Hugging Face's logo

U4R
/
Chimera-Reasoner-4B

Safetensors
chimera
custom_code
Model card Files Community
Potentialts commited on
Commit
5f8ed3a
·
1 Parent(s): 814186b

upload model weights

Browse files
Files changed (11) hide show
  1. added_tokens.json +27 -0
  2. config.json +628 -0
  3. generation_config.json +9 -0
  4. model-00001-of-00003.safetensors +3 -0
  5. model-00002-of-00003.safetensors +3 -0
  6. model-00003-of-00003.safetensors +3 -0
  7. model.safetensors.index.json +0 -0
  8. special_tokens_map.json +46 -0
  9. tokenizer.json +0 -0
  10. tokenizer.model +3 -0
  11. tokenizer_config.json +259 -0
added_tokens.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</box>": 32019,
3
+ "</domain>": 32021,
4
+ "</img>": 32012,
5
+ "</quad>": 32015,
6
+ "</ref>": 32017,
7
+ "<DOMAIN_0_CONTEXT>": 32022,
8
+ "<DOMAIN_1_CONTEXT>": 32023,
9
+ "<DOMAIN_2_CONTEXT>": 32024,
10
+ "<IMG_CONTEXT>": 32013,
11
+ "<box>": 32018,
12
+ "<domain>": 32020,
13
+ "<img>": 32011,
14
+ "<quad>": 32014,
15
+ "<ref>": 32016,
16
+ "<|assistant|>": 32001,
17
+ "<|endoftext|>": 32000,
18
+ "<|end|>": 32007,
19
+ "<|placeholder1|>": 32002,
20
+ "<|placeholder2|>": 32003,
21
+ "<|placeholder3|>": 32004,
22
+ "<|placeholder4|>": 32005,
23
+ "<|placeholder5|>": 32008,
24
+ "<|placeholder6|>": 32009,
25
+ "<|system|>": 32006,
26
+ "<|user|>": 32010
27
+ }
config.json ADDED
@@ -0,0 +1,628 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_commit_hash": null,
3
+ "_name_or_path": "/mnt/workspace/Chimera/checkpoints/solva-reasoner-4B",
4
+ "architectures": [
5
+ "ChimeraChatModel"
6
+ ],
7
+ "auto_map": {
8
+ "AutoConfig": "configuration_chimera.ChimeraChatConfig",
9
+ "AutoModel": "modeling_chimera.ChimeraChatModel",
10
+ "AutoModelForCausalLM": "modeling_chimera.ChimeraChatModel"
11
+ },
12
+ "downsample_ratio": 0.5,
13
+ "dynamic_image_size": true,
14
+ "expert_encoder_config": {
15
+ "_attn_implementation_internal": null,
16
+ "_commit_hash": null,
17
+ "_name_or_path": "",
18
+ "add_cross_attention": false,
19
+ "architectures": null,
20
+ "bad_words_ids": null,
21
+ "begin_suppress_tokens": null,
22
+ "bos_token_id": null,
23
+ "chunk_size_feed_forward": 0,
24
+ "config_list": [
25
+ {
26
+ "config": {
27
+ "_name_or_path": "/cpfs01/shared/ADLab/datasets/science_dataset/solva_modules/table_kosmos",
28
+ "add_cross_attention": false,
29
+ "architectures": [
30
+ "Kosmos2_5VisionModel"
31
+ ],
32
+ "attention_dropout": 0.0,
33
+ "bad_words_ids": null,
34
+ "begin_suppress_tokens": null,
35
+ "bos_token_id": null,
36
+ "chunk_size_feed_forward": 0,
37
+ "cross_attention_hidden_size": null,
38
+ "d_ff": 3968,
39
+ "d_kv": 64,
40
+ "decoder_start_token_id": null,
41
+ "dense_act_fn": "gelu_new",
42
+ "diversity_penalty": 0.0,
43
+ "do_sample": false,
44
+ "dropout_rate": 0.0,
45
+ "early_stopping": false,
46
+ "encoder_no_repeat_ngram_size": 0,
47
+ "eos_token_id": null,
48
+ "exponential_decay_length_penalty": null,
49
+ "finetuning_task": null,
50
+ "forced_bos_token_id": null,
51
+ "forced_eos_token_id": null,
52
+ "hidden_size": 1536,
53
+ "id2label": {
54
+ "0": "LABEL_0",
55
+ "1": "LABEL_1"
56
+ },
57
+ "initializer_factor": 1.0,
58
+ "initializer_range": 1e-10,
59
+ "is_decoder": false,
60
+ "is_encoder_decoder": false,
61
+ "label2id": {
62
+ "LABEL_0": 0,
63
+ "LABEL_1": 1
64
+ },
65
+ "layer_norm_eps": 1e-06,
66
+ "length_penalty": 1.0,
67
+ "max_length": 4096,
68
+ "min_length": 0,
69
+ "model_type": "kosmos_2_5_vision_model",
70
+ "no_repeat_ngram_size": 0,
71
+ "num_attention_heads": 24,
72
+ "num_beam_groups": 1,
73
+ "num_beams": 1,
74
+ "num_hidden_layers": 18,
75
+ "num_return_sequences": 1,
76
+ "output_attentions": false,
77
+ "output_hidden_states": false,
78
+ "output_scores": false,
79
+ "pad_token_id": null,
80
+ "patch_embed_hidden_size": 768,
81
+ "prefix": null,
82
+ "problem_type": null,
83
+ "pruned_heads": {},
84
+ "remove_invalid_values": false,
85
+ "repetition_penalty": 1.0,
86
+ "return_dict": true,
87
+ "return_dict_in_generate": false,
88
+ "sep_token_id": null,
89
+ "seq_len": 4096,
90
+ "suppress_tokens": null,
91
+ "task_specific_params": null,
92
+ "temperature": 1.0,
93
+ "tf_legacy_loss": false,
94
+ "tie_encoder_decoder": false,
95
+ "tie_word_embeddings": true,
96
+ "tokenizer_class": null,
97
+ "top_k": 50,
98
+ "top_p": 1.0,
99
+ "torch_dtype": "float32",
100
+ "torchscript": false,
101
+ "transformers_version": "4.44.2",
102
+ "typical_p": 1.0,
103
+ "use_bfloat16": false
104
+ },
105
+ "domain": "table",
106
+ "index": 0,
107
+ "select_layer": -1
108
+ },
109
+ {
110
+ "config": {
111
+ "_name_or_path": "/cpfs01/shared/ADLab/datasets/science_dataset/solva_modules/math_clip",
112
+ "add_cross_attention": false,
113
+ "architectures": null,
114
+ "attention_dropout": 0.0,
115
+ "bad_words_ids": null,
116
+ "begin_suppress_tokens": null,
117
+ "bos_token_id": null,
118
+ "chunk_size_feed_forward": 0,
119
+ "cross_attention_hidden_size": null,
120
+ "decoder_start_token_id": null,
121
+ "diversity_penalty": 0.0,
122
+ "do_sample": false,
123
+ "dropout": 0.0,
124
+ "early_stopping": false,
125
+ "encoder_no_repeat_ngram_size": 0,
126
+ "eos_token_id": null,
127
+ "exponential_decay_length_penalty": null,
128
+ "finetuning_task": null,
129
+ "forced_bos_token_id": null,
130
+ "forced_eos_token_id": null,
131
+ "hidden_act": "quick_gelu",
132
+ "hidden_size": 1024,
133
+ "id2label": {
134
+ "0": "LABEL_0",
135
+ "1": "LABEL_1"
136
+ },
137
+ "image_size": 336,
138
+ "initializer_factor": 1.0,
139
+ "initializer_range": 0.02,
140
+ "intermediate_size": 4096,
141
+ "is_decoder": false,
142
+ "is_encoder_decoder": false,
143
+ "label2id": {
144
+ "LABEL_0": 0,
145
+ "LABEL_1": 1
146
+ },
147
+ "layer_norm_eps": 1e-05,
148
+ "length_penalty": 1.0,
149
+ "max_length": 20,
150
+ "min_length": 0,
151
+ "model_type": "clip_vision_model",
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_dim": 768,
167
+ "pruned_heads": {},
168
+ "remove_invalid_values": false,
169
+ "repetition_penalty": 1.0,
170
+ "return_dict": true,
171
+ "return_dict_in_generate": false,
172
+ "sep_token_id": null,
173
+ "suppress_tokens": null,
174
+ "task_specific_params": null,
175
+ "temperature": 1.0,
176
+ "tf_legacy_loss": false,
177
+ "tie_encoder_decoder": false,
178
+ "tie_word_embeddings": true,
179
+ "tokenizer_class": null,
180
+ "top_k": 50,
181
+ "top_p": 1.0,
182
+ "torch_dtype": null,
183
+ "torchscript": false,
184
+ "transformers_version": "4.44.2",
185
+ "typical_p": 1.0,
186
+ "use_bfloat16": false
187
+ },
188
+ "domain": "math",
189
+ "index": 1,
190
+ "select_layer": -2
191
+ },
192
+ {
193
+ "config": {
194
+ "_name_or_path": "/cpfs01/shared/ADLab/datasets/science_dataset/solva_modules/chart_p2s",
195
+ "add_cross_attention": false,
196
+ "architectures": [
197
+ "Pix2StructVisionModel"
198
+ ],
199
+ "attention_dropout": 0.0,
200
+ "bad_words_ids": null,
201
+ "begin_suppress_tokens": null,
202
+ "bos_token_id": null,
203
+ "chunk_size_feed_forward": 0,
204
+ "cross_attention_hidden_size": null,
205
+ "d_ff": 3968,
206
+ "d_kv": 64,
207
+ "decoder_start_token_id": null,
208
+ "dense_act_fn": "gelu_new",
209
+ "diversity_penalty": 0.0,
210
+ "do_sample": false,
211
+ "dropout_rate": 0.0,
212
+ "early_stopping": false,
213
+ "encoder_no_repeat_ngram_size": 0,
214
+ "eos_token_id": null,
215
+ "exponential_decay_length_penalty": null,
216
+ "finetuning_task": null,
217
+ "forced_bos_token_id": null,
218
+ "forced_eos_token_id": null,
219
+ "hidden_size": 1536,
220
+ "id2label": {
221
+ "0": "LABEL_0",
222
+ "1": "LABEL_1"
223
+ },
224
+ "initializer_factor": 1.0,
225
+ "initializer_range": 0.02,
226
+ "is_decoder": false,
227
+ "is_encoder_decoder": false,
228
+ "label2id": {
229
+ "LABEL_0": 0,
230
+ "LABEL_1": 1
231
+ },
232
+ "layer_norm_bias": false,
233
+ "layer_norm_eps": 1e-06,
234
+ "length_penalty": 1.0,
235
+ "max_length": 20,
236
+ "min_length": 0,
237
+ "model_type": "pix2struct_vision_model",
238
+ "no_repeat_ngram_size": 0,
239
+ "num_attention_heads": 24,
240
+ "num_beam_groups": 1,
241
+ "num_beams": 1,
242
+ "num_channels": 3,
243
+ "num_hidden_layers": 18,
244
+ "num_return_sequences": 1,
245
+ "output_attentions": false,
246
+ "output_hidden_states": false,
247
+ "output_scores": false,
248
+ "pad_token_id": null,
249
+ "patch_embed_hidden_size": 768,
250
+ "patch_size": 16,
251
+ "prefix": null,
252
+ "problem_type": null,
253
+ "projection_dim": 768,
254
+ "pruned_heads": {},
255
+ "relative_attention_max_distance": 128,
256
+ "relative_attention_num_buckets": 32,
257
+ "remove_invalid_values": false,
258
+ "repetition_penalty": 1.0,
259
+ "return_dict": true,
260
+ "return_dict_in_generate": false,
261
+ "sep_token_id": null,
262
+ "seq_len": 4096,
263
+ "suppress_tokens": null,
264
+ "task_specific_params": null,
265
+ "temperature": 1.0,
266
+ "tf_legacy_loss": false,
267
+ "tie_encoder_decoder": false,
268
+ "tie_word_embeddings": true,
269
+ "tokenizer_class": null,
270
+ "top_k": 50,
271
+ "top_p": 1.0,
272
+ "torch_dtype": "float32",
273
+ "torchscript": false,
274
+ "transformers_version": "4.44.2",
275
+ "typical_p": 1.0,
276
+ "use_bfloat16": false
277
+ },
278
+ "domain": "chart",
279
+ "index": 2,
280
+ "select_layer": -1
281
+ }
282
+ ],
283
+ "cross_attention_hidden_size": null,
284
+ "decoder_start_token_id": null,
285
+ "diversity_penalty": 0.0,
286
+ "do_sample": false,
287
+ "early_stopping": false,
288
+ "encoder_no_repeat_ngram_size": 0,
289
+ "eos_token_id": null,
290
+ "exponential_decay_length_penalty": null,
291
+ "finetuning_task": null,
292
+ "forced_bos_token_id": null,
293
+ "forced_eos_token_id": null,
294
+ "id2label": {
295
+ "0": "LABEL_0",
296
+ "1": "LABEL_1"
297
+ },
298
+ "is_decoder": false,
299
+ "is_encoder_decoder": false,
300
+ "label2id": {
301
+ "LABEL_0": 0,
302
+ "LABEL_1": 1
303
+ },
304
+ "length_penalty": 1.0,
305
+ "llm_hidden_size": 3072,
306
+ "max_length": 20,
307
+ "min_length": 0,
308
+ "no_repeat_ngram_size": 0,
309
+ "num_beam_groups": 1,
310
+ "num_beams": 1,
311
+ "num_encoder": 3,
312
+ "num_return_sequences": 1,
313
+ "output_attentions": false,
314
+ "output_hidden_states": false,
315
+ "output_scores": false,
316
+ "pad_token_id": null,
317
+ "prefix": null,
318
+ "problem_type": null,
319
+ "pruned_heads": {},
320
+ "remove_invalid_values": false,
321
+ "repetition_penalty": 1.0,
322
+ "return_dict": true,
323
+ "return_dict_in_generate": false,
324
+ "sep_token_id": null,
325
+ "separate_load": false,
326
+ "suppress_tokens": null,
327
+ "task_specific_params": null,
328
+ "temperature": 1.0,
329
+ "tf_legacy_loss": false,
330
+ "tie_encoder_decoder": false,
331
+ "tie_word_embeddings": true,
332
+ "tokenizer_class": null,
333
+ "top_k": 50,
334
+ "top_p": 1.0,
335
+ "torch_dtype": null,
336
+ "torchscript": false,
337
+ "transformers_version": null,
338
+ "typical_p": 1.0,
339
+ "use_bfloat16": false
340
+ },
341
+ "force_image_size": 448,
342
+ "llm_config": {
343
+ "_name_or_path": "microsoft/Phi-3-mini-128k-instruct",
344
+ "add_cross_attention": false,
345
+ "architectures": [
346
+ "Phi3ForCausalLM"
347
+ ],
348
+ "attention_dropout": 0.0,
349
+ "auto_map": {
350
+ "AutoConfig": "configuration_phi3.Phi3Config",
351
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
352
+ },
353
+ "bad_words_ids": null,
354
+ "begin_suppress_tokens": null,
355
+ "bos_token_id": 1,
356
+ "chunk_size_feed_forward": 0,
357
+ "cross_attention_hidden_size": null,
358
+ "decoder_start_token_id": null,
359
+ "diversity_penalty": 0.0,
360
+ "do_sample": false,
361
+ "early_stopping": false,
362
+ "embd_pdrop": 0.0,
363
+ "encoder_no_repeat_ngram_size": 0,
364
+ "eos_token_id": 32000,
365
+ "exponential_decay_length_penalty": null,
366
+ "finetuning_task": null,
367
+ "forced_bos_token_id": null,
368
+ "forced_eos_token_id": null,
369
+ "hidden_act": "silu",
370
+ "hidden_size": 3072,
371
+ "id2label": {
372
+ "0": "LABEL_0",
373
+ "1": "LABEL_1"
374
+ },
375
+ "initializer_range": 0.02,
376
+ "intermediate_size": 8192,
377
+ "is_decoder": false,
378
+ "is_encoder_decoder": false,
379
+ "label2id": {
380
+ "LABEL_0": 0,
381
+ "LABEL_1": 1
382
+ },
383
+ "length_penalty": 1.0,
384
+ "max_length": 20,
385
+ "max_position_embeddings": 131072,
386
+ "min_length": 0,
387
+ "model_type": "phi3",
388
+ "no_repeat_ngram_size": 0,
389
+ "num_attention_heads": 32,
390
+ "num_beam_groups": 1,
391
+ "num_beams": 1,
392
+ "num_hidden_layers": 32,
393
+ "num_key_value_heads": 32,
394
+ "num_return_sequences": 1,
395
+ "original_max_position_embeddings": 4096,
396
+ "output_attentions": false,
397
+ "output_hidden_states": false,
398
+ "output_scores": false,
399
+ "pad_token_id": 32000,
400
+ "prefix": null,
401
+ "problem_type": null,
402
+ "pruned_heads": {},
403
+ "remove_invalid_values": false,
404
+ "repetition_penalty": 1.0,
405
+ "resid_pdrop": 0.0,
406
+ "return_dict": true,
407
+ "return_dict_in_generate": false,
408
+ "rms_norm_eps": 1e-05,
409
+ "rope_scaling": {
410
+ "long_factor": [
411
+ 1.0299999713897705,
412
+ 1.0499999523162842,
413
+ 1.0499999523162842,
414
+ 1.0799999237060547,
415
+ 1.2299998998641968,
416
+ 1.2299998998641968,
417
+ 1.2999999523162842,
418
+ 1.4499999284744263,
419
+ 1.5999999046325684,
420
+ 1.6499998569488525,
421
+ 1.8999998569488525,
422
+ 2.859999895095825,
423
+ 3.68999981880188,
424
+ 5.419999599456787,
425
+ 5.489999771118164,
426
+ 5.489999771118164,
427
+ 9.09000015258789,
428
+ 11.579999923706055,
429
+ 15.65999984741211,
430
+ 15.769999504089355,
431
+ 15.789999961853027,
432
+ 18.360000610351562,
433
+ 21.989999771118164,
434
+ 23.079999923706055,
435
+ 30.009998321533203,
436
+ 32.35000228881836,
437
+ 32.590003967285156,
438
+ 35.56000518798828,
439
+ 39.95000457763672,
440
+ 53.840003967285156,
441
+ 56.20000457763672,
442
+ 57.95000457763672,
443
+ 59.29000473022461,
444
+ 59.77000427246094,
445
+ 59.920005798339844,
446
+ 61.190006256103516,
447
+ 61.96000671386719,
448
+ 62.50000762939453,
449
+ 63.3700065612793,
450
+ 63.48000717163086,
451
+ 63.48000717163086,
452
+ 63.66000747680664,
453
+ 63.850006103515625,
454
+ 64.08000946044922,
455
+ 64.760009765625,
456
+ 64.80001068115234,
457
+ 64.81001281738281,
458
+ 64.81001281738281
459
+ ],
460
+ "short_factor": [
461
+ 1.05,
462
+ 1.05,
463
+ 1.05,
464
+ 1.1,
465
+ 1.1,
466
+ 1.1500000000000001,
467
+ 1.2000000000000002,
468
+ 1.2500000000000002,
469
+ 1.3000000000000003,
470
+ 1.3500000000000003,
471
+ 1.5000000000000004,
472
+ 2.000000000000001,
473
+ 2.000000000000001,
474
+ 2.000000000000001,
475
+ 2.000000000000001,
476
+ 2.000000000000001,
477
+ 2.000000000000001,
478
+ 2.000000000000001,
479
+ 2.000000000000001,
480
+ 2.000000000000001,
481
+ 2.000000000000001,
482
+ 2.000000000000001,
483
+ 2.000000000000001,
484
+ 2.000000000000001,
485
+ 2.000000000000001,
486
+ 2.000000000000001,
487
+ 2.000000000000001,
488
+ 2.000000000000001,
489
+ 2.000000000000001,
490
+ 2.000000000000001,
491
+ 2.000000000000001,
492
+ 2.000000000000001,
493
+ 2.0500000000000007,
494
+ 2.0500000000000007,
495
+ 2.0500000000000007,
496
+ 2.1000000000000005,
497
+ 2.1000000000000005,
498
+ 2.1000000000000005,
499
+ 2.1500000000000004,
500
+ 2.1500000000000004,
501
+ 2.3499999999999996,
502
+ 2.549999999999999,
503
+ 2.5999999999999988,
504
+ 2.5999999999999988,
505
+ 2.7499999999999982,
506
+ 2.849999999999998,
507
+ 2.849999999999998,
508
+ 2.9499999999999975
509
+ ],
510
+ "type": "su"
511
+ },
512
+ "rope_theta": 10000.0,
513
+ "sep_token_id": null,
514
+ "sliding_window": 262144,
515
+ "suppress_tokens": null,
516
+ "task_specific_params": null,
517
+ "temperature": 1.0,
518
+ "tf_legacy_loss": false,
519
+ "tie_encoder_decoder": false,
520
+ "tie_word_embeddings": false,
521
+ "tokenizer_class": null,
522
+ "top_k": 50,
523
+ "top_p": 1.0,
524
+ "torch_dtype": "bfloat16",
525
+ "torchscript": false,
526
+ "transformers_version": "4.44.2",
527
+ "typical_p": 1.0,
528
+ "use_bfloat16": true,
529
+ "use_cache": false,
530
+ "vocab_size": 32025
531
+ },
532
+ "max_dynamic_patch": 6,
533
+ "min_dynamic_patch": 1,
534
+ "model_type": "chimera",
535
+ "pad2square": false,
536
+ "ps_version": "solva-chat",
537
+ "select_layer": -1,
538
+ "template": "solva-chat-phi3",
539
+ "torch_dtype": "bfloat16",
540
+ "transformers_version": null,
541
+ "use_backbone_lora": 0,
542
+ "use_llm_lora": 0,
543
+ "use_thumbnail": true,
544
+ "vision_config": {
545
+ "_name_or_path": "",
546
+ "add_cross_attention": false,
547
+ "architectures": [
548
+ "InternVisionModel"
549
+ ],
550
+ "attention_dropout": 0.0,
551
+ "bad_words_ids": null,
552
+ "begin_suppress_tokens": null,
553
+ "bos_token_id": null,
554
+ "chunk_size_feed_forward": 0,
555
+ "cross_attention_hidden_size": null,
556
+ "decoder_start_token_id": null,
557
+ "diversity_penalty": 0.0,
558
+ "do_sample": false,
559
+ "drop_path_rate": 0.1,
560
+ "dropout": 0.0,
561
+ "early_stopping": false,
562
+ "encoder_no_repeat_ngram_size": 0,
563
+ "eos_token_id": null,
564
+ "exponential_decay_length_penalty": null,
565
+ "finetuning_task": null,
566
+ "forced_bos_token_id": null,
567
+ "forced_eos_token_id": null,
568
+ "hidden_act": "gelu",
569
+ "hidden_size": 1024,
570
+ "id2label": {
571
+ "0": "LABEL_0",
572
+ "1": "LABEL_1"
573
+ },
574
+ "image_size": 448,
575
+ "initializer_factor": 1.0,
576
+ "initializer_range": 0.02,
577
+ "intermediate_size": 4096,
578
+ "is_decoder": false,
579
+ "is_encoder_decoder": false,
580
+ "label2id": {
581
+ "LABEL_0": 0,
582
+ "LABEL_1": 1
583
+ },
584
+ "layer_norm_eps": 1e-06,
585
+ "length_penalty": 1.0,
586
+ "max_length": 20,
587
+ "min_length": 0,
588
+ "model_type": "intern_vit_6b",
589
+ "no_repeat_ngram_size": 0,
590
+ "norm_type": "layer_norm",
591
+ "num_attention_heads": 16,
592
+ "num_beam_groups": 1,
593
+ "num_beams": 1,
594
+ "num_channels": 3,
595
+ "num_hidden_layers": 24,
596
+ "num_return_sequences": 1,
597
+ "output_attentions": false,
598
+ "output_hidden_states": false,
599
+ "output_scores": false,
600
+ "pad_token_id": null,
601
+ "patch_size": 14,
602
+ "prefix": null,
603
+ "problem_type": null,
604
+ "pruned_heads": {},
605
+ "qk_normalization": false,
606
+ "qkv_bias": true,
607
+ "remove_invalid_values": false,
608
+ "repetition_penalty": 1.0,
609
+ "return_dict": true,
610
+ "return_dict_in_generate": false,
611
+ "sep_token_id": null,
612
+ "suppress_tokens": null,
613
+ "task_specific_params": null,
614
+ "temperature": 1.0,
615
+ "tf_legacy_loss": false,
616
+ "tie_encoder_decoder": false,
617
+ "tie_word_embeddings": true,
618
+ "tokenizer_class": null,
619
+ "top_k": 50,
620
+ "top_p": 1.0,
621
+ "torch_dtype": "bfloat16",
622
+ "torchscript": false,
623
+ "transformers_version": "4.44.2",
624
+ "typical_p": 1.0,
625
+ "use_bfloat16": true,
626
+ "use_flash_attn": true
627
+ }
628
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": [
4
+ 2,
5
+ 32000,
6
+ 32007
7
+ ],
8
+ "transformers_version": "4.44.2"
9
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70184329f9c10dfa6b916e708c5aff16a77f07a6752511e836e7746b1431f1e3
3
+ size 4979577304
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a0ef166c6873ccf4421525cd5c5b159cf09e26ea6a8287f395d58a0ec1653b2
3
+ size 4983120928
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b1ad38eaea6eb07c5c7879bd9bf60222dfaa483074c57a8354a70dadd640dfb
3
+ size 1071361592
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<img>",
4
+ "</img>",
5
+ "<IMG_CONTEXT>",
6
+ "<quad>",
7
+ "</quad>",
8
+ "<ref>",
9
+ "</ref>",
10
+ "<box>",
11
+ "</box>",
12
+ "<domain>",
13
+ "</domain>",
14
+ "<DOMAIN_0_CONTEXT>",
15
+ "<DOMAIN_1_CONTEXT>",
16
+ "<DOMAIN_2_CONTEXT>"
17
+ ],
18
+ "bos_token": {
19
+ "content": "<s>",
20
+ "lstrip": false,
21
+ "normalized": false,
22
+ "rstrip": false,
23
+ "single_word": false
24
+ },
25
+ "eos_token": {
26
+ "content": "</s>",
27
+ "lstrip": false,
28
+ "normalized": false,
29
+ "rstrip": true,
30
+ "single_word": false
31
+ },
32
+ "pad_token": {
33
+ "content": "</s>",
34
+ "lstrip": false,
35
+ "normalized": false,
36
+ "rstrip": true,
37
+ "single_word": false
38
+ },
39
+ "unk_token": {
40
+ "content": "<unk>",
41
+ "lstrip": false,
42
+ "normalized": false,
43
+ "rstrip": false,
44
+ "single_word": false
45
+ }
46
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,259 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": true,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<|endoftext|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<|assistant|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": true,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "32002": {
47
+ "content": "<|placeholder1|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": true,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "32003": {
55
+ "content": "<|placeholder2|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": true,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "32004": {
63
+ "content": "<|placeholder3|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": true,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "32005": {
71
+ "content": "<|placeholder4|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": true,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "32006": {
79
+ "content": "<|system|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": true,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "32007": {
87
+ "content": "<|end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": true,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "32008": {
95
+ "content": "<|placeholder5|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": true,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "32009": {
103
+ "content": "<|placeholder6|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": true,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "32010": {
111
+ "content": "<|user|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": true,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "32011": {
119
+ "content": "<img>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": true
125
+ },
126
+ "32012": {
127
+ "content": "</img>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": true
133
+ },
134
+ "32013": {
135
+ "content": "<IMG_CONTEXT>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": true
141
+ },
142
+ "32014": {
143
+ "content": "<quad>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": true
149
+ },
150
+ "32015": {
151
+ "content": "</quad>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": true
157
+ },
158
+ "32016": {
159
+ "content": "<ref>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": true
165
+ },
166
+ "32017": {
167
+ "content": "</ref>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": true
173
+ },
174
+ "32018": {
175
+ "content": "<box>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": true
181
+ },
182
+ "32019": {
183
+ "content": "</box>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": true
189
+ },
190
+ "32020": {
191
+ "content": "<domain>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": true
197
+ },
198
+ "32021": {
199
+ "content": "</domain>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": true
205
+ },
206
+ "32022": {
207
+ "content": "<DOMAIN_0_CONTEXT>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": true
213
+ },
214
+ "32023": {
215
+ "content": "<DOMAIN_1_CONTEXT>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "32024": {
223
+ "content": "<DOMAIN_2_CONTEXT>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": true
229
+ }
230
+ },
231
+ "additional_special_tokens": [
232
+ "<img>",
233
+ "</img>",
234
+ "<IMG_CONTEXT>",
235
+ "<quad>",
236
+ "</quad>",
237
+ "<ref>",
238
+ "</ref>",
239
+ "<box>",
240
+ "</box>",
241
+ "<domain>",
242
+ "</domain>",
243
+ "<DOMAIN_0_CONTEXT>",
244
+ "<DOMAIN_1_CONTEXT>",
245
+ "<DOMAIN_2_CONTEXT>"
246
+ ],
247
+ "bos_token": "<s>",
248
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') %}{{'<|user|>' + '\n' + message['content'] + '<|end|>' + '\n' + '<|assistant|>' + '\n'}}{% elif (message['role'] == 'assistant') %}{{message['content'] + '<|end|>' + '\n'}}{% endif %}{% endfor %}",
249
+ "clean_up_tokenization_spaces": false,
250
+ "eos_token": "</s>",
251
+ "legacy": false,
252
+ "model_max_length": 8192,
253
+ "pad_token": "</s>",
254
+ "sp_model_kwargs": {},
255
+ "spaces_between_special_tokens": false,
256
+ "tokenizer_class": "LlamaTokenizer",
257
+ "unk_token": "<unk>",
258
+ "use_default_system_prompt": false
259
+ }