015.safetensors",
877
+ "visual_tokenizer.backbone.trunk.blocks.20.norm_1.weight": "model-00014-of-00015.safetensors",
878
+ "visual_tokenizer.backbone.trunk.blocks.20.norm_2.weight": "model-00014-of-00015.safetensors",
879
+ "visual_tokenizer.backbone.trunk.blocks.21.attn.proj.weight": "model-00014-of-00015.safetensors",
880
+ "visual_tokenizer.backbone.trunk.blocks.21.attn.qkv.weight": "model-00014-of-00015.safetensors",
881
+ "visual_tokenizer.backbone.trunk.blocks.21.mlp.fc1.weight": "model-00014-of-00015.safetensors",
882
+ "visual_tokenizer.backbone.trunk.blocks.21.mlp.fc2.weight": "model-00014-of-00015.safetensors",
883
+ "visual_tokenizer.backbone.trunk.blocks.21.mlp.fc3.weight": "model-00014-of-00015.safetensors",
884
+ "visual_tokenizer.backbone.trunk.blocks.21.norm_1.weight": "model-00014-of-00015.safetensors",
885
+ "visual_tokenizer.backbone.trunk.blocks.21.norm_2.weight": "model-00014-of-00015.safetensors",
886
+ "visual_tokenizer.backbone.trunk.blocks.22.attn.proj.weight": "model-00014-of-00015.safetensors",
887
+ "visual_tokenizer.backbone.trunk.blocks.22.attn.qkv.weight": "model-00014-of-00015.safetensors",
888
+ "visual_tokenizer.backbone.trunk.blocks.22.mlp.fc1.weight": "model-00014-of-00015.safetensors",
889
+ "visual_tokenizer.backbone.trunk.blocks.22.mlp.fc2.weight": "model-00014-of-00015.safetensors",
890
+ "visual_tokenizer.backbone.trunk.blocks.22.mlp.fc3.weight": "model-00014-of-00015.safetensors",
891
+ "visual_tokenizer.backbone.trunk.blocks.22.norm_1.weight": "model-00014-of-00015.safetensors",
892
+ "visual_tokenizer.backbone.trunk.blocks.22.norm_2.weight": "model-00014-of-00015.safetensors",
893
+ "visual_tokenizer.backbone.trunk.blocks.23.attn.proj.weight": "model-00014-of-00015.safetensors",
894
+ "visual_tokenizer.backbone.trunk.blocks.23.attn.qkv.weight": "model-00014-of-00015.safetensors",
895
+ "visual_tokenizer.backbone.trunk.blocks.23.mlp.fc1.weight": "model-00014-of-00015.safetensors",
896
+ "visual_tokenizer.backbone.trunk.blocks.23.mlp.fc2.weight": "model-00014-of-00015.safetensors",
897
+ "visual_tokenizer.backbone.trunk.blocks.23.mlp.fc3.weight": "model-00014-of-00015.safetensors",
898
+ "visual_tokenizer.backbone.trunk.blocks.23.norm_1.weight": "model-00014-of-00015.safetensors",
899
+ "visual_tokenizer.backbone.trunk.blocks.23.norm_2.weight": "model-00014-of-00015.safetensors",
900
+ "visual_tokenizer.backbone.trunk.blocks.3.attn.proj.weight": "model-00014-of-00015.safetensors",
901
+ "visual_tokenizer.backbone.trunk.blocks.3.attn.qkv.weight": "model-00014-of-00015.safetensors",
902
+ "visual_tokenizer.backbone.trunk.blocks.3.mlp.fc1.weight": "model-00014-of-00015.safetensors",
903
+ "visual_tokenizer.backbone.trunk.blocks.3.mlp.fc2.weight": "model-00014-of-00015.safetensors",
904
+ "visual_tokenizer.backbone.trunk.blocks.3.mlp.fc3.weight": "model-00014-of-00015.safetensors",
905
+ "visual_tokenizer.backbone.trunk.blocks.3.norm_1.weight": "model-00014-of-00015.safetensors",
906
+ "visual_tokenizer.backbone.trunk.blocks.3.norm_2.weight": "model-00014-of-00015.safetensors",
907
+ "visual_tokenizer.backbone.trunk.blocks.4.attn.proj.weight": "model-00014-of-00015.safetensors",
908
+ "visual_tokenizer.backbone.trunk.blocks.4.attn.qkv.weight": "model-00014-of-00015.safetensors",
909
+ "visual_tokenizer.backbone.trunk.blocks.4.mlp.fc1.weight": "model-00014-of-00015.safetensors",
910
+ "visual_tokenizer.backbone.trunk.blocks.4.mlp.fc2.weight": "model-00014-of-00015.safetensors",
911
+ "visual_tokenizer.backbone.trunk.blocks.4.mlp.fc3.weight": "model-00014-of-00015.safetensors",
912
+ "visual_tokenizer.backbone.trunk.blocks.4.norm_1.weight": "model-00014-of-00015.safetensors",
913
+ "visual_tokenizer.backbone.trunk.blocks.4.norm_2.weight": "model-00014-of-00015.safetensors",
914
+ "visual_tokenizer.backbone.trunk.blocks.5.attn.proj.weight": "model-00014-of-00015.safetensors",
915
+ "visual_tokenizer.backbone.trunk.blocks.5.attn.qkv.weight": "model-00014-of-00015.safetensors",
916
+ "visual_tokenizer.backbone.trunk.blocks.5.mlp.fc1.weight": "model-00014-of-00015.safetensors",
917
+ "visual_tokenizer.backbone.trunk.blocks.5.mlp.fc2.weight": "model-00014-of-00015.safetensors",
918
+ "visual_tokenizer.backbone.trunk.blocks.5.mlp.fc3.weight": "model-00014-of-00015.safetensors",
919
+ "visual_tokenizer.backbone.trunk.blocks.5.norm_1.weight": "model-00014-of-00015.safetensors",
920
+ "visual_tokenizer.backbone.trunk.blocks.5.norm_2.weight": "model-00014-of-00015.safetensors",
921
+ "visual_tokenizer.backbone.trunk.blocks.6.attn.proj.weight": "model-00014-of-00015.safetensors",
922
+ "visual_tokenizer.backbone.trunk.blocks.6.attn.qkv.weight": "model-00014-of-00015.safetensors",
923
+ "visual_tokenizer.backbone.trunk.blocks.6.mlp.fc1.weight": "model-00014-of-00015.safetensors",
924
+ "visual_tokenizer.backbone.trunk.blocks.6.mlp.fc2.weight": "model-00014-of-00015.safetensors",
925
+ "visual_tokenizer.backbone.trunk.blocks.6.mlp.fc3.weight": "model-00014-of-00015.safetensors",
926
+ "visual_tokenizer.backbone.trunk.blocks.6.norm_1.weight": "model-00014-of-00015.safetensors",
927
+ "visual_tokenizer.backbone.trunk.blocks.6.norm_2.weight": "model-00014-of-00015.safetensors",
928
+ "visual_tokenizer.backbone.trunk.blocks.7.attn.proj.weight": "model-00014-of-00015.safetensors",
929
+ "visual_tokenizer.backbone.trunk.blocks.7.attn.qkv.weight": "model-00014-of-00015.safetensors",
930
+ "visual_tokenizer.backbone.trunk.blocks.7.mlp.fc1.weight": "model-00014-of-00015.safetensors",
931
+ "visual_tokenizer.backbone.trunk.blocks.7.mlp.fc2.weight": "model-00014-of-00015.safetensors",
932
+ "visual_tokenizer.backbone.trunk.blocks.7.mlp.fc3.weight": "model-00014-of-00015.safetensors",
933
+ "visual_tokenizer.backbone.trunk.blocks.7.norm_1.weight": "model-00014-of-00015.safetensors",
934
+ "visual_tokenizer.backbone.trunk.blocks.7.norm_2.weight": "model-00014-of-00015.safetensors",
935
+ "visual_tokenizer.backbone.trunk.blocks.8.attn.proj.weight": "model-00014-of-00015.safetensors",
936
+ "visual_tokenizer.backbone.trunk.blocks.8.attn.qkv.weight": "model-00014-of-00015.safetensors",
937
+ "visual_tokenizer.backbone.trunk.blocks.8.mlp.fc1.weight": "model-00014-of-00015.safetensors",
938
+ "visual_tokenizer.backbone.trunk.blocks.8.mlp.fc2.weight": "model-00014-of-00015.safetensors",
939
+ "visual_tokenizer.backbone.trunk.blocks.8.mlp.fc3.weight": "model-00014-of-00015.safetensors",
940
+ "visual_tokenizer.backbone.trunk.blocks.8.norm_1.weight": "model-00014-of-00015.safetensors",
941
+ "visual_tokenizer.backbone.trunk.blocks.8.norm_2.weight": "model-00014-of-00015.safetensors",
942
+ "visual_tokenizer.backbone.trunk.blocks.9.attn.proj.weight": "model-00014-of-00015.safetensors",
943
+ "visual_tokenizer.backbone.trunk.blocks.9.attn.qkv.weight": "model-00014-of-00015.safetensors",
944
+ "visual_tokenizer.backbone.trunk.blocks.9.mlp.fc1.weight": "model-00014-of-00015.safetensors",
945
+ "visual_tokenizer.backbone.trunk.blocks.9.mlp.fc2.weight": "model-00014-of-00015.safetensors",
946
+ "visual_tokenizer.backbone.trunk.blocks.9.mlp.fc3.weight": "model-00014-of-00015.safetensors",
947
+ "visual_tokenizer.backbone.trunk.blocks.9.norm_1.weight": "model-00014-of-00015.safetensors",
948
+ "visual_tokenizer.backbone.trunk.blocks.9.norm_2.weight": "model-00014-of-00015.safetensors",
949
+ "visual_tokenizer.backbone.trunk.post_trunk_norm.weight": "model-00014-of-00015.safetensors",
950
+ "visual_tokenizer.head.0.weight": "model-00015-of-00015.safetensors",
951
+ "visual_tokenizer.head.1.bias": "model-00015-of-00015.safetensors",
952
+ "visual_tokenizer.head.1.weight": "model-00015-of-00015.safetensors",
953
+ "vte.weight": "model-00015-of-00015.safetensors"
954
+ }
955
+ }
modeling_aimv2.py ADDED
@@ -0,0 +1,198 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # adapted from https://huggingface.co/apple/aimv2-huge-patch14-448 (modification: add gradient checkpoint support)
2
+ from typing import Optional, Tuple, Union
3
+
4
+ import torch
5
+ from .configuration_aimv2 import AIMv2Config
6
+ from torch import nn
7
+ from torch.nn import functional as F
8
+ from transformers.modeling_outputs import BaseModelOutputWithNoAttention
9
+ from transformers.modeling_utils import PreTrainedModel
10
+
11
+ __all__ = ["AIMv2Model"]
12
+
13
+
14
+ class RMSNorm(nn.Module):
15
+ def __init__(self, dim: int, eps: float = 1e-6):
16
+ super().__init__()
17
+ self.weight = nn.Parameter(torch.ones(dim))
18
+ self.eps = eps
19
+
20
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
21
+ output = self._norm(x.float()).type_as(x)
22
+ return output * self.weight
23
+
24
+ def extra_repr(self) -> str:
25
+ return f"{tuple(self.weight.shape)}, eps={self.eps}"
26
+
27
+ def _norm(self, x: torch.Tensor) -> torch.Tensor:
28
+ return x * torch.rsqrt(x.pow(2).mean(-1, keepdim=True) + self.eps)
29
+
30
+
31
+ class AIMv2SwiGLUFFN(nn.Module):
32
+ def __init__(self, config: AIMv2Config):
33
+ super().__init__()
34
+ hidden_features = config.intermediate_size
35
+ in_features = config.hidden_size
36
+ bias = config.use_bias
37
+
38
+ self.fc1 = nn.Linear(in_features, hidden_features, bias=bias)
39
+ self.fc2 = nn.Linear(hidden_features, in_features, bias=bias)
40
+ self.fc3 = nn.Linear(in_features, hidden_features, bias=bias)
41
+
42
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
43
+ x = F.silu(self.fc1(x)) * self.fc3(x)
44
+ x = self.fc2(x)
45
+ return x
46
+
47
+
48
+ class AIMv2PatchEmbed(nn.Module):
49
+ def __init__(self, config: AIMv2Config):
50
+ super().__init__()
51
+ self.proj = nn.Conv2d(
52
+ config.num_channels,
53
+ config.hidden_size,
54
+ kernel_size=(config.patch_size, config.patch_size),
55
+ stride=(config.patch_size, config.patch_size),
56
+ )
57
+ self.norm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
58
+
59
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
60
+ x = self.proj(x).flatten(2).transpose(1, 2)
61
+ x = self.norm(x)
62
+ return x
63
+
64
+
65
+ class AIMv2ViTPreprocessor(nn.Module):
66
+ def __init__(self, config: AIMv2Config):
67
+ super().__init__()
68
+ num_patches = (config.image_size // config.patch_size) ** 2
69
+
70
+ self.patchifier = AIMv2PatchEmbed(config)
71
+ self.pos_embed = nn.Parameter(torch.zeros((1, num_patches, config.hidden_size)))
72
+
73
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
74
+ tokens = self.patchifier(x)
75
+ _, N, _ = tokens.shape
76
+ pos_embed = self.pos_embed.to(tokens.device)
77
+ tokens = tokens + pos_embed[:, :N]
78
+ return tokens
79
+
80
+
81
+ class AIMv2Attention(nn.Module):
82
+ def __init__(self, config: AIMv2Config):
83
+ super().__init__()
84
+ dim = config.hidden_size
85
+
86
+ self.num_heads = config.num_attention_heads
87
+ self.qkv = nn.Linear(dim, dim * 3, bias=config.qkv_bias)
88
+ self.attn_drop = nn.Dropout(config.attention_dropout)
89
+ self.proj = nn.Linear(dim, dim, bias=config.use_bias)
90
+ self.proj_drop = nn.Dropout(config.projection_dropout)
91
+
92
+ def forward(
93
+ self, x: torch.Tensor, mask: Optional[torch.Tensor] = None
94
+ ) -> torch.Tensor:
95
+ B, N, C = x.shape
96
+ qkv = (
97
+ self.qkv(x)
98
+ .reshape(B, N, 3, self.num_heads, C // self.num_heads)
99
+ .permute(2, 0, 3, 1, 4)
100
+ )
101
+ q, k, v = qkv.unbind(0)
102
+
103
+ x = F.scaled_dot_product_attention(q, k, v, attn_mask=mask)
104
+ x = x.transpose(1, 2).contiguous().reshape(B, N, C)
105
+ x = self.proj(x)
106
+ x = self.proj_drop(x)
107
+ return x
108
+
109
+
110
+ class AIMv2Block(nn.Module):
111
+ def __init__(self, config: AIMv2Config):
112
+ super().__init__()
113
+ self.attn = AIMv2Attention(config)
114
+ self.norm_1 = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
115
+ self.mlp = AIMv2SwiGLUFFN(config)
116
+ self.norm_2 = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
117
+
118
+ def forward(
119
+ self, x: torch.Tensor, mask: Optional[torch.Tensor] = None
120
+ ) -> torch.Tensor:
121
+ x = x + self.attn(self.norm_1(x), mask)
122
+ x = x + self.mlp(self.norm_2(x))
123
+ return x
124
+
125
+
126
+ class AIMv2Transformer(nn.Module):
127
+ def __init__(self, config: AIMv2Config):
128
+ super().__init__()
129
+ self.blocks = nn.ModuleList(
130
+ [AIMv2Block(config) for _ in range(config.num_hidden_layers)]
131
+ )
132
+ self.post_trunk_norm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
133
+ self.gradient_checkpointing = False
134
+
135
+ def forward(
136
+ self,
137
+ tokens: torch.Tensor,
138
+ mask: Optional[torch.Tensor] = None,
139
+ output_hidden_states: bool = False,
140
+ ) -> Tuple[torch.Tensor, Optional[Tuple[torch.Tensor, ...]]]:
141
+ hidden_states = () if output_hidden_states else None
142
+ for block in self.blocks:
143
+ if self.gradient_checkpointing and self.training:
144
+ tokens = self._gradient_checkpointing_func(block.__call__, tokens, mask)
145
+ else:
146
+ tokens = block(tokens, mask)
147
+ if output_hidden_states:
148
+ hidden_states += (tokens,)
149
+ tokens = self.post_trunk_norm(tokens)
150
+ return tokens, hidden_states
151
+
152
+
153
+ class AIMv2PretrainedModel(PreTrainedModel):
154
+ config_class = AIMv2Config
155
+ base_model_prefix = "aimv2"
156
+ supports_gradient_checkpointing = True
157
+ main_input_name = "pixel_values"
158
+ _no_split_modules = ["AIMv2ViTPreprocessor", "AIMv2Block"]
159
+ _supports_sdpa = True
160
+
161
+
162
+ class AIMv2Model(AIMv2PretrainedModel):
163
+ def __init__(self, config: AIMv2Config):
164
+ super().__init__(config)
165
+ self.preprocessor = AIMv2ViTPreprocessor(config)
166
+ self.trunk = AIMv2Transformer(config)
167
+
168
+ def forward(
169
+ self,
170
+ pixel_values: torch.Tensor,
171
+ mask: Optional[torch.Tensor] = None,
172
+ output_hidden_states: Optional[bool] = None,
173
+ return_dict: Optional[bool] = None,
174
+ ) -> Union[
175
+ Tuple[torch.Tensor],
176
+ Tuple[torch.Tensor, Tuple[torch.Tensor, ...]],
177
+ BaseModelOutputWithNoAttention,
178
+ ]:
179
+ if output_hidden_states is None:
180
+ output_hidden_states = self.config.output_hidden_states
181
+ if return_dict is None:
182
+ return_dict = self.config.use_return_dict
183
+
184
+ x = self.preprocessor(pixel_values)
185
+ x, hidden_states = self.trunk(
186
+ x, mask, output_hidden_states=output_hidden_states
187
+ )
188
+
189
+ if not return_dict:
190
+ res = (x,)
191
+ res += (hidden_states,) if output_hidden_states else ()
192
+ return res
193
+
194
+ return BaseModelOutputWithNoAttention(
195
+ last_hidden_state=x,
196
+ hidden_states=hidden_states,
197
+ )
198
+
preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 448,
4
+ "width": 448
5
+ },
6
+ "do_center_crop": false,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 448
26
+ }
27
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
tokenizer_config.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\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 {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "Qwen2Tokenizer",
206
+ "unk_token": null
207
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff