anananan116 commited on
Commit
5b064cd
·
verified ·
1 Parent(s): 8249d39

Upload AtriVLM

Browse files
Files changed (2) hide show
  1. config.json +301 -305
  2. model.safetensors +1 -1
config.json CHANGED
@@ -1,305 +1,301 @@
1
- {
2
- "_name_or_path": "results/checkpoint-2000",
3
- "adjust_embedding_len": 128260,
4
- "architectures": [
5
- "AtriVLM"
6
- ],
7
- "auto_map": {
8
- "AutoConfig": "configuration_llama.VLMConfig",
9
- "AutoModelForCausalLM": "modeling_VLM.AtriVLM"
10
- },
11
- "attention_bias": false,
12
- "attention_dropout": 0.0,
13
- "bos_token_id": 128000,
14
- "encoded_image_dimention": 1024,
15
- "eos_token_id": [
16
- 128001,
17
- 128008,
18
- 128009
19
- ],
20
- "flashattention": false,
21
- "head_dim": 64,
22
- "hidden_act": "silu",
23
- "hidden_size": 2048,
24
- "hugging_face_token": null,
25
- "initializer_range": 0.02,
26
- "intermediate_size": 8192,
27
- "load_vision_model": true,
28
- "lora": false,
29
- "lora_alpha": 256,
30
- "lora_modules": [
31
- "q",
32
- "k",
33
- "v",
34
- "embed_tokens",
35
- "lm_head"
36
- ],
37
- "lora_rank": 256,
38
- "max_position_embeddings": 131072,
39
- "mlp_bias": false,
40
- "model_type": "llama",
41
- "num_attention_heads": 32,
42
- "num_hidden_layers": 16,
43
- "num_key_value_heads": 8,
44
- "num_patches": 64,
45
- "pretrained_model": "results/checkpoint-2000",
46
- "pretrained_vision_model": "openai/clip-vit-large-patch14-336",
47
- "pretraining_tp": 1,
48
- "rms_norm_eps": 1e-05,
49
- "rope_scaling": {
50
- "factor": 32.0,
51
- "high_freq_factor": 4.0,
52
- "low_freq_factor": 1.0,
53
- "original_max_position_embeddings": 8192,
54
- "rope_type": "llama3"
55
- },
56
- "rope_theta": 500000.0,
57
- "special_token_map": {
58
- "Caption": [
59
- "<Caption>",
60
- 128258
61
- ],
62
- "Image": [
63
- "<Image>",
64
- 128256
65
- ],
66
- "Image_End": [
67
- "<Image_End>",
68
- 128257
69
- ],
70
- "Image_Token": [
71
- "<Image_Token>",
72
- 128259
73
- ]
74
- },
75
- "task_type": "CAUSAL_LM",
76
- "tie_word_embeddings": true,
77
- "torch_dtype": "float16",
78
- "transformers_version": "4.46.1",
79
- "use_cache": true,
80
- "visual_config": {
81
- "_attn_implementation_autoset": true,
82
- "_name_or_path": "openai/clip-vit-large-patch14-336",
83
- "add_cross_attention": false,
84
- "architectures": [
85
- "CLIPModel"
86
- ],
87
- "bad_words_ids": null,
88
- "begin_suppress_tokens": null,
89
- "bos_token_id": null,
90
- "chunk_size_feed_forward": 0,
91
- "cross_attention_hidden_size": null,
92
- "decoder_start_token_id": null,
93
- "diversity_penalty": 0.0,
94
- "do_sample": false,
95
- "early_stopping": false,
96
- "encoder_no_repeat_ngram_size": 0,
97
- "eos_token_id": null,
98
- "exponential_decay_length_penalty": null,
99
- "finetuning_task": null,
100
- "forced_bos_token_id": null,
101
- "forced_eos_token_id": null,
102
- "id2label": {
103
- "0": "LABEL_0",
104
- "1": "LABEL_1"
105
- },
106
- "initializer_factor": 1.0,
107
- "is_decoder": false,
108
- "is_encoder_decoder": false,
109
- "label2id": {
110
- "LABEL_0": 0,
111
- "LABEL_1": 1
112
- },
113
- "length_penalty": 1.0,
114
- "logit_scale_init_value": 2.6592,
115
- "max_length": 20,
116
- "min_length": 0,
117
- "model_type": "clip",
118
- "no_repeat_ngram_size": 0,
119
- "num_beam_groups": 1,
120
- "num_beams": 1,
121
- "num_return_sequences": 1,
122
- "output_attentions": false,
123
- "output_hidden_states": false,
124
- "output_scores": false,
125
- "pad_token_id": null,
126
- "prefix": null,
127
- "problem_type": null,
128
- "projection_dim": 768,
129
- "pruned_heads": {},
130
- "remove_invalid_values": false,
131
- "repetition_penalty": 1.0,
132
- "return_dict": true,
133
- "return_dict_in_generate": false,
134
- "sep_token_id": null,
135
- "suppress_tokens": null,
136
- "task_specific_params": null,
137
- "temperature": 1.0,
138
- "text_config": {
139
- "_attn_implementation_autoset": false,
140
- "_name_or_path": "",
141
- "add_cross_attention": false,
142
- "architectures": null,
143
- "attention_dropout": 0.0,
144
- "bad_words_ids": null,
145
- "begin_suppress_tokens": null,
146
- "bos_token_id": 49406,
147
- "chunk_size_feed_forward": 0,
148
- "cross_attention_hidden_size": null,
149
- "decoder_start_token_id": null,
150
- "diversity_penalty": 0.0,
151
- "do_sample": false,
152
- "dropout": 0.0,
153
- "early_stopping": false,
154
- "encoder_no_repeat_ngram_size": 0,
155
- "eos_token_id": 49407,
156
- "exponential_decay_length_penalty": null,
157
- "finetuning_task": null,
158
- "forced_bos_token_id": null,
159
- "forced_eos_token_id": null,
160
- "hidden_act": "quick_gelu",
161
- "hidden_size": 768,
162
- "id2label": {
163
- "0": "LABEL_0",
164
- "1": "LABEL_1"
165
- },
166
- "initializer_factor": 1.0,
167
- "initializer_range": 0.02,
168
- "intermediate_size": 3072,
169
- "is_decoder": false,
170
- "is_encoder_decoder": false,
171
- "label2id": {
172
- "LABEL_0": 0,
173
- "LABEL_1": 1
174
- },
175
- "layer_norm_eps": 1e-05,
176
- "length_penalty": 1.0,
177
- "max_length": 20,
178
- "max_position_embeddings": 77,
179
- "min_length": 0,
180
- "model_type": "clip_text_model",
181
- "no_repeat_ngram_size": 0,
182
- "num_attention_heads": 12,
183
- "num_beam_groups": 1,
184
- "num_beams": 1,
185
- "num_hidden_layers": 12,
186
- "num_return_sequences": 1,
187
- "output_attentions": false,
188
- "output_hidden_states": false,
189
- "output_scores": false,
190
- "pad_token_id": 1,
191
- "prefix": null,
192
- "problem_type": null,
193
- "projection_dim": 768,
194
- "pruned_heads": {},
195
- "remove_invalid_values": false,
196
- "repetition_penalty": 1.0,
197
- "return_dict": true,
198
- "return_dict_in_generate": false,
199
- "sep_token_id": null,
200
- "suppress_tokens": null,
201
- "task_specific_params": null,
202
- "temperature": 1.0,
203
- "tf_legacy_loss": false,
204
- "tie_encoder_decoder": false,
205
- "tie_word_embeddings": true,
206
- "tokenizer_class": null,
207
- "top_k": 50,
208
- "top_p": 1.0,
209
- "torch_dtype": null,
210
- "torchscript": false,
211
- "typical_p": 1.0,
212
- "use_bfloat16": false,
213
- "vocab_size": 49408
214
- },
215
- "tf_legacy_loss": false,
216
- "tie_encoder_decoder": false,
217
- "tie_word_embeddings": true,
218
- "tokenizer_class": null,
219
- "top_k": 50,
220
- "top_p": 1.0,
221
- "torch_dtype": "float32",
222
- "torchscript": false,
223
- "typical_p": 1.0,
224
- "use_bfloat16": false,
225
- "vision_config": {
226
- "_attn_implementation_autoset": false,
227
- "_name_or_path": "",
228
- "add_cross_attention": false,
229
- "architectures": null,
230
- "attention_dropout": 0.0,
231
- "bad_words_ids": null,
232
- "begin_suppress_tokens": null,
233
- "bos_token_id": null,
234
- "chunk_size_feed_forward": 0,
235
- "cross_attention_hidden_size": null,
236
- "decoder_start_token_id": null,
237
- "diversity_penalty": 0.0,
238
- "do_sample": false,
239
- "dropout": 0.0,
240
- "early_stopping": false,
241
- "encoder_no_repeat_ngram_size": 0,
242
- "eos_token_id": null,
243
- "exponential_decay_length_penalty": null,
244
- "finetuning_task": null,
245
- "forced_bos_token_id": null,
246
- "forced_eos_token_id": null,
247
- "hidden_act": "quick_gelu",
248
- "hidden_size": 1024,
249
- "id2label": {
250
- "0": "LABEL_0",
251
- "1": "LABEL_1"
252
- },
253
- "image_size": 336,
254
- "initializer_factor": 1.0,
255
- "initializer_range": 0.02,
256
- "intermediate_size": 4096,
257
- "is_decoder": false,
258
- "is_encoder_decoder": false,
259
- "label2id": {
260
- "LABEL_0": 0,
261
- "LABEL_1": 1
262
- },
263
- "layer_norm_eps": 1e-05,
264
- "length_penalty": 1.0,
265
- "max_length": 20,
266
- "min_length": 0,
267
- "model_type": "clip_vision_model",
268
- "no_repeat_ngram_size": 0,
269
- "num_attention_heads": 16,
270
- "num_beam_groups": 1,
271
- "num_beams": 1,
272
- "num_channels": 3,
273
- "num_hidden_layers": 24,
274
- "num_return_sequences": 1,
275
- "output_attentions": false,
276
- "output_hidden_states": false,
277
- "output_scores": false,
278
- "pad_token_id": null,
279
- "patch_size": 14,
280
- "prefix": null,
281
- "problem_type": null,
282
- "projection_dim": 768,
283
- "pruned_heads": {},
284
- "remove_invalid_values": false,
285
- "repetition_penalty": 1.0,
286
- "return_dict": true,
287
- "return_dict_in_generate": false,
288
- "sep_token_id": null,
289
- "suppress_tokens": null,
290
- "task_specific_params": null,
291
- "temperature": 1.0,
292
- "tf_legacy_loss": false,
293
- "tie_encoder_decoder": false,
294
- "tie_word_embeddings": true,
295
- "tokenizer_class": null,
296
- "top_k": 50,
297
- "top_p": 1.0,
298
- "torch_dtype": null,
299
- "torchscript": false,
300
- "typical_p": 1.0,
301
- "use_bfloat16": false
302
- }
303
- },
304
- "vocab_size": 128260
305
- }
 
1
+ {
2
+ "_name_or_path": "results/checkpoint-5000",
3
+ "adjust_embedding_len": 128260,
4
+ "architectures": [
5
+ "AtriVLM"
6
+ ],
7
+ "attention_bias": false,
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 128000,
10
+ "encoded_image_dimention": 1024,
11
+ "eos_token_id": [
12
+ 128001,
13
+ 128008,
14
+ 128009
15
+ ],
16
+ "flashattention": false,
17
+ "head_dim": 64,
18
+ "hidden_act": "silu",
19
+ "hidden_size": 2048,
20
+ "hugging_face_token": null,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 8192,
23
+ "load_vision_model": true,
24
+ "lora": false,
25
+ "lora_alpha": 256,
26
+ "lora_modules": [
27
+ "q",
28
+ "k",
29
+ "v",
30
+ "embed_tokens",
31
+ "lm_head"
32
+ ],
33
+ "lora_rank": 256,
34
+ "max_position_embeddings": 131072,
35
+ "mlp_bias": false,
36
+ "model_type": "llama",
37
+ "num_attention_heads": 32,
38
+ "num_hidden_layers": 16,
39
+ "num_key_value_heads": 8,
40
+ "num_patches": 64,
41
+ "pretrained_model": "results/checkpoint-5000",
42
+ "pretrained_vision_model": "openai/clip-vit-large-patch14-336",
43
+ "pretraining_tp": 1,
44
+ "rms_norm_eps": 1e-05,
45
+ "rope_scaling": {
46
+ "factor": 32.0,
47
+ "high_freq_factor": 4.0,
48
+ "low_freq_factor": 1.0,
49
+ "original_max_position_embeddings": 8192,
50
+ "rope_type": "llama3"
51
+ },
52
+ "rope_theta": 500000.0,
53
+ "special_token_map": {
54
+ "Caption": [
55
+ "<Caption>",
56
+ 128258
57
+ ],
58
+ "Image": [
59
+ "<Image>",
60
+ 128256
61
+ ],
62
+ "Image_End": [
63
+ "<Image_End>",
64
+ 128257
65
+ ],
66
+ "Image_Token": [
67
+ "<Image_Token>",
68
+ 128259
69
+ ]
70
+ },
71
+ "task_type": "CAUSAL_LM",
72
+ "tie_word_embeddings": true,
73
+ "torch_dtype": "float16",
74
+ "transformers_version": "4.46.1",
75
+ "use_cache": true,
76
+ "visual_config": {
77
+ "_attn_implementation_autoset": true,
78
+ "_name_or_path": "openai/clip-vit-large-patch14-336",
79
+ "add_cross_attention": false,
80
+ "architectures": [
81
+ "CLIPModel"
82
+ ],
83
+ "bad_words_ids": null,
84
+ "begin_suppress_tokens": null,
85
+ "bos_token_id": null,
86
+ "chunk_size_feed_forward": 0,
87
+ "cross_attention_hidden_size": null,
88
+ "decoder_start_token_id": null,
89
+ "diversity_penalty": 0.0,
90
+ "do_sample": false,
91
+ "early_stopping": false,
92
+ "encoder_no_repeat_ngram_size": 0,
93
+ "eos_token_id": null,
94
+ "exponential_decay_length_penalty": null,
95
+ "finetuning_task": null,
96
+ "forced_bos_token_id": null,
97
+ "forced_eos_token_id": null,
98
+ "id2label": {
99
+ "0": "LABEL_0",
100
+ "1": "LABEL_1"
101
+ },
102
+ "initializer_factor": 1.0,
103
+ "is_decoder": false,
104
+ "is_encoder_decoder": false,
105
+ "label2id": {
106
+ "LABEL_0": 0,
107
+ "LABEL_1": 1
108
+ },
109
+ "length_penalty": 1.0,
110
+ "logit_scale_init_value": 2.6592,
111
+ "max_length": 20,
112
+ "min_length": 0,
113
+ "model_type": "clip",
114
+ "no_repeat_ngram_size": 0,
115
+ "num_beam_groups": 1,
116
+ "num_beams": 1,
117
+ "num_return_sequences": 1,
118
+ "output_attentions": false,
119
+ "output_hidden_states": false,
120
+ "output_scores": false,
121
+ "pad_token_id": null,
122
+ "prefix": null,
123
+ "problem_type": null,
124
+ "projection_dim": 768,
125
+ "pruned_heads": {},
126
+ "remove_invalid_values": false,
127
+ "repetition_penalty": 1.0,
128
+ "return_dict": true,
129
+ "return_dict_in_generate": false,
130
+ "sep_token_id": null,
131
+ "suppress_tokens": null,
132
+ "task_specific_params": null,
133
+ "temperature": 1.0,
134
+ "text_config": {
135
+ "_attn_implementation_autoset": false,
136
+ "_name_or_path": "",
137
+ "add_cross_attention": false,
138
+ "architectures": null,
139
+ "attention_dropout": 0.0,
140
+ "bad_words_ids": null,
141
+ "begin_suppress_tokens": null,
142
+ "bos_token_id": 49406,
143
+ "chunk_size_feed_forward": 0,
144
+ "cross_attention_hidden_size": null,
145
+ "decoder_start_token_id": null,
146
+ "diversity_penalty": 0.0,
147
+ "do_sample": false,
148
+ "dropout": 0.0,
149
+ "early_stopping": false,
150
+ "encoder_no_repeat_ngram_size": 0,
151
+ "eos_token_id": 49407,
152
+ "exponential_decay_length_penalty": null,
153
+ "finetuning_task": null,
154
+ "forced_bos_token_id": null,
155
+ "forced_eos_token_id": null,
156
+ "hidden_act": "quick_gelu",
157
+ "hidden_size": 768,
158
+ "id2label": {
159
+ "0": "LABEL_0",
160
+ "1": "LABEL_1"
161
+ },
162
+ "initializer_factor": 1.0,
163
+ "initializer_range": 0.02,
164
+ "intermediate_size": 3072,
165
+ "is_decoder": false,
166
+ "is_encoder_decoder": false,
167
+ "label2id": {
168
+ "LABEL_0": 0,
169
+ "LABEL_1": 1
170
+ },
171
+ "layer_norm_eps": 1e-05,
172
+ "length_penalty": 1.0,
173
+ "max_length": 20,
174
+ "max_position_embeddings": 77,
175
+ "min_length": 0,
176
+ "model_type": "clip_text_model",
177
+ "no_repeat_ngram_size": 0,
178
+ "num_attention_heads": 12,
179
+ "num_beam_groups": 1,
180
+ "num_beams": 1,
181
+ "num_hidden_layers": 12,
182
+ "num_return_sequences": 1,
183
+ "output_attentions": false,
184
+ "output_hidden_states": false,
185
+ "output_scores": false,
186
+ "pad_token_id": 1,
187
+ "prefix": null,
188
+ "problem_type": null,
189
+ "projection_dim": 768,
190
+ "pruned_heads": {},
191
+ "remove_invalid_values": false,
192
+ "repetition_penalty": 1.0,
193
+ "return_dict": true,
194
+ "return_dict_in_generate": false,
195
+ "sep_token_id": null,
196
+ "suppress_tokens": null,
197
+ "task_specific_params": null,
198
+ "temperature": 1.0,
199
+ "tf_legacy_loss": false,
200
+ "tie_encoder_decoder": false,
201
+ "tie_word_embeddings": true,
202
+ "tokenizer_class": null,
203
+ "top_k": 50,
204
+ "top_p": 1.0,
205
+ "torch_dtype": null,
206
+ "torchscript": false,
207
+ "typical_p": 1.0,
208
+ "use_bfloat16": false,
209
+ "vocab_size": 49408
210
+ },
211
+ "tf_legacy_loss": false,
212
+ "tie_encoder_decoder": false,
213
+ "tie_word_embeddings": true,
214
+ "tokenizer_class": null,
215
+ "top_k": 50,
216
+ "top_p": 1.0,
217
+ "torch_dtype": "float32",
218
+ "torchscript": false,
219
+ "typical_p": 1.0,
220
+ "use_bfloat16": false,
221
+ "vision_config": {
222
+ "_attn_implementation_autoset": false,
223
+ "_name_or_path": "",
224
+ "add_cross_attention": false,
225
+ "architectures": null,
226
+ "attention_dropout": 0.0,
227
+ "bad_words_ids": null,
228
+ "begin_suppress_tokens": null,
229
+ "bos_token_id": null,
230
+ "chunk_size_feed_forward": 0,
231
+ "cross_attention_hidden_size": null,
232
+ "decoder_start_token_id": null,
233
+ "diversity_penalty": 0.0,
234
+ "do_sample": false,
235
+ "dropout": 0.0,
236
+ "early_stopping": false,
237
+ "encoder_no_repeat_ngram_size": 0,
238
+ "eos_token_id": null,
239
+ "exponential_decay_length_penalty": null,
240
+ "finetuning_task": null,
241
+ "forced_bos_token_id": null,
242
+ "forced_eos_token_id": null,
243
+ "hidden_act": "quick_gelu",
244
+ "hidden_size": 1024,
245
+ "id2label": {
246
+ "0": "LABEL_0",
247
+ "1": "LABEL_1"
248
+ },
249
+ "image_size": 336,
250
+ "initializer_factor": 1.0,
251
+ "initializer_range": 0.02,
252
+ "intermediate_size": 4096,
253
+ "is_decoder": false,
254
+ "is_encoder_decoder": false,
255
+ "label2id": {
256
+ "LABEL_0": 0,
257
+ "LABEL_1": 1
258
+ },
259
+ "layer_norm_eps": 1e-05,
260
+ "length_penalty": 1.0,
261
+ "max_length": 20,
262
+ "min_length": 0,
263
+ "model_type": "clip_vision_model",
264
+ "no_repeat_ngram_size": 0,
265
+ "num_attention_heads": 16,
266
+ "num_beam_groups": 1,
267
+ "num_beams": 1,
268
+ "num_channels": 3,
269
+ "num_hidden_layers": 24,
270
+ "num_return_sequences": 1,
271
+ "output_attentions": false,
272
+ "output_hidden_states": false,
273
+ "output_scores": false,
274
+ "pad_token_id": null,
275
+ "patch_size": 14,
276
+ "prefix": null,
277
+ "problem_type": null,
278
+ "projection_dim": 768,
279
+ "pruned_heads": {},
280
+ "remove_invalid_values": false,
281
+ "repetition_penalty": 1.0,
282
+ "return_dict": true,
283
+ "return_dict_in_generate": false,
284
+ "sep_token_id": null,
285
+ "suppress_tokens": null,
286
+ "task_specific_params": null,
287
+ "temperature": 1.0,
288
+ "tf_legacy_loss": false,
289
+ "tie_encoder_decoder": false,
290
+ "tie_word_embeddings": true,
291
+ "tokenizer_class": null,
292
+ "top_k": 50,
293
+ "top_p": 1.0,
294
+ "torch_dtype": null,
295
+ "torchscript": false,
296
+ "typical_p": 1.0,
297
+ "use_bfloat16": false
298
+ }
299
+ },
300
+ "vocab_size": 128260
301
+ }
 
 
 
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c3c3983f8c9110a3821c20a71a8303beb492d851f6852fdd0aa07aa9d179cb01
3
  size 3148589970
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:197c6fdc090a5bce6c2821e208ad0363f1fabc16534186cd65dadf14ac2b39e1
3
  size 3148589970