Hugging Face's logo

wwwyyy
/
pretrain-RealSimEEFJoint-eef-relative

Safetensors
beingh
Model card Files
xet
Community
1
wwwyyy commited on
Commit
ccab54b
·
verified ·
1 Parent(s): c7034c5

Upload 9 files

Browse files
Files changed (10) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +60 -0
  3. chat_template.jinja +6 -0
  4. config.json +433 -0
  5. merges.txt +0 -0
  6. model.safetensors +3 -0
  7. special_tokens_map.json +40 -0
  8. tokenizer.json +3 -0
  9. tokenizer_config.json +506 -0
  10. 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
added_tokens.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</box>": 151677,
3
+ "</img>": 151670,
4
+ "</quad>": 151673,
5
+ "</ref>": 151675,
6
+ "</think>": 151668,
7
+ "</tool_call>": 151658,
8
+ "</tool_response>": 151666,
9
+ "<ACTION_TOKEN_0>": 151685,
10
+ "<ACTION_TOKEN_10>": 151695,
11
+ "<ACTION_TOKEN_11>": 151696,
12
+ "<ACTION_TOKEN_12>": 151697,
13
+ "<ACTION_TOKEN_13>": 151698,
14
+ "<ACTION_TOKEN_14>": 151699,
15
+ "<ACTION_TOKEN_15>": 151700,
16
+ "<ACTION_TOKEN_1>": 151686,
17
+ "<ACTION_TOKEN_2>": 151687,
18
+ "<ACTION_TOKEN_3>": 151688,
19
+ "<ACTION_TOKEN_4>": 151689,
20
+ "<ACTION_TOKEN_5>": 151690,
21
+ "<ACTION_TOKEN_6>": 151691,
22
+ "<ACTION_TOKEN_7>": 151692,
23
+ "<ACTION_TOKEN_8>": 151693,
24
+ "<ACTION_TOKEN_9>": 151694,
25
+ "<IMG_CONTEXT>": 151671,
26
+ "<PROP_CONTEXT>": 151684,
27
+ "<box>": 151676,
28
+ "<img>": 151669,
29
+ "<quad>": 151672,
30
+ "<ref>": 151674,
31
+ "<think>": 151667,
32
+ "<tool_call>": 151657,
33
+ "<tool_response>": 151665,
34
+ "<vision_end>": 151679,
35
+ "<vision_start>": 151678,
36
+ "<|action_end|>": 151681,
37
+ "<|action_start|>": 151680,
38
+ "<|box_end|>": 151649,
39
+ "<|box_start|>": 151648,
40
+ "<|endoftext|>": 151643,
41
+ "<|file_sep|>": 151664,
42
+ "<|fim_middle|>": 151660,
43
+ "<|fim_pad|>": 151662,
44
+ "<|fim_prefix|>": 151659,
45
+ "<|fim_suffix|>": 151661,
46
+ "<|im_end|>": 151645,
47
+ "<|im_start|>": 151644,
48
+ "<|image_pad|>": 151655,
49
+ "<|object_ref_end|>": 151647,
50
+ "<|object_ref_start|>": 151646,
51
+ "<|quad_end|>": 151651,
52
+ "<|quad_start|>": 151650,
53
+ "<|repo_name|>": 151663,
54
+ "<|state_end|>": 151683,
55
+ "<|state_start|>": 151682,
56
+ "<|video_pad|>": 151656,
57
+ "<|vision_end|>": 151653,
58
+ "<|vision_pad|>": 151654,
59
+ "<|vision_start|>": 151652
60
+ }
chat_template.jinja ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {% for message in messages %}{{'<|im_start|>' + message['role'] + '
2
+ '}}{% if message['content'] is string %}{{ message['content'] }}{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' %}{{ '<image>
3
+ ' }}{% elif content['type'] == 'video' %}{{ '<video>
4
+ ' }}{% elif content['type'] == 'text' %}{{ content['text'] }}{% endif %}{% endfor %}{% endif %}{{'<|im_end|>
5
+ '}}{% endfor %}{% if add_generation_prompt %}{{'<|im_start|>assistant
6
+ ' }}{% endif %}
config.json ADDED
@@ -0,0 +1,433 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "return_dict": true,
3
+ "output_hidden_states": false,
4
+ "torchscript": false,
5
+ "dtype": null,
6
+ "pruned_heads": {},
7
+ "tie_word_embeddings": true,
8
+ "chunk_size_feed_forward": 0,
9
+ "is_encoder_decoder": false,
10
+ "is_decoder": false,
11
+ "cross_attention_hidden_size": null,
12
+ "add_cross_attention": false,
13
+ "tie_encoder_decoder": false,
14
+ "architectures": null,
15
+ "finetuning_task": null,
16
+ "id2label": {
17
+ "0": "LABEL_0",
18
+ "1": "LABEL_1"
19
+ },
20
+ "label2id": {
21
+ "LABEL_0": 0,
22
+ "LABEL_1": 1
23
+ },
24
+ "task_specific_params": null,
25
+ "problem_type": null,
26
+ "tokenizer_class": null,
27
+ "prefix": null,
28
+ "bos_token_id": null,
29
+ "pad_token_id": null,
30
+ "eos_token_id": null,
31
+ "sep_token_id": null,
32
+ "decoder_start_token_id": null,
33
+ "max_length": 20,
34
+ "min_length": 0,
35
+ "do_sample": false,
36
+ "early_stopping": false,
37
+ "num_beams": 1,
38
+ "temperature": 1.0,
39
+ "top_k": 50,
40
+ "top_p": 1.0,
41
+ "typical_p": 1.0,
42
+ "repetition_penalty": 1.0,
43
+ "length_penalty": 1.0,
44
+ "no_repeat_ngram_size": 0,
45
+ "encoder_no_repeat_ngram_size": 0,
46
+ "bad_words_ids": null,
47
+ "num_return_sequences": 1,
48
+ "output_scores": false,
49
+ "return_dict_in_generate": false,
50
+ "forced_bos_token_id": null,
51
+ "forced_eos_token_id": null,
52
+ "remove_invalid_values": false,
53
+ "exponential_decay_length_penalty": null,
54
+ "suppress_tokens": null,
55
+ "begin_suppress_tokens": null,
56
+ "num_beam_groups": 1,
57
+ "diversity_penalty": 0.0,
58
+ "_name_or_path": "",
59
+ "transformers_version": "4.57.1",
60
+ "use_dig_flow": false,
61
+ "dig_flow_num_projections": 32,
62
+ "dig_flow_lambda": 0.1,
63
+ "dig_flow_use_stop_gradient": true,
64
+ "dig_flow_refinement_iters": 1,
65
+ "dig_flow_gate_temperature": 2.0,
66
+ "use_training_time_rtc": false,
67
+ "simulated_delay": 0,
68
+ "rtc_delay_exp_weight": true,
69
+ "use_inference_prefix_overwrite": true,
70
+ "tf_legacy_loss": false,
71
+ "use_bfloat16": false,
72
+ "llm_config": {
73
+ "vocab_size": 151701,
74
+ "max_position_embeddings": 40960,
75
+ "hidden_size": 2048,
76
+ "intermediate_size": 6144,
77
+ "num_hidden_layers": 28,
78
+ "num_attention_heads": 16,
79
+ "use_sliding_window": false,
80
+ "sliding_window": null,
81
+ "max_window_layers": 28,
82
+ "num_key_value_heads": 8,
83
+ "hidden_act": "silu",
84
+ "initializer_range": 0.02,
85
+ "rms_norm_eps": 1e-06,
86
+ "use_cache": false,
87
+ "rope_theta": 1000000,
88
+ "rope_scaling": null,
89
+ "attention_bias": false,
90
+ "attention_dropout": 0.0,
91
+ "is_causal": true,
92
+ "return_dict": true,
93
+ "output_hidden_states": false,
94
+ "torchscript": false,
95
+ "dtype": "bfloat16",
96
+ "pruned_heads": {},
97
+ "tie_word_embeddings": false,
98
+ "chunk_size_feed_forward": 0,
99
+ "is_encoder_decoder": false,
100
+ "is_decoder": false,
101
+ "cross_attention_hidden_size": null,
102
+ "add_cross_attention": false,
103
+ "tie_encoder_decoder": false,
104
+ "architectures": [
105
+ "Qwen3ForCausalLM"
106
+ ],
107
+ "finetuning_task": null,
108
+ "id2label": {
109
+ "0": "LABEL_0",
110
+ "1": "LABEL_1"
111
+ },
112
+ "label2id": {
113
+ "LABEL_0": 0,
114
+ "LABEL_1": 1
115
+ },
116
+ "task_specific_params": null,
117
+ "problem_type": null,
118
+ "tokenizer_class": null,
119
+ "prefix": null,
120
+ "bos_token_id": 151643,
121
+ "pad_token_id": null,
122
+ "eos_token_id": 151645,
123
+ "sep_token_id": null,
124
+ "decoder_start_token_id": null,
125
+ "max_length": 20,
126
+ "min_length": 0,
127
+ "do_sample": false,
128
+ "early_stopping": false,
129
+ "num_beams": 1,
130
+ "temperature": 1.0,
131
+ "top_k": 50,
132
+ "top_p": 1.0,
133
+ "typical_p": 1.0,
134
+ "repetition_penalty": 1.0,
135
+ "length_penalty": 1.0,
136
+ "no_repeat_ngram_size": 0,
137
+ "encoder_no_repeat_ngram_size": 0,
138
+ "bad_words_ids": null,
139
+ "num_return_sequences": 1,
140
+ "output_scores": false,
141
+ "return_dict_in_generate": false,
142
+ "forced_bos_token_id": null,
143
+ "forced_eos_token_id": null,
144
+ "remove_invalid_values": false,
145
+ "exponential_decay_length_penalty": null,
146
+ "suppress_tokens": null,
147
+ "begin_suppress_tokens": null,
148
+ "num_beam_groups": 1,
149
+ "diversity_penalty": 0.0,
150
+ "_name_or_path": "/root/codespace/checkpoints/Qwen3-1.7B",
151
+ "debug": false,
152
+ "ep_size": 1,
153
+ "head_dim": 128,
154
+ "micro_forward": false,
155
+ "model_type": "qwen3",
156
+ "skip_checkpoint": false,
157
+ "use_deepep": false,
158
+ "tf_legacy_loss": false,
159
+ "use_bfloat16": false,
160
+ "qk_norm": true,
161
+ "layer_module": "Qwen3MoTDecoderLayer",
162
+ "freeze_und": false,
163
+ "use_mot": true,
164
+ "expert_config": {
165
+ "vocab_size": 151936,
166
+ "max_position_embeddings": 40960,
167
+ "hidden_size": 1024,
168
+ "intermediate_size": 3072,
169
+ "num_hidden_layers": 28,
170
+ "num_attention_heads": 16,
171
+ "use_sliding_window": false,
172
+ "sliding_window": null,
173
+ "max_window_layers": 28,
174
+ "num_key_value_heads": 8,
175
+ "hidden_act": "silu",
176
+ "initializer_range": 0.02,
177
+ "rms_norm_eps": 1e-06,
178
+ "use_cache": true,
179
+ "rope_theta": 1000000,
180
+ "rope_scaling": null,
181
+ "attention_bias": false,
182
+ "attention_dropout": 0.0,
183
+ "is_causal": true,
184
+ "return_dict": true,
185
+ "output_hidden_states": false,
186
+ "torchscript": false,
187
+ "dtype": "bfloat16",
188
+ "pruned_heads": {},
189
+ "tie_word_embeddings": true,
190
+ "chunk_size_feed_forward": 0,
191
+ "is_encoder_decoder": false,
192
+ "is_decoder": false,
193
+ "cross_attention_hidden_size": null,
194
+ "add_cross_attention": false,
195
+ "tie_encoder_decoder": false,
196
+ "architectures": [
197
+ "Qwen3ForCausalLM"
198
+ ],
199
+ "finetuning_task": null,
200
+ "id2label": {
201
+ "0": "LABEL_0",
202
+ "1": "LABEL_1"
203
+ },
204
+ "label2id": {
205
+ "LABEL_0": 0,
206
+ "LABEL_1": 1
207
+ },
208
+ "task_specific_params": null,
209
+ "problem_type": null,
210
+ "tokenizer_class": null,
211
+ "prefix": null,
212
+ "bos_token_id": 151643,
213
+ "pad_token_id": null,
214
+ "eos_token_id": 151645,
215
+ "sep_token_id": null,
216
+ "decoder_start_token_id": null,
217
+ "max_length": 20,
218
+ "min_length": 0,
219
+ "do_sample": false,
220
+ "early_stopping": false,
221
+ "num_beams": 1,
222
+ "temperature": 1.0,
223
+ "top_k": 50,
224
+ "top_p": 1.0,
225
+ "typical_p": 1.0,
226
+ "repetition_penalty": 1.0,
227
+ "length_penalty": 1.0,
228
+ "no_repeat_ngram_size": 0,
229
+ "encoder_no_repeat_ngram_size": 0,
230
+ "bad_words_ids": null,
231
+ "num_return_sequences": 1,
232
+ "output_scores": false,
233
+ "return_dict_in_generate": false,
234
+ "forced_bos_token_id": null,
235
+ "forced_eos_token_id": null,
236
+ "remove_invalid_values": false,
237
+ "exponential_decay_length_penalty": null,
238
+ "suppress_tokens": null,
239
+ "begin_suppress_tokens": null,
240
+ "num_beam_groups": 1,
241
+ "diversity_penalty": 0.0,
242
+ "_name_or_path": "",
243
+ "head_dim": 128,
244
+ "model_type": "qwen3",
245
+ "tf_legacy_loss": false,
246
+ "use_bfloat16": false,
247
+ "qk_norm": true,
248
+ "layer_module": "Qwen3DecoderLayer",
249
+ "freeze_und": false,
250
+ "moe_layer_ids": [],
251
+ "num_experts": 5,
252
+ "num_experts_per_tok": 2,
253
+ "output_attentions": false
254
+ },
255
+ "attn_implementation": "flash_attention_2",
256
+ "output_attentions": false
257
+ },
258
+ "vit_config": {
259
+ "return_dict": true,
260
+ "output_hidden_states": false,
261
+ "torchscript": false,
262
+ "dtype": "bfloat16",
263
+ "pruned_heads": {},
264
+ "tie_word_embeddings": true,
265
+ "chunk_size_feed_forward": 0,
266
+ "is_encoder_decoder": false,
267
+ "is_decoder": false,
268
+ "cross_attention_hidden_size": null,
269
+ "add_cross_attention": false,
270
+ "tie_encoder_decoder": false,
271
+ "architectures": [
272
+ "InternVisionModel"
273
+ ],
274
+ "finetuning_task": null,
275
+ "id2label": {
276
+ "0": "LABEL_0",
277
+ "1": "LABEL_1"
278
+ },
279
+ "label2id": {
280
+ "LABEL_0": 0,
281
+ "LABEL_1": 1
282
+ },
283
+ "task_specific_params": null,
284
+ "problem_type": null,
285
+ "tokenizer_class": null,
286
+ "prefix": null,
287
+ "bos_token_id": null,
288
+ "pad_token_id": null,
289
+ "eos_token_id": null,
290
+ "sep_token_id": null,
291
+ "decoder_start_token_id": null,
292
+ "max_length": 20,
293
+ "min_length": 0,
294
+ "do_sample": false,
295
+ "early_stopping": false,
296
+ "num_beams": 1,
297
+ "temperature": 1.0,
298
+ "top_k": 50,
299
+ "top_p": 1.0,
300
+ "typical_p": 1.0,
301
+ "repetition_penalty": 1.0,
302
+ "length_penalty": 1.0,
303
+ "no_repeat_ngram_size": 0,
304
+ "encoder_no_repeat_ngram_size": 0,
305
+ "bad_words_ids": null,
306
+ "num_return_sequences": 1,
307
+ "output_scores": false,
308
+ "return_dict_in_generate": false,
309
+ "forced_bos_token_id": null,
310
+ "forced_eos_token_id": null,
311
+ "remove_invalid_values": false,
312
+ "exponential_decay_length_penalty": null,
313
+ "suppress_tokens": null,
314
+ "begin_suppress_tokens": null,
315
+ "num_beam_groups": 1,
316
+ "diversity_penalty": 0.0,
317
+ "_name_or_path": "",
318
+ "auto_map": {
319
+ "AutoConfig": "configuration_intern_vit.InternVisionConfig",
320
+ "AutoModel": "modeling_intern_vit.InternVisionModel"
321
+ },
322
+ "model_type": "intern_vit_6b",
323
+ "use_fa3": false,
324
+ "tf_legacy_loss": false,
325
+ "use_bfloat16": false,
326
+ "hidden_size": 1024,
327
+ "intermediate_size": 4096,
328
+ "dropout": 0.0,
329
+ "drop_path_rate": 0.0,
330
+ "num_hidden_layers": 24,
331
+ "num_attention_heads": 16,
332
+ "num_channels": 3,
333
+ "patch_size": 14,
334
+ "image_size": 224,
335
+ "initializer_range": 0.02,
336
+ "initializer_factor": 1.0,
337
+ "attention_dropout": 0.0,
338
+ "layer_norm_eps": 1e-06,
339
+ "hidden_act": "gelu",
340
+ "norm_type": "layer_norm",
341
+ "qkv_bias": true,
342
+ "qk_normalization": false,
343
+ "use_flash_attn": true,
344
+ "select_layer": -1,
345
+ "downsample_ratio": 0.5,
346
+ "output_attentions": false
347
+ },
348
+ "connector_arch": "internvl_connector",
349
+ "template": "internvl2_5",
350
+ "select_layer": -1,
351
+ "force_image_size": 224,
352
+ "downsample_ratio": 0.5,
353
+ "max_state_dim": 64,
354
+ "max_action_dim": 32,
355
+ "action_chunk_length": 15,
356
+ "loss_func": "l1",
357
+ "gen_action_type": "prop_hidden",
358
+ "layer_select_for_action": -1,
359
+ "action_token_num": 15,
360
+ "k_token_one_action": 1,
361
+ "learnable_action_query": false,
362
+ "bid_attn": false,
363
+ "use_features": "raw_features",
364
+ "prompt_template": "long",
365
+ "use_expert": true,
366
+ "use_flow_matching": true,
367
+ "attn_mode": "causal",
368
+ "prop_context_token_id": 151684,
369
+ "action_token_ids": [
370
+ 151685,
371
+ 151686,
372
+ 151687,
373
+ 151688,
374
+ 151689,
375
+ 151690,
376
+ 151691,
377
+ 151692,
378
+ 151693,
379
+ 151694,
380
+ 151695,
381
+ 151696,
382
+ 151697,
383
+ 151698,
384
+ 151699,
385
+ 151700
386
+ ],
387
+ "max_num_embodiments": 32,
388
+ "noise_beta_alpha": 1.5,
389
+ "noise_beta_beta": 1.0,
390
+ "num_timestep_buckets": 1000,
391
+ "noise_s": 0.999,
392
+ "num_inference_timesteps": 4,
393
+ "action_head_cfg": {
394
+ "action_dim": 32,
395
+ "action_horizon": 15,
396
+ "add_pos_embed": true,
397
+ "backbone_embedding_dim": 2048,
398
+ "diffusion_model_cfg": {
399
+ "attention_head_dim": 48,
400
+ "cross_attention_dim": 2048,
401
+ "dropout": 0.2,
402
+ "final_dropout": true,
403
+ "interleave_self_attention": true,
404
+ "norm_type": "ada_norm",
405
+ "num_attention_heads": 32,
406
+ "num_layers": 16,
407
+ "output_dim": 1024
408
+ },
409
+ "hidden_size": 1024,
410
+ "input_embedding_dim": 1536,
411
+ "max_state_dim": 64,
412
+ "max_action_dim": 32,
413
+ "model_dtype": "float32",
414
+ "noise_beta_alpha": 1.5,
415
+ "noise_beta_beta": 1.0,
416
+ "noise_s": 0.999,
417
+ "num_inference_timesteps": 4,
418
+ "num_target_vision_tokens": 32,
419
+ "num_timestep_buckets": 1000,
420
+ "max_num_embodiments": 1,
421
+ "use_vlln": true,
422
+ "vl_self_attention_cfg": {
423
+ "attention_head_dim": 64,
424
+ "dropout": 0.2,
425
+ "final_dropout": true,
426
+ "num_attention_heads": 32,
427
+ "num_layers": 4
428
+ }
429
+ },
430
+ "img_context_token_id": 151671,
431
+ "model_type": "beingh",
432
+ "output_attentions": false
433
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24253b867acfc4ab0d015cfb44c4c0c802546ae6d0f00b576cc8567e29226684
3
+ size 11174148080
special_tokens_map.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "<img>",
17
+ "</img>",
18
+ "<IMG_CONTEXT>",
19
+ "<quad>",
20
+ "</quad>",
21
+ "<ref>",
22
+ "</ref>",
23
+ "<box>",
24
+ "</box>"
25
+ ],
26
+ "eos_token": {
27
+ "content": "<|im_end|>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ },
33
+ "pad_token": {
34
+ "content": "<|endoftext|>",
35
+ "lstrip": false,
36
+ "normalized": false,
37
+ "rstrip": false,
38
+ "single_word": false
39
+ }
40
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9a9c73ae68252161d0994241d80dea8376cab0c4c0719785d3611fdfa5ccf4a
3
+ size 11428731
tokenizer_config.json ADDED
@@ -0,0 +1,506 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<|im_start|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "151645": {
23
+ "content": "<|im_end|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "151646": {
31
+ "content": "<|object_ref_start|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|object_ref_end|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "151648": {
47
+ "content": "<|box_start|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "151649": {
55
+ "content": "<|box_end|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": true,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": false
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": true,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": false
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ },
182
+ "151665": {
183
+ "content": "<tool_response>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": false
189
+ },
190
+ "151666": {
191
+ "content": "</tool_response>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": false
197
+ },
198
+ "151667": {
199
+ "content": "<think>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": false
205
+ },
206
+ "151668": {
207
+ "content": "</think>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": false
213
+ },
214
+ "151669": {
215
+ "content": "<img>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "151670": {
223
+ "content": "</img>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": true
229
+ },
230
+ "151671": {
231
+ "content": "<IMG_CONTEXT>",
232
+ "lstrip": false,
233
+ "normalized": true,
234
+ "rstrip": false,
235
+ "single_word": false,
236
+ "special": false
237
+ },
238
+ "151672": {
239
+ "content": "<quad>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": false,
243
+ "single_word": false,
244
+ "special": true
245
+ },
246
+ "151673": {
247
+ "content": "</quad>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": false,
251
+ "single_word": false,
252
+ "special": true
253
+ },
254
+ "151674": {
255
+ "content": "<ref>",
256
+ "lstrip": false,
257
+ "normalized": true,
258
+ "rstrip": false,
259
+ "single_word": false,
260
+ "special": false
261
+ },
262
+ "151675": {
263
+ "content": "</ref>",
264
+ "lstrip": false,
265
+ "normalized": true,
266
+ "rstrip": false,
267
+ "single_word": false,
268
+ "special": false
269
+ },
270
+ "151676": {
271
+ "content": "<box>",
272
+ "lstrip": false,
273
+ "normalized": true,
274
+ "rstrip": false,
275
+ "single_word": false,
276
+ "special": false
277
+ },
278
+ "151677": {
279
+ "content": "</box>",
280
+ "lstrip": false,
281
+ "normalized": true,
282
+ "rstrip": false,
283
+ "single_word": false,
284
+ "special": false
285
+ },
286
+ "151678": {
287
+ "content": "<vision_start>",
288
+ "lstrip": false,
289
+ "normalized": true,
290
+ "rstrip": false,
291
+ "single_word": false,
292
+ "special": false
293
+ },
294
+ "151679": {
295
+ "content": "<vision_end>",
296
+ "lstrip": false,
297
+ "normalized": true,
298
+ "rstrip": false,
299
+ "single_word": false,
300
+ "special": false
301
+ },
302
+ "151680": {
303
+ "content": "<|action_start|>",
304
+ "lstrip": false,
305
+ "normalized": true,
306
+ "rstrip": false,
307
+ "single_word": false,
308
+ "special": false
309
+ },
310
+ "151681": {
311
+ "content": "<|action_end|>",
312
+ "lstrip": false,
313
+ "normalized": true,
314
+ "rstrip": false,
315
+ "single_word": false,
316
+ "special": false
317
+ },
318
+ "151682": {
319
+ "content": "<|state_start|>",
320
+ "lstrip": false,
321
+ "normalized": true,
322
+ "rstrip": false,
323
+ "single_word": false,
324
+ "special": false
325
+ },
326
+ "151683": {
327
+ "content": "<|state_end|>",
328
+ "lstrip": false,
329
+ "normalized": true,
330
+ "rstrip": false,
331
+ "single_word": false,
332
+ "special": false
333
+ },
334
+ "151684": {
335
+ "content": "<PROP_CONTEXT>",
336
+ "lstrip": false,
337
+ "normalized": true,
338
+ "rstrip": false,
339
+ "single_word": false,
340
+ "special": false
341
+ },
342
+ "151685": {
343
+ "content": "<ACTION_TOKEN_0>",
344
+ "lstrip": false,
345
+ "normalized": true,
346
+ "rstrip": false,
347
+ "single_word": false,
348
+ "special": false
349
+ },
350
+ "151686": {
351
+ "content": "<ACTION_TOKEN_1>",
352
+ "lstrip": false,
353
+ "normalized": true,
354
+ "rstrip": false,
355
+ "single_word": false,
356
+ "special": false
357
+ },
358
+ "151687": {
359
+ "content": "<ACTION_TOKEN_2>",
360
+ "lstrip": false,
361
+ "normalized": true,
362
+ "rstrip": false,
363
+ "single_word": false,
364
+ "special": false
365
+ },
366
+ "151688": {
367
+ "content": "<ACTION_TOKEN_3>",
368
+ "lstrip": false,
369
+ "normalized": true,
370
+ "rstrip": false,
371
+ "single_word": false,
372
+ "special": false
373
+ },
374
+ "151689": {
375
+ "content": "<ACTION_TOKEN_4>",
376
+ "lstrip": false,
377
+ "normalized": true,
378
+ "rstrip": false,
379
+ "single_word": false,
380
+ "special": false
381
+ },
382
+ "151690": {
383
+ "content": "<ACTION_TOKEN_5>",
384
+ "lstrip": false,
385
+ "normalized": true,
386
+ "rstrip": false,
387
+ "single_word": false,
388
+ "special": false
389
+ },
390
+ "151691": {
391
+ "content": "<ACTION_TOKEN_6>",
392
+ "lstrip": false,
393
+ "normalized": true,
394
+ "rstrip": false,
395
+ "single_word": false,
396
+ "special": false
397
+ },
398
+ "151692": {
399
+ "content": "<ACTION_TOKEN_7>",
400
+ "lstrip": false,
401
+ "normalized": true,
402
+ "rstrip": false,
403
+ "single_word": false,
404
+ "special": false
405
+ },
406
+ "151693": {
407
+ "content": "<ACTION_TOKEN_8>",
408
+ "lstrip": false,
409
+ "normalized": true,
410
+ "rstrip": false,
411
+ "single_word": false,
412
+ "special": false
413
+ },
414
+ "151694": {
415
+ "content": "<ACTION_TOKEN_9>",
416
+ "lstrip": false,
417
+ "normalized": true,
418
+ "rstrip": false,
419
+ "single_word": false,
420
+ "special": false
421
+ },
422
+ "151695": {
423
+ "content": "<ACTION_TOKEN_10>",
424
+ "lstrip": false,
425
+ "normalized": true,
426
+ "rstrip": false,
427
+ "single_word": false,
428
+ "special": false
429
+ },
430
+ "151696": {
431
+ "content": "<ACTION_TOKEN_11>",
432
+ "lstrip": false,
433
+ "normalized": true,
434
+ "rstrip": false,
435
+ "single_word": false,
436
+ "special": false
437
+ },
438
+ "151697": {
439
+ "content": "<ACTION_TOKEN_12>",
440
+ "lstrip": false,
441
+ "normalized": true,
442
+ "rstrip": false,
443
+ "single_word": false,
444
+ "special": false
445
+ },
446
+ "151698": {
447
+ "content": "<ACTION_TOKEN_13>",
448
+ "lstrip": false,
449
+ "normalized": true,
450
+ "rstrip": false,
451
+ "single_word": false,
452
+ "special": false
453
+ },
454
+ "151699": {
455
+ "content": "<ACTION_TOKEN_14>",
456
+ "lstrip": false,
457
+ "normalized": true,
458
+ "rstrip": false,
459
+ "single_word": false,
460
+ "special": false
461
+ },
462
+ "151700": {
463
+ "content": "<ACTION_TOKEN_15>",
464
+ "lstrip": false,
465
+ "normalized": true,
466
+ "rstrip": false,
467
+ "single_word": false,
468
+ "special": false
469
+ }
470
+ },
471
+ "additional_special_tokens": [
472
+ "<|im_start|>",
473
+ "<|im_end|>",
474
+ "<|object_ref_start|>",
475
+ "<|object_ref_end|>",
476
+ "<|box_start|>",
477
+ "<|box_end|>",
478
+ "<|quad_start|>",
479
+ "<|quad_end|>",
480
+ "<|vision_start|>",
481
+ "<|vision_end|>",
482
+ "<|vision_pad|>",
483
+ "<|image_pad|>",
484
+ "<|video_pad|>",
485
+ "<img>",
486
+ "</img>",
487
+ "<IMG_CONTEXT>",
488
+ "<quad>",
489
+ "</quad>",
490
+ "<ref>",
491
+ "</ref>",
492
+ "<box>",
493
+ "</box>"
494
+ ],
495
+ "bos_token": null,
496
+ "clean_up_tokenization_spaces": false,
497
+ "eos_token": "<|im_end|>",
498
+ "errors": "replace",
499
+ "extra_special_tokens": {},
500
+ "model_max_length": 16384,
501
+ "pad_token": "<|endoftext|>",
502
+ "processor_class": "InternVLProcessor",
503
+ "split_special_tokens": false,
504
+ "tokenizer_class": "Qwen2Tokenizer",
505
+ "unk_token": null
506
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff