Upload model weights.
Browse files- .gitattributes +1 -0
- added_tokens.json +21 -0
- args.json +304 -0
- chat_template.json +3 -0
- config.json +53 -0
- generation_config.json +13 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- preprocessor_config.json +29 -0
- special_tokens_map.json +47 -0
- tokenizer.json +3 -0
- tokenizer_config.json +177 -0
- trainer_state.json +0 -0
- training_args.bin +3 -0
- 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 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|box_end|>": 151649,
|
3 |
+
"<|box_start|>": 151648,
|
4 |
+
"<|diff_marker|>": 151658,
|
5 |
+
"<|endoftext|>": 151643,
|
6 |
+
"<|im_end|>": 151645,
|
7 |
+
"<|im_start|>": 151644,
|
8 |
+
"<|image_pad|>": 151655,
|
9 |
+
"<|object_ref_end|>": 151647,
|
10 |
+
"<|object_ref_start|>": 151646,
|
11 |
+
"<|pointer_end|>": 151660,
|
12 |
+
"<|pointer_pad|>": 151661,
|
13 |
+
"<|pointer_start|>": 151659,
|
14 |
+
"<|quad_end|>": 151651,
|
15 |
+
"<|quad_start|>": 151650,
|
16 |
+
"<|recipient|>": 151657,
|
17 |
+
"<|video_pad|>": 151656,
|
18 |
+
"<|vision_end|>": 151653,
|
19 |
+
"<|vision_pad|>": 151654,
|
20 |
+
"<|vision_start|>": 151652
|
21 |
+
}
|
args.json
ADDED
@@ -0,0 +1,304 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"model_config": {
|
3 |
+
"vocab_size": 151662,
|
4 |
+
"max_position_embeddings": 32768,
|
5 |
+
"hidden_size": 1536,
|
6 |
+
"intermediate_size": 8960,
|
7 |
+
"num_hidden_layers": 28,
|
8 |
+
"num_attention_heads": 12,
|
9 |
+
"use_sliding_window": false,
|
10 |
+
"sliding_window": 32768,
|
11 |
+
"max_window_layers": 28,
|
12 |
+
"num_key_value_heads": 2,
|
13 |
+
"hidden_act": "silu",
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"rms_norm_eps": 1e-06,
|
16 |
+
"use_cache": false,
|
17 |
+
"rope_theta": 1000000.0,
|
18 |
+
"attention_dropout": 0.0,
|
19 |
+
"rope_scaling": {
|
20 |
+
"mrope_section": [
|
21 |
+
16,
|
22 |
+
24,
|
23 |
+
24
|
24 |
+
],
|
25 |
+
"rope_type": "default",
|
26 |
+
"type": "default"
|
27 |
+
},
|
28 |
+
"return_dict": true,
|
29 |
+
"output_hidden_states": false,
|
30 |
+
"output_attentions": false,
|
31 |
+
"torchscript": false,
|
32 |
+
"use_bfloat16": false,
|
33 |
+
"tf_legacy_loss": false,
|
34 |
+
"pruned_heads": {},
|
35 |
+
"tie_word_embeddings": true,
|
36 |
+
"chunk_size_feed_forward": 0,
|
37 |
+
"is_encoder_decoder": false,
|
38 |
+
"is_decoder": false,
|
39 |
+
"cross_attention_hidden_size": null,
|
40 |
+
"add_cross_attention": false,
|
41 |
+
"tie_encoder_decoder": false,
|
42 |
+
"max_length": 20,
|
43 |
+
"min_length": 0,
|
44 |
+
"do_sample": false,
|
45 |
+
"early_stopping": false,
|
46 |
+
"num_beams": 1,
|
47 |
+
"num_beam_groups": 1,
|
48 |
+
"diversity_penalty": 0.0,
|
49 |
+
"temperature": 1.0,
|
50 |
+
"top_k": 50,
|
51 |
+
"top_p": 1.0,
|
52 |
+
"typical_p": 1.0,
|
53 |
+
"repetition_penalty": 1.0,
|
54 |
+
"length_penalty": 1.0,
|
55 |
+
"no_repeat_ngram_size": 0,
|
56 |
+
"encoder_no_repeat_ngram_size": 0,
|
57 |
+
"bad_words_ids": null,
|
58 |
+
"num_return_sequences": 1,
|
59 |
+
"output_scores": false,
|
60 |
+
"return_dict_in_generate": false,
|
61 |
+
"forced_bos_token_id": null,
|
62 |
+
"forced_eos_token_id": null,
|
63 |
+
"remove_invalid_values": false,
|
64 |
+
"exponential_decay_length_penalty": null,
|
65 |
+
"suppress_tokens": null,
|
66 |
+
"begin_suppress_tokens": null,
|
67 |
+
"architectures": [
|
68 |
+
"Qwen2VLForConditionalGenerationWithPointer"
|
69 |
+
],
|
70 |
+
"finetuning_task": null,
|
71 |
+
"id2label": {
|
72 |
+
"0": "LABEL_0",
|
73 |
+
"1": "LABEL_1"
|
74 |
+
},
|
75 |
+
"label2id": {
|
76 |
+
"LABEL_0": 0,
|
77 |
+
"LABEL_1": 1
|
78 |
+
},
|
79 |
+
"tokenizer_class": null,
|
80 |
+
"prefix": null,
|
81 |
+
"bos_token_id": 151643,
|
82 |
+
"pad_token_id": null,
|
83 |
+
"eos_token_id": [
|
84 |
+
151658
|
85 |
+
],
|
86 |
+
"sep_token_id": null,
|
87 |
+
"decoder_start_token_id": null,
|
88 |
+
"task_specific_params": null,
|
89 |
+
"problem_type": null,
|
90 |
+
"_name_or_path": "/mnt/checkpoints/aguvis_final/qwen2vl7binstruct_stage1_ep1_lr0.0001_bs1_mp5720064_mml24576_ufallFalse_ufpnTrue_uflmFalse_ufbmFalse_ufntTrue_ufvFalse_ploss1.0_lmloss-1.0_stage0_all_params_uground_a100_n8_final_fixed_2B/",
|
91 |
+
"_commit_hash": null,
|
92 |
+
"_attn_implementation_internal": "flash_attention_2",
|
93 |
+
"_attn_implementation_autoset": true,
|
94 |
+
"transformers_version": "4.47.1",
|
95 |
+
"image_token_id": 151655,
|
96 |
+
"model_type": "qwen2_vl",
|
97 |
+
"pointer_end_token_id": 151660,
|
98 |
+
"pointer_pad_token_id": 151661,
|
99 |
+
"pointer_start_token_id": 151659,
|
100 |
+
"video_token_id": 151656,
|
101 |
+
"vision_end_token_id": 151653,
|
102 |
+
"vision_start_token_id": 151652,
|
103 |
+
"vision_token_id": 151654
|
104 |
+
},
|
105 |
+
"data_processor_config": {
|
106 |
+
"image_token": "<|image_pad|>",
|
107 |
+
"video_token": "<|video_pad|>",
|
108 |
+
"chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
|
109 |
+
},
|
110 |
+
"image_processor_config": {
|
111 |
+
"_processor_class": "Qwen2VLProcessor",
|
112 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
113 |
+
"size": {
|
114 |
+
"min_pixels": 3136,
|
115 |
+
"max_pixels": 5720064
|
116 |
+
},
|
117 |
+
"do_resize": true,
|
118 |
+
"resample": 3,
|
119 |
+
"do_rescale": true,
|
120 |
+
"rescale_factor": 0.00392156862745098,
|
121 |
+
"do_normalize": true,
|
122 |
+
"image_mean": [
|
123 |
+
0.48145466,
|
124 |
+
0.4578275,
|
125 |
+
0.40821073
|
126 |
+
],
|
127 |
+
"image_std": [
|
128 |
+
0.26862954,
|
129 |
+
0.26130258,
|
130 |
+
0.27577711
|
131 |
+
],
|
132 |
+
"min_pixels": 3136,
|
133 |
+
"max_pixels": 5720064,
|
134 |
+
"patch_size": 14,
|
135 |
+
"temporal_patch_size": 2,
|
136 |
+
"merge_size": 2,
|
137 |
+
"do_convert_rgb": true
|
138 |
+
},
|
139 |
+
"model_args": {
|
140 |
+
"model_name_or_path": "/mnt/checkpoints/aguvis_final/qwen2vl7binstruct_stage1_ep1_lr0.0001_bs1_mp5720064_mml24576_ufallFalse_ufpnTrue_uflmFalse_ufbmFalse_ufntTrue_ufvFalse_ploss1.0_lmloss-1.0_stage0_all_params_uground_a100_n8_final_fixed_2B/",
|
141 |
+
"flash_attn_2_enabled": true
|
142 |
+
},
|
143 |
+
"data_args": {
|
144 |
+
"data_path": "data/amlt_stage1.yaml",
|
145 |
+
"early_mix_text": false,
|
146 |
+
"image_folder": "/mnt/data/aguvis",
|
147 |
+
"min_pixels": 3136,
|
148 |
+
"max_pixels": 5720064,
|
149 |
+
"max_conv_turns": 10
|
150 |
+
},
|
151 |
+
"training_args": {
|
152 |
+
"output_dir": "/mnt/checkpoints/aguvis_final/qwen2vl7binstruct_stage1_ep1_lr5e-06_bs1_mp5720064_mml24576_ufallTrue_ufpnFalse_uflmFalse_ufbmFalse_ufntFalse_ufvFalse_ploss1.0_lmloss1.0_stage1_all_params_uground_a100_n8_final_fixed_2B",
|
153 |
+
"overwrite_output_dir": false,
|
154 |
+
"do_train": false,
|
155 |
+
"do_eval": false,
|
156 |
+
"do_predict": false,
|
157 |
+
"eval_strategy": "no",
|
158 |
+
"prediction_loss_only": false,
|
159 |
+
"per_device_train_batch_size": 1,
|
160 |
+
"per_device_eval_batch_size": 4,
|
161 |
+
"per_gpu_train_batch_size": null,
|
162 |
+
"per_gpu_eval_batch_size": null,
|
163 |
+
"gradient_accumulation_steps": 1,
|
164 |
+
"eval_accumulation_steps": null,
|
165 |
+
"eval_delay": 0,
|
166 |
+
"torch_empty_cache_steps": null,
|
167 |
+
"learning_rate": 5e-06,
|
168 |
+
"weight_decay": 0.0,
|
169 |
+
"adam_beta1": 0.9,
|
170 |
+
"adam_beta2": 0.999,
|
171 |
+
"adam_epsilon": 1e-08,
|
172 |
+
"max_grad_norm": 1.0,
|
173 |
+
"num_train_epochs": 1.0,
|
174 |
+
"max_steps": -1,
|
175 |
+
"lr_scheduler_type": "cosine",
|
176 |
+
"lr_scheduler_kwargs": {},
|
177 |
+
"warmup_ratio": 0.03,
|
178 |
+
"warmup_steps": 0,
|
179 |
+
"log_level": "passive",
|
180 |
+
"log_level_replica": "warning",
|
181 |
+
"log_on_each_node": true,
|
182 |
+
"logging_dir": "/mnt/checkpoints/aguvis_final/qwen2vl7binstruct_stage1_ep1_lr5e-06_bs1_mp5720064_mml24576_ufallTrue_ufpnFalse_uflmFalse_ufbmFalse_ufntFalse_ufvFalse_ploss1.0_lmloss1.0_stage1_all_params_uground_a100_n8_final_fixed_2B/runs/May22_17-23-22_node-7",
|
183 |
+
"logging_strategy": "steps",
|
184 |
+
"logging_first_step": false,
|
185 |
+
"logging_steps": 10,
|
186 |
+
"logging_nan_inf_filter": true,
|
187 |
+
"save_strategy": "steps",
|
188 |
+
"save_steps": 1000,
|
189 |
+
"save_total_limit": null,
|
190 |
+
"save_safetensors": true,
|
191 |
+
"save_on_each_node": false,
|
192 |
+
"save_only_model": false,
|
193 |
+
"restore_callback_states_from_checkpoint": false,
|
194 |
+
"no_cuda": false,
|
195 |
+
"use_cpu": false,
|
196 |
+
"use_mps_device": false,
|
197 |
+
"seed": 42,
|
198 |
+
"data_seed": null,
|
199 |
+
"jit_mode_eval": false,
|
200 |
+
"use_ipex": false,
|
201 |
+
"bf16": true,
|
202 |
+
"fp16": false,
|
203 |
+
"fp16_opt_level": "O1",
|
204 |
+
"half_precision_backend": "auto",
|
205 |
+
"bf16_full_eval": false,
|
206 |
+
"fp16_full_eval": false,
|
207 |
+
"tf32": true,
|
208 |
+
"local_rank": 0,
|
209 |
+
"ddp_backend": null,
|
210 |
+
"tpu_num_cores": null,
|
211 |
+
"tpu_metrics_debug": false,
|
212 |
+
"debug": [],
|
213 |
+
"dataloader_drop_last": false,
|
214 |
+
"eval_steps": null,
|
215 |
+
"dataloader_num_workers": 8,
|
216 |
+
"dataloader_prefetch_factor": null,
|
217 |
+
"past_index": -1,
|
218 |
+
"run_name": "qwen2vl7binstruct_stage1_ep1_lr5e-06_bs1_mp5720064_mml24576_ufallTrue_ufpnFalse_uflmFalse_ufbmFalse_ufntFalse_ufvFalse_ploss1.0_lmloss1.0_stage1_all_params_uground_a100_n8_final_fixed_2B",
|
219 |
+
"disable_tqdm": false,
|
220 |
+
"remove_unused_columns": true,
|
221 |
+
"label_names": null,
|
222 |
+
"load_best_model_at_end": false,
|
223 |
+
"metric_for_best_model": null,
|
224 |
+
"greater_is_better": null,
|
225 |
+
"ignore_data_skip": false,
|
226 |
+
"fsdp": [],
|
227 |
+
"fsdp_min_num_params": 0,
|
228 |
+
"fsdp_config": {
|
229 |
+
"min_num_params": 0,
|
230 |
+
"xla": false,
|
231 |
+
"xla_fsdp_v2": false,
|
232 |
+
"xla_fsdp_grad_ckpt": false
|
233 |
+
},
|
234 |
+
"fsdp_transformer_layer_cls_to_wrap": null,
|
235 |
+
"deepspeed": "./scripts/zero3.json",
|
236 |
+
"label_smoothing_factor": 0.0,
|
237 |
+
"optim": "adamw_torch",
|
238 |
+
"optim_args": null,
|
239 |
+
"adafactor": false,
|
240 |
+
"group_by_length": false,
|
241 |
+
"length_column_name": "length",
|
242 |
+
"report_to": [
|
243 |
+
"wandb"
|
244 |
+
],
|
245 |
+
"ddp_find_unused_parameters": null,
|
246 |
+
"ddp_bucket_cap_mb": null,
|
247 |
+
"ddp_broadcast_buffers": null,
|
248 |
+
"dataloader_pin_memory": true,
|
249 |
+
"dataloader_persistent_workers": false,
|
250 |
+
"skip_memory_metrics": true,
|
251 |
+
"use_legacy_prediction_loop": false,
|
252 |
+
"push_to_hub": false,
|
253 |
+
"resume_from_checkpoint": null,
|
254 |
+
"hub_model_id": null,
|
255 |
+
"hub_strategy": "every_save",
|
256 |
+
"hub_token": null,
|
257 |
+
"hub_private_repo": null,
|
258 |
+
"hub_always_push": false,
|
259 |
+
"gradient_checkpointing": true,
|
260 |
+
"gradient_checkpointing_kwargs": null,
|
261 |
+
"include_inputs_for_metrics": false,
|
262 |
+
"include_for_metrics": [],
|
263 |
+
"eval_do_concat_batches": true,
|
264 |
+
"fp16_backend": "auto",
|
265 |
+
"evaluation_strategy": null,
|
266 |
+
"push_to_hub_model_id": null,
|
267 |
+
"push_to_hub_organization": null,
|
268 |
+
"push_to_hub_token": null,
|
269 |
+
"mp_parameters": "",
|
270 |
+
"auto_find_batch_size": false,
|
271 |
+
"full_determinism": false,
|
272 |
+
"torchdynamo": null,
|
273 |
+
"ray_scope": "last",
|
274 |
+
"ddp_timeout": 1800,
|
275 |
+
"torch_compile": false,
|
276 |
+
"torch_compile_backend": null,
|
277 |
+
"torch_compile_mode": null,
|
278 |
+
"dispatch_batches": null,
|
279 |
+
"split_batches": null,
|
280 |
+
"include_tokens_per_second": false,
|
281 |
+
"include_num_input_tokens_seen": false,
|
282 |
+
"neftune_noise_alpha": null,
|
283 |
+
"optim_target_modules": null,
|
284 |
+
"batch_eval_metrics": false,
|
285 |
+
"eval_on_start": false,
|
286 |
+
"use_liger_kernel": false,
|
287 |
+
"eval_use_gather_object": false,
|
288 |
+
"average_tokens_across_devices": false,
|
289 |
+
"cache_dir": null,
|
290 |
+
"model_max_length": 24576,
|
291 |
+
"group_by_modality_length": true,
|
292 |
+
"verbose_logging": false,
|
293 |
+
"unfreeze_all_parameters": true,
|
294 |
+
"unfreeze_pointer_head": false,
|
295 |
+
"unfreeze_lm_head": false,
|
296 |
+
"unfreeze_base_model": false,
|
297 |
+
"unfreeze_last_n_layers": -1,
|
298 |
+
"unfreeze_new_tokens": false,
|
299 |
+
"unfreeze_visual": false,
|
300 |
+
"pointer_loss_weight": 1.0,
|
301 |
+
"lm_loss_weight": 1.0,
|
302 |
+
"_n_gpu": 1
|
303 |
+
}
|
304 |
+
}
|
chat_template.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
|
3 |
+
}
|
config.json
ADDED
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/mnt/checkpoints/aguvis_final/qwen2vl7binstruct_stage1_ep1_lr0.0001_bs1_mp5720064_mml24576_ufallFalse_ufpnTrue_uflmFalse_ufbmFalse_ufntTrue_ufvFalse_ploss1.0_lmloss-1.0_stage0_all_params_uground_a100_n8_final_fixed_2B/",
|
3 |
+
"architectures": [
|
4 |
+
"Qwen2VLForConditionalGenerationWithPointer"
|
5 |
+
],
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"bos_token_id": 151643,
|
8 |
+
"eos_token_id": [
|
9 |
+
151658
|
10 |
+
],
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 1536,
|
13 |
+
"image_token_id": 151655,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 8960,
|
16 |
+
"max_position_embeddings": 32768,
|
17 |
+
"max_window_layers": 28,
|
18 |
+
"model_type": "qwen2_vl",
|
19 |
+
"num_attention_heads": 12,
|
20 |
+
"num_hidden_layers": 28,
|
21 |
+
"num_key_value_heads": 2,
|
22 |
+
"pointer_end_token_id": 151660,
|
23 |
+
"pointer_pad_token_id": 151661,
|
24 |
+
"pointer_start_token_id": 151659,
|
25 |
+
"rms_norm_eps": 1e-06,
|
26 |
+
"rope_scaling": {
|
27 |
+
"mrope_section": [
|
28 |
+
16,
|
29 |
+
24,
|
30 |
+
24
|
31 |
+
],
|
32 |
+
"rope_type": "default",
|
33 |
+
"type": "default"
|
34 |
+
},
|
35 |
+
"rope_theta": 1000000.0,
|
36 |
+
"sliding_window": 32768,
|
37 |
+
"tie_word_embeddings": true,
|
38 |
+
"torch_dtype": "bfloat16",
|
39 |
+
"transformers_version": "4.47.1",
|
40 |
+
"use_cache": true,
|
41 |
+
"use_sliding_window": false,
|
42 |
+
"video_token_id": 151656,
|
43 |
+
"vision_config": {
|
44 |
+
"hidden_size": 1536,
|
45 |
+
"in_chans": 3,
|
46 |
+
"model_type": "qwen2_vl",
|
47 |
+
"spatial_patch_size": 14
|
48 |
+
},
|
49 |
+
"vision_end_token_id": 151653,
|
50 |
+
"vision_start_token_id": 151652,
|
51 |
+
"vision_token_id": 151654,
|
52 |
+
"vocab_size": 151662
|
53 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"attn_implementation": "flash_attention_2",
|
3 |
+
"bos_token_id": 151643,
|
4 |
+
"do_sample": true,
|
5 |
+
"eos_token_id": [
|
6 |
+
151658
|
7 |
+
],
|
8 |
+
"pad_token_id": 151643,
|
9 |
+
"temperature": 0.01,
|
10 |
+
"top_k": 1,
|
11 |
+
"top_p": 0.001,
|
12 |
+
"transformers_version": "4.47.1"
|
13 |
+
}
|
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 sha256:63c798ba4fe73c9a92c4d15fa58ca9739774d83bfcd7911d5d2c58310fbd9146
|
3 |
+
size 4454990312
|
preprocessor_config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_convert_rgb": true,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_rescale": true,
|
5 |
+
"do_resize": true,
|
6 |
+
"image_mean": [
|
7 |
+
0.48145466,
|
8 |
+
0.4578275,
|
9 |
+
0.40821073
|
10 |
+
],
|
11 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
12 |
+
"image_std": [
|
13 |
+
0.26862954,
|
14 |
+
0.26130258,
|
15 |
+
0.27577711
|
16 |
+
],
|
17 |
+
"max_pixels": 5720064,
|
18 |
+
"merge_size": 2,
|
19 |
+
"min_pixels": 3136,
|
20 |
+
"patch_size": 14,
|
21 |
+
"processor_class": "Qwen2VLProcessor",
|
22 |
+
"resample": 3,
|
23 |
+
"rescale_factor": 0.00392156862745098,
|
24 |
+
"size": {
|
25 |
+
"max_pixels": 5720064,
|
26 |
+
"min_pixels": 3136
|
27 |
+
},
|
28 |
+
"temporal_patch_size": 2
|
29 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
{
|
4 |
+
"content": "<|recipient|>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false
|
9 |
+
},
|
10 |
+
{
|
11 |
+
"content": "<|diff_marker|>",
|
12 |
+
"lstrip": false,
|
13 |
+
"normalized": false,
|
14 |
+
"rstrip": false,
|
15 |
+
"single_word": false
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"content": "<|pointer_start|>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": false,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"content": "<|pointer_end|>",
|
26 |
+
"lstrip": false,
|
27 |
+
"normalized": false,
|
28 |
+
"rstrip": false,
|
29 |
+
"single_word": false
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"content": "<|pointer_pad|>",
|
33 |
+
"lstrip": false,
|
34 |
+
"normalized": false,
|
35 |
+
"rstrip": false,
|
36 |
+
"single_word": false
|
37 |
+
}
|
38 |
+
],
|
39 |
+
"eos_token": "<|diff_marker|>",
|
40 |
+
"pad_token": {
|
41 |
+
"content": "<|endoftext|>",
|
42 |
+
"lstrip": false,
|
43 |
+
"normalized": false,
|
44 |
+
"rstrip": false,
|
45 |
+
"single_word": false
|
46 |
+
}
|
47 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9ca6580f6054f3fd7a7bd12c9a6319757c8e6f1106844b6fc595aebca7e9c9dd
|
3 |
+
size 11421331
|
tokenizer_config.json
ADDED
@@ -0,0 +1,177 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"151643": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": false,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
},
|
12 |
+
"151644": {
|
13 |
+
"content": "<|im_start|>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": true
|
19 |
+
},
|
20 |
+
"151645": {
|
21 |
+
"content": "<|im_end|>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false,
|
26 |
+
"special": true
|
27 |
+
},
|
28 |
+
"151646": {
|
29 |
+
"content": "<|object_ref_start|>",
|
30 |
+
"lstrip": false,
|
31 |
+
"normalized": false,
|
32 |
+
"rstrip": false,
|
33 |
+
"single_word": false,
|
34 |
+
"special": true
|
35 |
+
},
|
36 |
+
"151647": {
|
37 |
+
"content": "<|object_ref_end|>",
|
38 |
+
"lstrip": false,
|
39 |
+
"normalized": false,
|
40 |
+
"rstrip": false,
|
41 |
+
"single_word": false,
|
42 |
+
"special": true
|
43 |
+
},
|
44 |
+
"151648": {
|
45 |
+
"content": "<|box_start|>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": false,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false,
|
50 |
+
"special": true
|
51 |
+
},
|
52 |
+
"151649": {
|
53 |
+
"content": "<|box_end|>",
|
54 |
+
"lstrip": false,
|
55 |
+
"normalized": false,
|
56 |
+
"rstrip": false,
|
57 |
+
"single_word": false,
|
58 |
+
"special": true
|
59 |
+
},
|
60 |
+
"151650": {
|
61 |
+
"content": "<|quad_start|>",
|
62 |
+
"lstrip": false,
|
63 |
+
"normalized": false,
|
64 |
+
"rstrip": false,
|
65 |
+
"single_word": false,
|
66 |
+
"special": true
|
67 |
+
},
|
68 |
+
"151651": {
|
69 |
+
"content": "<|quad_end|>",
|
70 |
+
"lstrip": false,
|
71 |
+
"normalized": false,
|
72 |
+
"rstrip": false,
|
73 |
+
"single_word": false,
|
74 |
+
"special": true
|
75 |
+
},
|
76 |
+
"151652": {
|
77 |
+
"content": "<|vision_start|>",
|
78 |
+
"lstrip": false,
|
79 |
+
"normalized": false,
|
80 |
+
"rstrip": false,
|
81 |
+
"single_word": false,
|
82 |
+
"special": true
|
83 |
+
},
|
84 |
+
"151653": {
|
85 |
+
"content": "<|vision_end|>",
|
86 |
+
"lstrip": false,
|
87 |
+
"normalized": false,
|
88 |
+
"rstrip": false,
|
89 |
+
"single_word": false,
|
90 |
+
"special": true
|
91 |
+
},
|
92 |
+
"151654": {
|
93 |
+
"content": "<|vision_pad|>",
|
94 |
+
"lstrip": false,
|
95 |
+
"normalized": false,
|
96 |
+
"rstrip": false,
|
97 |
+
"single_word": false,
|
98 |
+
"special": true
|
99 |
+
},
|
100 |
+
"151655": {
|
101 |
+
"content": "<|image_pad|>",
|
102 |
+
"lstrip": false,
|
103 |
+
"normalized": false,
|
104 |
+
"rstrip": false,
|
105 |
+
"single_word": false,
|
106 |
+
"special": true
|
107 |
+
},
|
108 |
+
"151656": {
|
109 |
+
"content": "<|video_pad|>",
|
110 |
+
"lstrip": false,
|
111 |
+
"normalized": false,
|
112 |
+
"rstrip": false,
|
113 |
+
"single_word": false,
|
114 |
+
"special": true
|
115 |
+
},
|
116 |
+
"151657": {
|
117 |
+
"content": "<|recipient|>",
|
118 |
+
"lstrip": false,
|
119 |
+
"normalized": false,
|
120 |
+
"rstrip": false,
|
121 |
+
"single_word": false,
|
122 |
+
"special": true
|
123 |
+
},
|
124 |
+
"151658": {
|
125 |
+
"content": "<|diff_marker|>",
|
126 |
+
"lstrip": false,
|
127 |
+
"normalized": false,
|
128 |
+
"rstrip": false,
|
129 |
+
"single_word": false,
|
130 |
+
"special": true
|
131 |
+
},
|
132 |
+
"151659": {
|
133 |
+
"content": "<|pointer_start|>",
|
134 |
+
"lstrip": false,
|
135 |
+
"normalized": false,
|
136 |
+
"rstrip": false,
|
137 |
+
"single_word": false,
|
138 |
+
"special": true
|
139 |
+
},
|
140 |
+
"151660": {
|
141 |
+
"content": "<|pointer_end|>",
|
142 |
+
"lstrip": false,
|
143 |
+
"normalized": false,
|
144 |
+
"rstrip": false,
|
145 |
+
"single_word": false,
|
146 |
+
"special": true
|
147 |
+
},
|
148 |
+
"151661": {
|
149 |
+
"content": "<|pointer_pad|>",
|
150 |
+
"lstrip": false,
|
151 |
+
"normalized": false,
|
152 |
+
"rstrip": false,
|
153 |
+
"single_word": false,
|
154 |
+
"special": true
|
155 |
+
}
|
156 |
+
},
|
157 |
+
"additional_special_tokens": [
|
158 |
+
"<|recipient|>",
|
159 |
+
"<|diff_marker|>",
|
160 |
+
"<|pointer_start|>",
|
161 |
+
"<|pointer_end|>",
|
162 |
+
"<|pointer_pad|>"
|
163 |
+
],
|
164 |
+
"bos_token": null,
|
165 |
+
"chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
|
166 |
+
"clean_up_tokenization_spaces": false,
|
167 |
+
"eos_token": "<|diff_marker|>",
|
168 |
+
"errors": "replace",
|
169 |
+
"extra_special_tokens": {},
|
170 |
+
"model_max_length": 24576,
|
171 |
+
"pad_token": "<|endoftext|>",
|
172 |
+
"padding_side": "right",
|
173 |
+
"processor_class": "Qwen2VLProcessor",
|
174 |
+
"split_special_tokens": false,
|
175 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
176 |
+
"unk_token": null
|
177 |
+
}
|
trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:919574f979bb70edaa6d1b576ac9b0febad4c854fa3305167e1847607f1b3994
|
3 |
+
size 8312
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|