ahishamm commited on
Commit
3290f93
·
verified ·
1 Parent(s): 8b02d7e

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +322 -0
config.json ADDED
@@ -0,0 +1,322 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": false,
3
+ "add_cross_attention": false,
4
+ "architectures": [
5
+ "Idefics3ForConditionalGeneration"
6
+ ],
7
+ "bad_words_ids": null,
8
+ "begin_suppress_tokens": null,
9
+ "bos_token_id": null,
10
+ "chunk_size_feed_forward": 0,
11
+ "cross_attention_hidden_size": null,
12
+ "decoder_start_token_id": null,
13
+ "diversity_penalty": 0.0,
14
+ "do_sample": false,
15
+ "early_stopping": false,
16
+ "encoder_no_repeat_ngram_size": 0,
17
+ "eos_token_id": null,
18
+ "exponential_decay_length_penalty": null,
19
+ "finetuning_task": null,
20
+ "forced_bos_token_id": null,
21
+ "forced_eos_token_id": null,
22
+ "id2label": {
23
+ "0": "LABEL_0",
24
+ "1": "LABEL_1"
25
+ },
26
+ "image_token_id": 49190,
27
+ "is_decoder": false,
28
+ "is_encoder_decoder": false,
29
+ "label2id": {
30
+ "LABEL_0": 0,
31
+ "LABEL_1": 1
32
+ },
33
+ "length_penalty": 1.0,
34
+ "max_length": 20,
35
+ "min_length": 0,
36
+ "model_type": "idefics3",
37
+ "no_repeat_ngram_size": 0,
38
+ "num_beam_groups": 1,
39
+ "num_beams": 1,
40
+ "num_return_sequences": 1,
41
+ "output_attentions": false,
42
+ "output_hidden_states": false,
43
+ "output_scores": false,
44
+ "pad_token_id": 128002,
45
+ "prefix": null,
46
+ "problem_type": null,
47
+ "pruned_heads": {},
48
+ "remove_invalid_values": false,
49
+ "repetition_penalty": 1.0,
50
+ "return_dict": true,
51
+ "return_dict_in_generate": false,
52
+ "scale_factor": 4,
53
+ "sep_token_id": null,
54
+ "suppress_tokens": null,
55
+ "task_specific_params": null,
56
+ "temperature": 1.0,
57
+ "text_config": {
58
+ "vocab_size": 49280,
59
+ "max_position_embeddings": 8192,
60
+ "hidden_size": 576,
61
+ "intermediate_size": 1536,
62
+ "num_hidden_layers": 30,
63
+ "num_attention_heads": 9,
64
+ "num_key_value_heads": 3,
65
+ "hidden_act": "silu",
66
+ "initializer_range": 0.041666666666666664,
67
+ "rms_norm_eps": 1e-05,
68
+ "pretraining_tp": 1,
69
+ "use_cache": true,
70
+ "rope_theta": 100000,
71
+ "rope_scaling": null,
72
+ "attention_bias": false,
73
+ "attention_dropout": 0.0,
74
+ "mlp_bias": false,
75
+ "head_dim": 64,
76
+ "return_dict": true,
77
+ "output_hidden_states": false,
78
+ "output_attentions": false,
79
+ "torchscript": false,
80
+ "torch_dtype": "bfloat16",
81
+ "use_bfloat16": false,
82
+ "tf_legacy_loss": false,
83
+ "pruned_heads": {},
84
+ "tie_word_embeddings": false,
85
+ "chunk_size_feed_forward": 0,
86
+ "is_encoder_decoder": false,
87
+ "is_decoder": false,
88
+ "cross_attention_hidden_size": null,
89
+ "add_cross_attention": false,
90
+ "tie_encoder_decoder": false,
91
+ "max_length": 20,
92
+ "min_length": 0,
93
+ "do_sample": false,
94
+ "early_stopping": false,
95
+ "num_beams": 1,
96
+ "num_beam_groups": 1,
97
+ "diversity_penalty": 0.0,
98
+ "temperature": 1.0,
99
+ "top_k": 50,
100
+ "top_p": 1.0,
101
+ "typical_p": 1.0,
102
+ "repetition_penalty": 1.0,
103
+ "length_penalty": 1.0,
104
+ "no_repeat_ngram_size": 0,
105
+ "encoder_no_repeat_ngram_size": 0,
106
+ "bad_words_ids": null,
107
+ "num_return_sequences": 1,
108
+ "output_scores": false,
109
+ "return_dict_in_generate": false,
110
+ "forced_bos_token_id": null,
111
+ "forced_eos_token_id": null,
112
+ "remove_invalid_values": false,
113
+ "exponential_decay_length_penalty": null,
114
+ "suppress_tokens": null,
115
+ "begin_suppress_tokens": null,
116
+ "architectures": [
117
+ "VLlama3ForCausalLM"
118
+ ],
119
+ "finetuning_task": null,
120
+ "id2label": {
121
+ "0": "LABEL_0",
122
+ "1": "LABEL_1"
123
+ },
124
+ "label2id": {
125
+ "LABEL_0": 0,
126
+ "LABEL_1": 1
127
+ },
128
+ "tokenizer_class": null,
129
+ "prefix": null,
130
+ "bos_token_id": 1,
131
+ "pad_token_id": 2,
132
+ "eos_token_id": 2,
133
+ "sep_token_id": null,
134
+ "decoder_start_token_id": null,
135
+ "task_specific_params": null,
136
+ "problem_type": null,
137
+ "_name_or_path": "None",
138
+ "_attn_implementation_autoset": false,
139
+ "_flash_attn_2_enabled": true,
140
+ "is_llama_config": true,
141
+ "model_type": "llama",
142
+ "neftune_noise_alpha": 0.0,
143
+ "perceiver_config": {
144
+ "_attn_implementation_autoset": false,
145
+ "_name_or_path": "",
146
+ "add_cross_attention": false,
147
+ "architectures": null,
148
+ "attention_dropout": 0.0,
149
+ "bad_words_ids": null,
150
+ "begin_suppress_tokens": null,
151
+ "bos_token_id": null,
152
+ "chunk_size_feed_forward": 0,
153
+ "cross_attention_hidden_size": null,
154
+ "decoder_start_token_id": null,
155
+ "diversity_penalty": 0.0,
156
+ "do_sample": false,
157
+ "early_stopping": false,
158
+ "encoder_no_repeat_ngram_size": 0,
159
+ "eos_token_id": null,
160
+ "exponential_decay_length_penalty": null,
161
+ "finetuning_task": null,
162
+ "forced_bos_token_id": null,
163
+ "forced_eos_token_id": null,
164
+ "hidden_act": "silu",
165
+ "id2label": {
166
+ "0": "LABEL_0",
167
+ "1": "LABEL_1"
168
+ },
169
+ "is_decoder": false,
170
+ "is_encoder_decoder": false,
171
+ "label2id": {
172
+ "LABEL_0": 0,
173
+ "LABEL_1": 1
174
+ },
175
+ "length_penalty": 1.0,
176
+ "max_length": 20,
177
+ "min_length": 0,
178
+ "model_type": "vllama3",
179
+ "no_repeat_ngram_size": 0,
180
+ "num_beam_groups": 1,
181
+ "num_beams": 1,
182
+ "num_key_value_heads": 1,
183
+ "num_return_sequences": 1,
184
+ "output_attentions": false,
185
+ "output_hidden_states": false,
186
+ "output_scores": false,
187
+ "pad_token_id": null,
188
+ "prefix": null,
189
+ "problem_type": null,
190
+ "pruned_heads": {},
191
+ "qk_layer_norms_perceiver": false,
192
+ "remove_invalid_values": false,
193
+ "repetition_penalty": 1.0,
194
+ "resampler_depth": 6,
195
+ "resampler_head_dim": 96,
196
+ "resampler_n_heads": 16,
197
+ "resampler_n_latents": 64,
198
+ "return_dict": true,
199
+ "return_dict_in_generate": false,
200
+ "sep_token_id": null,
201
+ "suppress_tokens": null,
202
+ "task_specific_params": null,
203
+ "temperature": 1.0,
204
+ "tf_legacy_loss": false,
205
+ "tie_encoder_decoder": false,
206
+ "tie_word_embeddings": true,
207
+ "tokenizer_class": null,
208
+ "top_k": 50,
209
+ "top_p": 1.0,
210
+ "torch_dtype": null,
211
+ "torchscript": false,
212
+ "transformers_version": "4.46.0",
213
+ "typical_p": 1.0,
214
+ "use_bfloat16": false
215
+ },
216
+ "pixel_shuffle_factor": 4,
217
+ "qk_layer_norms": false,
218
+ "rope_interleaved": false,
219
+ "transformers.js_config": {
220
+ "kv_cache_dtype": {
221
+ "fp16": "float16",
222
+ "q4f16": "float16"
223
+ }
224
+ },
225
+ "use_resampler": false
226
+ },
227
+ "tf_legacy_loss": false,
228
+ "tie_encoder_decoder": false,
229
+ "tie_word_embeddings": false,
230
+ "tokenizer_class": null,
231
+ "top_k": 50,
232
+ "top_p": 1.0,
233
+ "torch_dtype": "bfloat16",
234
+ "torchscript": false,
235
+ "transformers_version": "4.49.0",
236
+ "typical_p": 1.0,
237
+ "use_bfloat16": false,
238
+ "use_cache": true,
239
+ "vision_config": {
240
+ "return_dict": true,
241
+ "output_hidden_states": false,
242
+ "output_attentions": false,
243
+ "torchscript": false,
244
+ "torch_dtype": "bfloat16",
245
+ "use_bfloat16": false,
246
+ "tf_legacy_loss": false,
247
+ "pruned_heads": {},
248
+ "tie_word_embeddings": false,
249
+ "chunk_size_feed_forward": 0,
250
+ "is_encoder_decoder": false,
251
+ "is_decoder": false,
252
+ "cross_attention_hidden_size": null,
253
+ "add_cross_attention": false,
254
+ "tie_encoder_decoder": false,
255
+ "max_length": 20,
256
+ "min_length": 0,
257
+ "do_sample": false,
258
+ "early_stopping": false,
259
+ "num_beams": 1,
260
+ "num_beam_groups": 1,
261
+ "diversity_penalty": 0.0,
262
+ "temperature": 1.0,
263
+ "top_k": 50,
264
+ "top_p": 1.0,
265
+ "typical_p": 1.0,
266
+ "repetition_penalty": 1.0,
267
+ "length_penalty": 1.0,
268
+ "no_repeat_ngram_size": 0,
269
+ "encoder_no_repeat_ngram_size": 0,
270
+ "bad_words_ids": null,
271
+ "num_return_sequences": 1,
272
+ "output_scores": false,
273
+ "return_dict_in_generate": false,
274
+ "forced_bos_token_id": null,
275
+ "forced_eos_token_id": null,
276
+ "remove_invalid_values": false,
277
+ "exponential_decay_length_penalty": null,
278
+ "suppress_tokens": null,
279
+ "begin_suppress_tokens": null,
280
+ "architectures": null,
281
+ "finetuning_task": null,
282
+ "id2label": {
283
+ "0": "LABEL_0",
284
+ "1": "LABEL_1"
285
+ },
286
+ "label2id": {
287
+ "LABEL_0": 0,
288
+ "LABEL_1": 1
289
+ },
290
+ "tokenizer_class": null,
291
+ "prefix": null,
292
+ "bos_token_id": null,
293
+ "pad_token_id": null,
294
+ "eos_token_id": null,
295
+ "sep_token_id": null,
296
+ "decoder_start_token_id": null,
297
+ "task_specific_params": null,
298
+ "problem_type": null,
299
+ "_name_or_path": "",
300
+ "_attn_implementation_autoset": false,
301
+ "max_image_size": {
302
+ "longest_edge": 512
303
+ },
304
+ "model_type": "idefics3_vision",
305
+ "size": {
306
+ "longest_edge": 2048
307
+ },
308
+ "use_base_siglip": true,
309
+ "hidden_size": 768,
310
+ "intermediate_size": 3072,
311
+ "num_hidden_layers": 12,
312
+ "num_attention_heads": 12,
313
+ "num_channels": 3,
314
+ "patch_size": 16,
315
+ "image_size": 512,
316
+ "attention_dropout": 0.0,
317
+ "layer_norm_eps": 1e-06,
318
+ "hidden_act": "gelu_pytorch_tanh",
319
+ "initializer_range": 0.02
320
+ },
321
+ "vocab_size": 49280
322
+ }