Hugging Face's logo

FreedomIntelligence
/
TierMem

Safetensors
Model card Files
xet
Community
Shunian commited on
Commit
13ea8c1
·
verified ·
1 Parent(s): b9e67a0

Upload qwen3_0.6b_lora model

Browse files
Files changed (14) hide show
  1. .gitattributes +1 -0
  2. model/qwen3_0.6b_lora/.claude/settings.local.json +9 -0
  3. model/qwen3_0.6b_lora/added_tokens.json +28 -0
  4. model/qwen3_0.6b_lora/args.json +527 -0
  5. model/qwen3_0.6b_lora/args.json.bak +527 -0
  6. model/qwen3_0.6b_lora/chat_template.jinja +89 -0
  7. model/qwen3_0.6b_lora/config.json +61 -0
  8. model/qwen3_0.6b_lora/generation_config.json +13 -0
  9. model/qwen3_0.6b_lora/merges.txt +0 -0
  10. model/qwen3_0.6b_lora/model.safetensors +3 -0
  11. model/qwen3_0.6b_lora/special_tokens_map.json +31 -0
  12. model/qwen3_0.6b_lora/tokenizer.json +3 -0
  13. model/qwen3_0.6b_lora/tokenizer_config.json +239 -0
  14. model/qwen3_0.6b_lora/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
+ model/qwen3_0.6b_lora/tokenizer.json filter=lfs diff=lfs merge=lfs -text
model/qwen3_0.6b_lora/.claude/settings.local.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "permissions": {
3
+ "allow": [
4
+ "Bash(pip show:*)",
5
+ "Bash(python3:*)",
6
+ "Bash(pip install:*)"
7
+ ]
8
+ }
9
+ }
model/qwen3_0.6b_lora/added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
model/qwen3_0.6b_lora/args.json ADDED
@@ -0,0 +1,527 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "output_dir": "[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952",
3
+ "overwrite_output_dir": false,
4
+ "do_train": false,
5
+ "do_eval": false,
6
+ "do_predict": false,
7
+ "eval_strategy": "steps",
8
+ "prediction_loss_only": false,
9
+ "per_device_train_batch_size": 24,
10
+ "per_device_eval_batch_size": 24,
11
+ "per_gpu_train_batch_size": null,
12
+ "per_gpu_eval_batch_size": null,
13
+ "gradient_accumulation_steps": 1,
14
+ "eval_accumulation_steps": null,
15
+ "eval_delay": 0,
16
+ "torch_empty_cache_steps": null,
17
+ "learning_rate": 5e-06,
18
+ "weight_decay": 0.1,
19
+ "adam_beta1": 0.9,
20
+ "adam_beta2": 0.95,
21
+ "adam_epsilon": 1e-08,
22
+ "max_grad_norm": 1.0,
23
+ "num_train_epochs": 5.0,
24
+ "max_steps": -1,
25
+ "lr_scheduler_type": "cosine",
26
+ "lr_scheduler_kwargs": null,
27
+ "warmup_ratio": 0.05,
28
+ "warmup_steps": 0,
29
+ "log_level": "passive",
30
+ "log_level_replica": "warning",
31
+ "log_on_each_node": true,
32
+ "logging_dir": "[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952/runs",
33
+ "logging_strategy": "steps",
34
+ "logging_first_step": true,
35
+ "logging_steps": 1,
36
+ "logging_nan_inf_filter": true,
37
+ "save_strategy": "steps",
38
+ "save_steps": 50.0,
39
+ "save_total_limit": null,
40
+ "save_safetensors": true,
41
+ "save_on_each_node": false,
42
+ "save_only_model": false,
43
+ "restore_callback_states_from_checkpoint": false,
44
+ "no_cuda": false,
45
+ "use_cpu": false,
46
+ "use_mps_device": false,
47
+ "seed": 42,
48
+ "data_seed": 42,
49
+ "jit_mode_eval": false,
50
+ "bf16": true,
51
+ "fp16": false,
52
+ "fp16_opt_level": "O1",
53
+ "half_precision_backend": "auto",
54
+ "bf16_full_eval": false,
55
+ "fp16_full_eval": false,
56
+ "tf32": null,
57
+ "local_rank": 0,
58
+ "ddp_backend": null,
59
+ "tpu_num_cores": null,
60
+ "tpu_metrics_debug": false,
61
+ "debug": null,
62
+ "dataloader_drop_last": false,
63
+ "eval_steps": 50.0,
64
+ "dataloader_num_workers": 4,
65
+ "dataloader_prefetch_factor": null,
66
+ "past_index": -1,
67
+ "run_name": "[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952",
68
+ "disable_tqdm": null,
69
+ "remove_unused_columns": false,
70
+ "label_names": null,
71
+ "load_best_model_at_end": false,
72
+ "metric_for_best_model": "loss",
73
+ "greater_is_better": false,
74
+ "ignore_data_skip": false,
75
+ "fsdp": [],
76
+ "fsdp_min_num_params": 0,
77
+ "fsdp_config": null,
78
+ "fsdp_transformer_layer_cls_to_wrap": null,
79
+ "accelerator_config": {
80
+ "dispatch_batches": false
81
+ },
82
+ "parallelism_config": null,
83
+ "deepspeed": {
84
+ "fp16": {
85
+ "enabled": "auto",
86
+ "loss_scale": 0,
87
+ "loss_scale_window": 1000,
88
+ "initial_scale_power": 16,
89
+ "hysteresis": 2,
90
+ "min_loss_scale": 1
91
+ },
92
+ "bf16": {
93
+ "enabled": "auto"
94
+ },
95
+ "zero_optimization": {
96
+ "stage": 2,
97
+ "offload_optimizer": {
98
+ "device": "none",
99
+ "pin_memory": true
100
+ },
101
+ "allgather_partitions": true,
102
+ "allgather_bucket_size": 200000000.0,
103
+ "overlap_comm": false,
104
+ "reduce_scatter": true,
105
+ "reduce_bucket_size": 200000000.0,
106
+ "contiguous_gradients": true
107
+ },
108
+ "gradient_accumulation_steps": "auto",
109
+ "gradient_clipping": "auto",
110
+ "steps_per_print": 2000,
111
+ "train_batch_size": "auto",
112
+ "train_micro_batch_size_per_gpu": "auto",
113
+ "wall_clock_breakdown": false
114
+ },
115
+ "label_smoothing_factor": 0.0,
116
+ "optim": "adamw_torch_fused",
117
+ "optim_args": null,
118
+ "adafactor": false,
119
+ "group_by_length": false,
120
+ "length_column_name": "length",
121
+ "report_to": [
122
+ "tensorboard"
123
+ ],
124
+ "project": "huggingface",
125
+ "trackio_space_id": "trackio",
126
+ "ddp_find_unused_parameters": null,
127
+ "ddp_bucket_cap_mb": null,
128
+ "ddp_broadcast_buffers": null,
129
+ "dataloader_pin_memory": true,
130
+ "dataloader_persistent_workers": false,
131
+ "skip_memory_metrics": true,
132
+ "use_legacy_prediction_loop": false,
133
+ "push_to_hub": false,
134
+ "resume_from_checkpoint": null,
135
+ "hub_model_id": null,
136
+ "hub_strategy": "every_save",
137
+ "hub_token": null,
138
+ "hub_private_repo": null,
139
+ "hub_always_push": false,
140
+ "hub_revision": null,
141
+ "gradient_checkpointing": true,
142
+ "gradient_checkpointing_kwargs": null,
143
+ "include_inputs_for_metrics": false,
144
+ "include_for_metrics": [],
145
+ "eval_do_concat_batches": true,
146
+ "fp16_backend": "auto",
147
+ "push_to_hub_model_id": null,
148
+ "push_to_hub_organization": null,
149
+ "push_to_hub_token": null,
150
+ "mp_parameters": "",
151
+ "auto_find_batch_size": false,
152
+ "full_determinism": false,
153
+ "torchdynamo": null,
154
+ "ray_scope": "last",
155
+ "ddp_timeout": 18000000,
156
+ "torch_compile": false,
157
+ "torch_compile_backend": null,
158
+ "torch_compile_mode": null,
159
+ "include_tokens_per_second": false,
160
+ "include_num_input_tokens_seen": false,
161
+ "neftune_noise_alpha": null,
162
+ "optim_target_modules": null,
163
+ "batch_eval_metrics": false,
164
+ "eval_on_start": false,
165
+ "use_liger_kernel": false,
166
+ "liger_kernel_config": null,
167
+ "eval_use_gather_object": false,
168
+ "average_tokens_across_devices": true,
169
+ "sortish_sampler": false,
170
+ "predict_with_generate": false,
171
+ "generation_max_length": null,
172
+ "generation_num_beams": null,
173
+ "generation_config": null,
174
+ "tuner_backend": "peft",
175
+ "vit_gradient_checkpointing": null,
176
+ "router_aux_loss_coef": 0.0,
177
+ "enable_dft_loss": false,
178
+ "enable_channel_loss": false,
179
+ "check_model": true,
180
+ "acc_strategy": "token",
181
+ "train_dataloader_shuffle": true,
182
+ "max_epochs": null,
183
+ "aligner_lr": null,
184
+ "vit_lr": null,
185
+ "use_logits_to_keep": null,
186
+ "ds3_gather_for_generation": true,
187
+ "resume_only_model": false,
188
+ "optimizer": null,
189
+ "loss_type": "grpo",
190
+ "metric": null,
191
+ "eval_use_evalscope": false,
192
+ "eval_dataset": [],
193
+ "eval_dataset_args": null,
194
+ "eval_limit": null,
195
+ "eval_generation_config": null,
196
+ "extra_eval_args": null,
197
+ "use_flash_ckpt": false,
198
+ "use_ray": false,
199
+ "ray_exp_name": null,
200
+ "device_groups": null,
201
+ "model": "[USER_HOME]/.cache/modelscope/hub/models/Qwen/Qwen3-0.6B",
202
+ "model_type": "qwen3",
203
+ "model_revision": null,
204
+ "task_type": "causal_lm",
205
+ "torch_dtype": "bfloat16",
206
+ "attn_impl": null,
207
+ "new_special_tokens": [],
208
+ "num_labels": null,
209
+ "problem_type": null,
210
+ "rope_scaling": null,
211
+ "device_map": null,
212
+ "max_memory": {},
213
+ "max_model_len": null,
214
+ "local_repo_path": null,
215
+ "init_strategy": null,
216
+ "template": "qwen3",
217
+ "system": null,
218
+ "max_length": 4096,
219
+ "truncation_strategy": "left",
220
+ "max_pixels": null,
221
+ "agent_template": null,
222
+ "norm_bbox": null,
223
+ "use_chat_template": true,
224
+ "padding_side": "right",
225
+ "padding_free": false,
226
+ "loss_scale": "last_round",
227
+ "sequence_parallel_size": 1,
228
+ "template_backend": "swift",
229
+ "response_prefix": null,
230
+ "enable_thinking": null,
231
+ "add_non_thinking_prefix": true,
232
+ "dataset": [
233
+ "[USER_HOME]/AGMS/data/router_grpo_v7/train_grpo.jsonl"
234
+ ],
235
+ "val_dataset": [
236
+ "[USER_HOME]/AGMS/data/router_grpo_v7/val_grpo.jsonl"
237
+ ],
238
+ "cached_dataset": [],
239
+ "cached_val_dataset": [],
240
+ "split_dataset_ratio": 0.0,
241
+ "dataset_num_proc": 1,
242
+ "load_from_cache_file": true,
243
+ "dataset_shuffle": true,
244
+ "val_dataset_shuffle": false,
245
+ "streaming": false,
246
+ "interleave_prob": null,
247
+ "stopping_strategy": "first_exhausted",
248
+ "shuffle_buffer_size": 1000,
249
+ "download_mode": "reuse_dataset_if_exists",
250
+ "columns": {},
251
+ "strict": false,
252
+ "model_name": null,
253
+ "model_author": null,
254
+ "custom_dataset_info": [],
255
+ "quant_method": null,
256
+ "quant_bits": null,
257
+ "hqq_axis": null,
258
+ "bnb_4bit_compute_dtype": "bfloat16",
259
+ "bnb_4bit_quant_type": "nf4",
260
+ "bnb_4bit_use_double_quant": true,
261
+ "bnb_4bit_quant_storage": null,
262
+ "max_new_tokens": 512,
263
+ "temperature": 0.9,
264
+ "top_k": 50,
265
+ "top_p": 0.9,
266
+ "repetition_penalty": 1.0,
267
+ "num_beams": 1,
268
+ "stream": false,
269
+ "stop_words": [],
270
+ "logprobs": false,
271
+ "top_logprobs": null,
272
+ "structured_outputs_regex": null,
273
+ "ckpt_dir": "[USER_HOME]/AGMS/output/router_sft_qwen3_0.6B/v1-20260112-153416/checkpoint-50",
274
+ "lora_modules": [],
275
+ "train_type": "lora",
276
+ "adapters": [
277
+ "[USER_HOME]/AGMS/output/router_sft_qwen3_0.6B/v1-20260112-153416/checkpoint-50"
278
+ ],
279
+ "external_plugins": [
280
+ "[USER_HOME]/AGMS/scripts/router_training/plugin/router_reward.py"
281
+ ],
282
+ "model_kwargs": {},
283
+ "load_args": false,
284
+ "load_data_args": false,
285
+ "packing": false,
286
+ "packing_length": null,
287
+ "packing_num_proc": 1,
288
+ "lazy_tokenize": false,
289
+ "custom_register_path": [],
290
+ "use_hf": false,
291
+ "ignore_args_error": false,
292
+ "use_swift_lora": false,
293
+ "freeze_parameters": [],
294
+ "freeze_parameters_regex": null,
295
+ "freeze_parameters_ratio": 0.0,
296
+ "trainable_parameters": [],
297
+ "trainable_parameters_regex": null,
298
+ "freeze_llm": false,
299
+ "freeze_vit": true,
300
+ "freeze_aligner": true,
301
+ "target_modules": [
302
+ "all-linear"
303
+ ],
304
+ "target_regex": null,
305
+ "target_parameters": null,
306
+ "modules_to_save": [],
307
+ "lora_rank": 16,
308
+ "lora_alpha": 32,
309
+ "lora_dropout": 0.05,
310
+ "lora_bias": "none",
311
+ "lora_dtype": null,
312
+ "lorap_lr_ratio": null,
313
+ "use_rslora": false,
314
+ "use_dora": false,
315
+ "lora_ga_batch_size": 2,
316
+ "lora_ga_iters": 2,
317
+ "lora_ga_max_length": 1024,
318
+ "lora_ga_direction": "ArB2r",
319
+ "lora_ga_scale": "stable",
320
+ "lora_ga_stable_gamma": 16,
321
+ "init_weights": true,
322
+ "fourier_n_frequency": 2000,
323
+ "fourier_scaling": 300.0,
324
+ "boft_block_size": 4,
325
+ "boft_block_num": 0,
326
+ "boft_n_butterfly_factor": 1,
327
+ "boft_dropout": 0.0,
328
+ "vera_rank": 256,
329
+ "vera_projection_prng_key": 0,
330
+ "vera_dropout": 0.0,
331
+ "vera_d_initial": 0.1,
332
+ "adapter_act": "gelu",
333
+ "adapter_length": 128,
334
+ "use_galore": false,
335
+ "galore_target_modules": null,
336
+ "galore_rank": 128,
337
+ "galore_update_proj_gap": 50,
338
+ "galore_scale": 1.0,
339
+ "galore_proj_type": "std",
340
+ "galore_optim_per_parameter": false,
341
+ "galore_with_embedding": false,
342
+ "galore_quantization": false,
343
+ "galore_proj_quant": false,
344
+ "galore_proj_bits": 4,
345
+ "galore_proj_group_size": 256,
346
+ "galore_cos_threshold": 0.4,
347
+ "galore_gamma_proj": 2,
348
+ "galore_queue_size": 5,
349
+ "adalora_target_r": 8,
350
+ "adalora_init_r": 12,
351
+ "adalora_tinit": 0,
352
+ "adalora_tfinal": 0,
353
+ "adalora_deltaT": 1,
354
+ "adalora_beta1": 0.85,
355
+ "adalora_beta2": 0.85,
356
+ "adalora_orth_reg_weight": 0.5,
357
+ "llamapro_num_new_blocks": 4,
358
+ "llamapro_num_groups": null,
359
+ "lisa_activated_layers": 0,
360
+ "lisa_step_interval": 20,
361
+ "reft_layer_key": null,
362
+ "reft_layers": null,
363
+ "reft_rank": 4,
364
+ "reft_intervention_type": "LoreftIntervention",
365
+ "reft_args": null,
366
+ "swanlab_token": null,
367
+ "swanlab_project": "ms-swift",
368
+ "swanlab_workspace": null,
369
+ "swanlab_exp_name": null,
370
+ "swanlab_notification_method": null,
371
+ "swanlab_webhook_url": null,
372
+ "swanlab_secret": null,
373
+ "swanlab_mode": "cloud",
374
+ "add_version": true,
375
+ "create_checkpoint_symlink": false,
376
+ "zero_hpz_partition_size": null,
377
+ "deepspeed_autotp_size": null,
378
+ "early_stop_interval": null,
379
+ "sft_alpha": 0,
380
+ "chord_sft_dataset": [],
381
+ "chord_sft_per_device_train_batch_size": null,
382
+ "chord_enable_phi_function": false,
383
+ "chord_mu_warmup_steps": null,
384
+ "chord_mu_decay_steps": null,
385
+ "chord_mu_peak": null,
386
+ "chord_mu_valley": null,
387
+ "reward_model": null,
388
+ "reward_adapters": [],
389
+ "reward_model_type": null,
390
+ "reward_model_revision": null,
391
+ "num_ppo_epochs": 4,
392
+ "whiten_rewards": false,
393
+ "kl_coef": 0.05,
394
+ "cliprange": 0.2,
395
+ "vf_coef": 0.1,
396
+ "cliprange_value": 0.2,
397
+ "gamma": 1.0,
398
+ "lam": 0.95,
399
+ "num_mini_batches": 1,
400
+ "local_rollout_forward_batch_size": 8,
401
+ "num_sample_generations": 10,
402
+ "response_length": 512,
403
+ "missing_eos_penalty": null,
404
+ "vllm_gpu_memory_utilization": 0.9,
405
+ "vllm_tensor_parallel_size": 1,
406
+ "vllm_pipeline_parallel_size": 1,
407
+ "vllm_enable_expert_parallel": false,
408
+ "vllm_max_num_seqs": null,
409
+ "vllm_max_model_len": null,
410
+ "vllm_disable_custom_all_reduce": true,
411
+ "vllm_enforce_eager": false,
412
+ "vllm_limit_mm_per_prompt": null,
413
+ "vllm_max_lora_rank": 16,
414
+ "vllm_enable_prefix_caching": true,
415
+ "vllm_use_async_engine": false,
416
+ "vllm_quantization": null,
417
+ "vllm_reasoning_parser": null,
418
+ "vllm_disable_cascade_attn": false,
419
+ "vllm_mm_processor_cache_gb": null,
420
+ "vllm_speculative_config": null,
421
+ "vllm_engine_kwargs": {},
422
+ "vllm_data_parallel_size": 1,
423
+ "use_vllm": false,
424
+ "vllm_mode": null,
425
+ "vllm_enable_lora": false,
426
+ "vllm_server_base_url": null,
427
+ "vllm_server_host": null,
428
+ "vllm_server_port": [
429
+ 8000
430
+ ],
431
+ "vllm_server_timeout": 240.0,
432
+ "vllm_server_group_port": null,
433
+ "enable_flattened_weight_sync": true,
434
+ "async_generate": false,
435
+ "sleep_level": 0,
436
+ "move_model_batches": null,
437
+ "offload_optimizer": false,
438
+ "offload_model": false,
439
+ "wandb_log_unique_prompts": null,
440
+ "epsilon": 0.2,
441
+ "epsilon_high": null,
442
+ "delta": null,
443
+ "cosine_min_len_value_wrong": -0.5,
444
+ "cosine_max_len_value_wrong": 0.0,
445
+ "cosine_min_len_value_correct": 1.0,
446
+ "cosine_max_len_value_correct": 0.5,
447
+ "cosine_max_len": null,
448
+ "repetition_n_grams": 3,
449
+ "repetition_max_penalty": -1.0,
450
+ "reward_model_plugin": null,
451
+ "sync_ref_model": false,
452
+ "ref_model_sync_steps": 512,
453
+ "ref_model_mixup_alpha": 0.6,
454
+ "multi_turn_scheduler": null,
455
+ "max_turns": null,
456
+ "completion_length_limit_scope": "per_round",
457
+ "vllm_server_pass_dataset": false,
458
+ "dynamic_sample": false,
459
+ "max_resample_times": 3,
460
+ "overlong_filter": false,
461
+ "soft_max_length": null,
462
+ "soft_cache_length": null,
463
+ "scale_rewards": "group",
464
+ "log_entropy": false,
465
+ "top_entropy_quantile": 1.0,
466
+ "importance_sampling_level": "token",
467
+ "tau_pos": 1.0,
468
+ "tau_neg": 1.05,
469
+ "advantage_estimator": "grpo",
470
+ "kl_in_reward": false,
471
+ "generation_batch_size": null,
472
+ "steps_per_generation": null,
473
+ "num_generations_eval": null,
474
+ "rollout_importance_sampling_mode": null,
475
+ "rollout_importance_sampling_threshold": 2.0,
476
+ "log_rollout_offpolicy_metrics": false,
477
+ "off_policy_sequence_mask_delta": null,
478
+ "num_generations": 12,
479
+ "reward_funcs": [
480
+ "router_final",
481
+ "router_format"
482
+ ],
483
+ "reward_weights": [
484
+ 0.95,
485
+ 0.05
486
+ ],
487
+ "log_completions": true,
488
+ "num_iterations": 1,
489
+ "teacher_model": null,
490
+ "teacher_adapters": [],
491
+ "teacher_model_type": null,
492
+ "teacher_model_revision": null,
493
+ "teacher_deepspeed": null,
494
+ "rlhf_type": "grpo",
495
+ "ref_model": null,
496
+ "ref_adapters": [],
497
+ "ref_model_type": null,
498
+ "ref_model_revision": null,
499
+ "beta": 0.01,
500
+ "label_smoothing": 0,
501
+ "max_completion_length": 512,
502
+ "rpo_alpha": null,
503
+ "ld_alpha": null,
504
+ "discopop_tau": 0.05,
505
+ "loss_weights": null,
506
+ "cpo_alpha": 1.0,
507
+ "simpo_gamma": 1,
508
+ "desirable_weight": 1.0,
509
+ "undesirable_weight": 1.0,
510
+ "center_rewards_coefficient": null,
511
+ "lmbda": 0.5,
512
+ "seq_kd": false,
513
+ "offload_teacher_model": false,
514
+ "rank": 0,
515
+ "global_world_size": 4,
516
+ "local_world_size": 4,
517
+ "model_suffix": "Qwen3-0.6B",
518
+ "model_info": "ModelInfo(model_type='qwen3', model_dir='[USER_HOME]/.cache/modelscope/hub/models/Qwen/Qwen3-0.6B', torch_dtype=torch.bfloat16, max_model_len=40960, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=False, config=None, task_type='causal_lm', num_labels=None)",
519
+ "model_meta": "ModelMeta(model_type='qwen3', model_groups=[ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3-0.6B-Base', hf_model_id='Qwen/Qwen3-0.6B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B-Base', hf_model_id='Qwen/Qwen3-1.7B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-Base', hf_model_id='Qwen/Qwen3-4B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-Base', hf_model_id='Qwen/Qwen3-8B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-Base', hf_model_id='Qwen/Qwen3-14B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-0.6B', hf_model_id='Qwen/Qwen3-0.6B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B', hf_model_id='Qwen/Qwen3-1.7B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B', hf_model_id='Qwen/Qwen3-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B', hf_model_id='Qwen/Qwen3-8B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B', hf_model_id='Qwen/Qwen3-14B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B', hf_model_id='Qwen/Qwen3-32B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-0.6B-FP8', hf_model_id='Qwen/Qwen3-0.6B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B-FP8', hf_model_id='Qwen/Qwen3-1.7B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-FP8', hf_model_id='Qwen/Qwen3-4B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-FP8', hf_model_id='Qwen/Qwen3-8B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-FP8', hf_model_id='Qwen/Qwen3-14B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B-FP8', hf_model_id='Qwen/Qwen3-32B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-AWQ', hf_model_id='Qwen/Qwen3-4B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-AWQ', hf_model_id='Qwen/Qwen3-8B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-AWQ', hf_model_id='Qwen/Qwen3-14B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B-AWQ', hf_model_id='Qwen/Qwen3-32B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='swift/Qwen3-32B-AWQ', hf_model_id=None, model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='qwen3', get_function=<function get_model_tokenizer_with_flash_attn at 0x7f9ba23d84a0>, model_arch=ModelKeys(arch_name='llama', embedding='model.embed_tokens', module_list='model.layers', lm_head='lm_head', q_proj='model.layers.{}.self_attn.q_proj', k_proj='model.layers.{}.self_attn.k_proj', v_proj='model.layers.{}.self_attn.v_proj', o_proj='model.layers.{}.self_attn.o_proj', attention='model.layers.{}.self_attn', mlp='model.layers.{}.mlp', down_proj='model.layers.{}.mlp.down_proj', qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None), architectures=['Qwen3ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=False, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers>=4.51'], tags=[])",
520
+ "model_dir": "[USER_HOME]/.cache/modelscope/hub/models/Qwen/Qwen3-0.6B",
521
+ "_val_dataset_exists": [
522
+ "[USER_HOME]/AGMS/data/router_grpo_v7/val_grpo.jsonl"
523
+ ],
524
+ "hub": "<class 'swift.hub.hub.MSHub'>",
525
+ "evaluation_strategy": "steps",
526
+ "training_args": "GRPOConfig(output_dir='[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=24, per_device_eval_batch_size=24, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=1, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=5.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.05, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=1, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=50, save_total_limit=None, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=True, eval_steps=50, dataloader_num_workers=4, dataloader_prefetch_factor=2, past_index=-1, run_name='[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), parallelism_config=None, deepspeed={'fp16': {'enabled': 'auto', 'loss_scale': 0, 'loss_scale_window': 1000, 'initial_scale_power': 16, 'hysteresis': 2, 'min_loss_scale': 1}, 'bf16': {'enabled': 'auto'}, 'zero_optimization': {'stage': 2, 'offload_optimizer': {'device': 'none', 'pin_memory': True}, 'allgather_partitions': True, 'allgather_bucket_size': 200000000.0, 'overlap_comm': False, 'reduce_scatter': True, 'reduce_bucket_size': 200000000.0, 'contiguous_gradients': True}, 'gradient_accumulation_steps': 'auto', 'gradient_clipping': 'auto', 'steps_per_print': 2000, 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'wall_clock_breakdown': False}, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH_FUSED: 'adamw_torch_fused'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], project='huggingface', trackio_space_id='trackio', ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, hub_revision=None, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, liger_kernel_config=None, eval_use_gather_object=False, average_tokens_across_devices=None, model_init_kwargs=None, disable_dropout=False, max_prompt_length=512, num_generations=12, max_completion_length=512, ds3_gather_for_generation=True, shuffle_dataset=True, generation_batch_size=96, steps_per_generation=1, temperature=0.9, top_p=0.9, top_k=50, min_p=None, generation_kwargs=None, repetition_penalty=1.0, use_transformers_paged=False, cache_implementation=None, use_vllm=False, vllm_mode=None, vllm_model_impl='vllm', vllm_enable_sleep_mode=False, vllm_guided_decoding_regex=None, vllm_server_base_url=None, vllm_server_host=None, vllm_server_port=[8000], vllm_server_timeout=240.0, vllm_gpu_memory_utilization=0.9, vllm_tensor_parallel_size=1, beta=0.01, num_iterations=1, epsilon=0.2, delta=None, epsilon_high=None, importance_sampling_level='token', reward_weights=[0.95, 0.05], scale_rewards='group', loss_type='grpo', mask_truncated_completions=False, sync_ref_model=False, ref_model_mixup_alpha=0.6, ref_model_sync_steps=512, top_entropy_quantile=1.0, use_liger_loss=False, vllm_importance_sampling_correction=True, vllm_importance_sampling_cap=2.0, log_completions=True, num_completions_to_print=None, wandb_log_unique_prompts=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=None, vit_lr=None, use_logits_to_keep=None, resume_only_model=False, optimizer=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, task_type='causal_lm', problem_type=None, vllm_pipeline_parallel_size=1, vllm_enable_expert_parallel=False, vllm_max_num_seqs=None, vllm_max_model_len=None, vllm_disable_custom_all_reduce=True, vllm_enforce_eager=False, vllm_limit_mm_per_prompt=None, vllm_max_lora_rank=16, vllm_enable_prefix_caching=True, vllm_use_async_engine=False, vllm_quantization=None, vllm_reasoning_parser=None, vllm_disable_cascade_attn=False, vllm_mm_processor_cache_gb=None, vllm_speculative_config=None, vllm_engine_kwargs={}, vllm_data_parallel_size=1, stop_words=[], vllm_enable_lora=False, lora_rank=16, vllm_server_group_port=None, enable_flattened_weight_sync=True, async_generate=False, structured_outputs_regex=None, sleep_level=0, move_model_batches=None, offload_optimizer=False, offload_model=False, cosine_min_len_value_wrong=-0.5, cosine_max_len_value_wrong=0.0, cosine_min_len_value_correct=1.0, cosine_max_len_value_correct=0.5, cosine_max_len=512, repetition_n_grams=3, repetition_max_penalty=-1.0, reward_model=None, reward_model_plugin=None, multi_turn_scheduler=None, max_turns=None, completion_length_limit_scope='per_round', vllm_server_pass_dataset=False, dynamic_sample=False, max_resample_times=3, overlong_filter=False, soft_max_length=None, soft_cache_length=None, log_entropy=False, tau_pos=1.0, tau_neg=1.05, advantage_estimator='grpo', kl_in_reward=False, num_generations_eval=None, dataset_shuffle=True, rollout_importance_sampling_mode=None, rollout_importance_sampling_threshold=2.0, log_rollout_offpolicy_metrics=False, off_policy_sequence_mask_delta=None)"
527
+ }
model/qwen3_0.6b_lora/args.json.bak ADDED
@@ -0,0 +1,527 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "output_dir": "[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952",
3
+ "overwrite_output_dir": false,
4
+ "do_train": false,
5
+ "do_eval": false,
6
+ "do_predict": false,
7
+ "eval_strategy": "steps",
8
+ "prediction_loss_only": false,
9
+ "per_device_train_batch_size": 24,
10
+ "per_device_eval_batch_size": 24,
11
+ "per_gpu_train_batch_size": null,
12
+ "per_gpu_eval_batch_size": null,
13
+ "gradient_accumulation_steps": 1,
14
+ "eval_accumulation_steps": null,
15
+ "eval_delay": 0,
16
+ "torch_empty_cache_steps": null,
17
+ "learning_rate": 5e-06,
18
+ "weight_decay": 0.1,
19
+ "adam_beta1": 0.9,
20
+ "adam_beta2": 0.95,
21
+ "adam_epsilon": 1e-08,
22
+ "max_grad_norm": 1.0,
23
+ "num_train_epochs": 5.0,
24
+ "max_steps": -1,
25
+ "lr_scheduler_type": "cosine",
26
+ "lr_scheduler_kwargs": null,
27
+ "warmup_ratio": 0.05,
28
+ "warmup_steps": 0,
29
+ "log_level": "passive",
30
+ "log_level_replica": "warning",
31
+ "log_on_each_node": true,
32
+ "logging_dir": "[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952/runs",
33
+ "logging_strategy": "steps",
34
+ "logging_first_step": true,
35
+ "logging_steps": 1,
36
+ "logging_nan_inf_filter": true,
37
+ "save_strategy": "steps",
38
+ "save_steps": 50.0,
39
+ "save_total_limit": null,
40
+ "save_safetensors": true,
41
+ "save_on_each_node": false,
42
+ "save_only_model": false,
43
+ "restore_callback_states_from_checkpoint": false,
44
+ "no_cuda": false,
45
+ "use_cpu": false,
46
+ "use_mps_device": false,
47
+ "seed": 42,
48
+ "data_seed": 42,
49
+ "jit_mode_eval": false,
50
+ "bf16": true,
51
+ "fp16": false,
52
+ "fp16_opt_level": "O1",
53
+ "half_precision_backend": "auto",
54
+ "bf16_full_eval": false,
55
+ "fp16_full_eval": false,
56
+ "tf32": null,
57
+ "local_rank": 0,
58
+ "ddp_backend": null,
59
+ "tpu_num_cores": null,
60
+ "tpu_metrics_debug": false,
61
+ "debug": null,
62
+ "dataloader_drop_last": false,
63
+ "eval_steps": 50.0,
64
+ "dataloader_num_workers": 4,
65
+ "dataloader_prefetch_factor": null,
66
+ "past_index": -1,
67
+ "run_name": "[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952",
68
+ "disable_tqdm": null,
69
+ "remove_unused_columns": false,
70
+ "label_names": null,
71
+ "load_best_model_at_end": false,
72
+ "metric_for_best_model": "loss",
73
+ "greater_is_better": false,
74
+ "ignore_data_skip": false,
75
+ "fsdp": [],
76
+ "fsdp_min_num_params": 0,
77
+ "fsdp_config": null,
78
+ "fsdp_transformer_layer_cls_to_wrap": null,
79
+ "accelerator_config": {
80
+ "dispatch_batches": false
81
+ },
82
+ "parallelism_config": null,
83
+ "deepspeed": {
84
+ "fp16": {
85
+ "enabled": "auto",
86
+ "loss_scale": 0,
87
+ "loss_scale_window": 1000,
88
+ "initial_scale_power": 16,
89
+ "hysteresis": 2,
90
+ "min_loss_scale": 1
91
+ },
92
+ "bf16": {
93
+ "enabled": "auto"
94
+ },
95
+ "zero_optimization": {
96
+ "stage": 2,
97
+ "offload_optimizer": {
98
+ "device": "none",
99
+ "pin_memory": true
100
+ },
101
+ "allgather_partitions": true,
102
+ "allgather_bucket_size": 200000000.0,
103
+ "overlap_comm": false,
104
+ "reduce_scatter": true,
105
+ "reduce_bucket_size": 200000000.0,
106
+ "contiguous_gradients": true
107
+ },
108
+ "gradient_accumulation_steps": "auto",
109
+ "gradient_clipping": "auto",
110
+ "steps_per_print": 2000,
111
+ "train_batch_size": "auto",
112
+ "train_micro_batch_size_per_gpu": "auto",
113
+ "wall_clock_breakdown": false
114
+ },
115
+ "label_smoothing_factor": 0.0,
116
+ "optim": "adamw_torch_fused",
117
+ "optim_args": null,
118
+ "adafactor": false,
119
+ "group_by_length": false,
120
+ "length_column_name": "length",
121
+ "report_to": [
122
+ "tensorboard"
123
+ ],
124
+ "project": "huggingface",
125
+ "trackio_space_id": "trackio",
126
+ "ddp_find_unused_parameters": null,
127
+ "ddp_bucket_cap_mb": null,
128
+ "ddp_broadcast_buffers": null,
129
+ "dataloader_pin_memory": true,
130
+ "dataloader_persistent_workers": false,
131
+ "skip_memory_metrics": true,
132
+ "use_legacy_prediction_loop": false,
133
+ "push_to_hub": false,
134
+ "resume_from_checkpoint": null,
135
+ "hub_model_id": null,
136
+ "hub_strategy": "every_save",
137
+ "hub_token": null,
138
+ "hub_private_repo": null,
139
+ "hub_always_push": false,
140
+ "hub_revision": null,
141
+ "gradient_checkpointing": true,
142
+ "gradient_checkpointing_kwargs": null,
143
+ "include_inputs_for_metrics": false,
144
+ "include_for_metrics": [],
145
+ "eval_do_concat_batches": true,
146
+ "fp16_backend": "auto",
147
+ "push_to_hub_model_id": null,
148
+ "push_to_hub_organization": null,
149
+ "push_to_hub_token": null,
150
+ "mp_parameters": "",
151
+ "auto_find_batch_size": false,
152
+ "full_determinism": false,
153
+ "torchdynamo": null,
154
+ "ray_scope": "last",
155
+ "ddp_timeout": 18000000,
156
+ "torch_compile": false,
157
+ "torch_compile_backend": null,
158
+ "torch_compile_mode": null,
159
+ "include_tokens_per_second": false,
160
+ "include_num_input_tokens_seen": false,
161
+ "neftune_noise_alpha": null,
162
+ "optim_target_modules": null,
163
+ "batch_eval_metrics": false,
164
+ "eval_on_start": false,
165
+ "use_liger_kernel": false,
166
+ "liger_kernel_config": null,
167
+ "eval_use_gather_object": false,
168
+ "average_tokens_across_devices": true,
169
+ "sortish_sampler": false,
170
+ "predict_with_generate": false,
171
+ "generation_max_length": null,
172
+ "generation_num_beams": null,
173
+ "generation_config": null,
174
+ "tuner_backend": "peft",
175
+ "vit_gradient_checkpointing": null,
176
+ "router_aux_loss_coef": 0.0,
177
+ "enable_dft_loss": false,
178
+ "enable_channel_loss": false,
179
+ "check_model": true,
180
+ "acc_strategy": "token",
181
+ "train_dataloader_shuffle": true,
182
+ "max_epochs": null,
183
+ "aligner_lr": null,
184
+ "vit_lr": null,
185
+ "use_logits_to_keep": null,
186
+ "ds3_gather_for_generation": true,
187
+ "resume_only_model": false,
188
+ "optimizer": null,
189
+ "loss_type": "grpo",
190
+ "metric": null,
191
+ "eval_use_evalscope": false,
192
+ "eval_dataset": [],
193
+ "eval_dataset_args": null,
194
+ "eval_limit": null,
195
+ "eval_generation_config": null,
196
+ "extra_eval_args": null,
197
+ "use_flash_ckpt": false,
198
+ "use_ray": false,
199
+ "ray_exp_name": null,
200
+ "device_groups": null,
201
+ "model": "[USER_HOME]/.cache/modelscope/hub/models/Qwen/Qwen3-0.6B",
202
+ "model_type": "qwen3",
203
+ "model_revision": null,
204
+ "task_type": "causal_lm",
205
+ "torch_dtype": "bfloat16",
206
+ "attn_impl": null,
207
+ "new_special_tokens": [],
208
+ "num_labels": null,
209
+ "problem_type": null,
210
+ "rope_scaling": null,
211
+ "device_map": null,
212
+ "max_memory": {},
213
+ "max_model_len": null,
214
+ "local_repo_path": null,
215
+ "init_strategy": null,
216
+ "template": "qwen3",
217
+ "system": null,
218
+ "max_length": 4096,
219
+ "truncation_strategy": "left",
220
+ "max_pixels": null,
221
+ "agent_template": null,
222
+ "norm_bbox": null,
223
+ "use_chat_template": true,
224
+ "padding_side": "right",
225
+ "padding_free": false,
226
+ "loss_scale": "last_round",
227
+ "sequence_parallel_size": 1,
228
+ "template_backend": "swift",
229
+ "response_prefix": null,
230
+ "enable_thinking": null,
231
+ "add_non_thinking_prefix": true,
232
+ "dataset": [
233
+ "[USER_HOME]/AGMS/data/router_grpo_v7/train_grpo.jsonl"
234
+ ],
235
+ "val_dataset": [
236
+ "[USER_HOME]/AGMS/data/router_grpo_v7/val_grpo.jsonl"
237
+ ],
238
+ "cached_dataset": [],
239
+ "cached_val_dataset": [],
240
+ "split_dataset_ratio": 0.0,
241
+ "dataset_num_proc": 1,
242
+ "load_from_cache_file": true,
243
+ "dataset_shuffle": true,
244
+ "val_dataset_shuffle": false,
245
+ "streaming": false,
246
+ "interleave_prob": null,
247
+ "stopping_strategy": "first_exhausted",
248
+ "shuffle_buffer_size": 1000,
249
+ "download_mode": "reuse_dataset_if_exists",
250
+ "columns": {},
251
+ "strict": false,
252
+ "model_name": null,
253
+ "model_author": null,
254
+ "custom_dataset_info": [],
255
+ "quant_method": null,
256
+ "quant_bits": null,
257
+ "hqq_axis": null,
258
+ "bnb_4bit_compute_dtype": "bfloat16",
259
+ "bnb_4bit_quant_type": "nf4",
260
+ "bnb_4bit_use_double_quant": true,
261
+ "bnb_4bit_quant_storage": null,
262
+ "max_new_tokens": 512,
263
+ "temperature": 0.9,
264
+ "top_k": 50,
265
+ "top_p": 0.9,
266
+ "repetition_penalty": 1.0,
267
+ "num_beams": 1,
268
+ "stream": false,
269
+ "stop_words": [],
270
+ "logprobs": false,
271
+ "top_logprobs": null,
272
+ "structured_outputs_regex": null,
273
+ "ckpt_dir": "[USER_HOME]/AGMS/output/router_sft_qwen3_0.6B/v1-20260112-153416/checkpoint-50",
274
+ "lora_modules": [],
275
+ "train_type": "lora",
276
+ "adapters": [
277
+ "[USER_HOME]/AGMS/output/router_sft_qwen3_0.6B/v1-20260112-153416/checkpoint-50"
278
+ ],
279
+ "external_plugins": [
280
+ "[USER_HOME]/AGMS/scripts/router_training/plugin/router_reward.py"
281
+ ],
282
+ "model_kwargs": {},
283
+ "load_args": false,
284
+ "load_data_args": false,
285
+ "packing": false,
286
+ "packing_length": null,
287
+ "packing_num_proc": 1,
288
+ "lazy_tokenize": false,
289
+ "custom_register_path": [],
290
+ "use_hf": false,
291
+ "ignore_args_error": false,
292
+ "use_swift_lora": false,
293
+ "freeze_parameters": [],
294
+ "freeze_parameters_regex": null,
295
+ "freeze_parameters_ratio": 0.0,
296
+ "trainable_parameters": [],
297
+ "trainable_parameters_regex": null,
298
+ "freeze_llm": false,
299
+ "freeze_vit": true,
300
+ "freeze_aligner": true,
301
+ "target_modules": [
302
+ "all-linear"
303
+ ],
304
+ "target_regex": null,
305
+ "target_parameters": null,
306
+ "modules_to_save": [],
307
+ "lora_rank": 16,
308
+ "lora_alpha": 32,
309
+ "lora_dropout": 0.05,
310
+ "lora_bias": "none",
311
+ "lora_dtype": null,
312
+ "lorap_lr_ratio": null,
313
+ "use_rslora": false,
314
+ "use_dora": false,
315
+ "lora_ga_batch_size": 2,
316
+ "lora_ga_iters": 2,
317
+ "lora_ga_max_length": 1024,
318
+ "lora_ga_direction": "ArB2r",
319
+ "lora_ga_scale": "stable",
320
+ "lora_ga_stable_gamma": 16,
321
+ "init_weights": true,
322
+ "fourier_n_frequency": 2000,
323
+ "fourier_scaling": 300.0,
324
+ "boft_block_size": 4,
325
+ "boft_block_num": 0,
326
+ "boft_n_butterfly_factor": 1,
327
+ "boft_dropout": 0.0,
328
+ "vera_rank": 256,
329
+ "vera_projection_prng_key": 0,
330
+ "vera_dropout": 0.0,
331
+ "vera_d_initial": 0.1,
332
+ "adapter_act": "gelu",
333
+ "adapter_length": 128,
334
+ "use_galore": false,
335
+ "galore_target_modules": null,
336
+ "galore_rank": 128,
337
+ "galore_update_proj_gap": 50,
338
+ "galore_scale": 1.0,
339
+ "galore_proj_type": "std",
340
+ "galore_optim_per_parameter": false,
341
+ "galore_with_embedding": false,
342
+ "galore_quantization": false,
343
+ "galore_proj_quant": false,
344
+ "galore_proj_bits": 4,
345
+ "galore_proj_group_size": 256,
346
+ "galore_cos_threshold": 0.4,
347
+ "galore_gamma_proj": 2,
348
+ "galore_queue_size": 5,
349
+ "adalora_target_r": 8,
350
+ "adalora_init_r": 12,
351
+ "adalora_tinit": 0,
352
+ "adalora_tfinal": 0,
353
+ "adalora_deltaT": 1,
354
+ "adalora_beta1": 0.85,
355
+ "adalora_beta2": 0.85,
356
+ "adalora_orth_reg_weight": 0.5,
357
+ "llamapro_num_new_blocks": 4,
358
+ "llamapro_num_groups": null,
359
+ "lisa_activated_layers": 0,
360
+ "lisa_step_interval": 20,
361
+ "reft_layer_key": null,
362
+ "reft_layers": null,
363
+ "reft_rank": 4,
364
+ "reft_intervention_type": "LoreftIntervention",
365
+ "reft_args": null,
366
+ "swanlab_token": null,
367
+ "swanlab_project": "ms-swift",
368
+ "swanlab_workspace": null,
369
+ "swanlab_exp_name": null,
370
+ "swanlab_notification_method": null,
371
+ "swanlab_webhook_url": null,
372
+ "swanlab_secret": null,
373
+ "swanlab_mode": "cloud",
374
+ "add_version": true,
375
+ "create_checkpoint_symlink": false,
376
+ "zero_hpz_partition_size": null,
377
+ "deepspeed_autotp_size": null,
378
+ "early_stop_interval": null,
379
+ "sft_alpha": 0,
380
+ "chord_sft_dataset": [],
381
+ "chord_sft_per_device_train_batch_size": null,
382
+ "chord_enable_phi_function": false,
383
+ "chord_mu_warmup_steps": null,
384
+ "chord_mu_decay_steps": null,
385
+ "chord_mu_peak": null,
386
+ "chord_mu_valley": null,
387
+ "reward_model": null,
388
+ "reward_adapters": [],
389
+ "reward_model_type": null,
390
+ "reward_model_revision": null,
391
+ "num_ppo_epochs": 4,
392
+ "whiten_rewards": false,
393
+ "kl_coef": 0.05,
394
+ "cliprange": 0.2,
395
+ "vf_coef": 0.1,
396
+ "cliprange_value": 0.2,
397
+ "gamma": 1.0,
398
+ "lam": 0.95,
399
+ "num_mini_batches": 1,
400
+ "local_rollout_forward_batch_size": 8,
401
+ "num_sample_generations": 10,
402
+ "response_length": 512,
403
+ "missing_eos_penalty": null,
404
+ "vllm_gpu_memory_utilization": 0.9,
405
+ "vllm_tensor_parallel_size": 1,
406
+ "vllm_pipeline_parallel_size": 1,
407
+ "vllm_enable_expert_parallel": false,
408
+ "vllm_max_num_seqs": null,
409
+ "vllm_max_model_len": null,
410
+ "vllm_disable_custom_all_reduce": true,
411
+ "vllm_enforce_eager": false,
412
+ "vllm_limit_mm_per_prompt": null,
413
+ "vllm_max_lora_rank": 16,
414
+ "vllm_enable_prefix_caching": true,
415
+ "vllm_use_async_engine": false,
416
+ "vllm_quantization": null,
417
+ "vllm_reasoning_parser": null,
418
+ "vllm_disable_cascade_attn": false,
419
+ "vllm_mm_processor_cache_gb": null,
420
+ "vllm_speculative_config": null,
421
+ "vllm_engine_kwargs": {},
422
+ "vllm_data_parallel_size": 1,
423
+ "use_vllm": false,
424
+ "vllm_mode": null,
425
+ "vllm_enable_lora": false,
426
+ "vllm_server_base_url": null,
427
+ "vllm_server_host": null,
428
+ "vllm_server_port": [
429
+ 8000
430
+ ],
431
+ "vllm_server_timeout": 240.0,
432
+ "vllm_server_group_port": null,
433
+ "enable_flattened_weight_sync": true,
434
+ "async_generate": false,
435
+ "sleep_level": 0,
436
+ "move_model_batches": null,
437
+ "offload_optimizer": false,
438
+ "offload_model": false,
439
+ "wandb_log_unique_prompts": null,
440
+ "epsilon": 0.2,
441
+ "epsilon_high": null,
442
+ "delta": null,
443
+ "cosine_min_len_value_wrong": -0.5,
444
+ "cosine_max_len_value_wrong": 0.0,
445
+ "cosine_min_len_value_correct": 1.0,
446
+ "cosine_max_len_value_correct": 0.5,
447
+ "cosine_max_len": null,
448
+ "repetition_n_grams": 3,
449
+ "repetition_max_penalty": -1.0,
450
+ "reward_model_plugin": null,
451
+ "sync_ref_model": false,
452
+ "ref_model_sync_steps": 512,
453
+ "ref_model_mixup_alpha": 0.6,
454
+ "multi_turn_scheduler": null,
455
+ "max_turns": null,
456
+ "completion_length_limit_scope": "per_round",
457
+ "vllm_server_pass_dataset": false,
458
+ "dynamic_sample": false,
459
+ "max_resample_times": 3,
460
+ "overlong_filter": false,
461
+ "soft_max_length": null,
462
+ "soft_cache_length": null,
463
+ "scale_rewards": "group",
464
+ "log_entropy": false,
465
+ "top_entropy_quantile": 1.0,
466
+ "importance_sampling_level": "token",
467
+ "tau_pos": 1.0,
468
+ "tau_neg": 1.05,
469
+ "advantage_estimator": "grpo",
470
+ "kl_in_reward": false,
471
+ "generation_batch_size": null,
472
+ "steps_per_generation": null,
473
+ "num_generations_eval": null,
474
+ "rollout_importance_sampling_mode": null,
475
+ "rollout_importance_sampling_threshold": 2.0,
476
+ "log_rollout_offpolicy_metrics": false,
477
+ "off_policy_sequence_mask_delta": null,
478
+ "num_generations": 12,
479
+ "reward_funcs": [
480
+ "router_final",
481
+ "router_format"
482
+ ],
483
+ "reward_weights": [
484
+ 0.95,
485
+ 0.05
486
+ ],
487
+ "log_completions": true,
488
+ "num_iterations": 1,
489
+ "teacher_model": null,
490
+ "teacher_adapters": [],
491
+ "teacher_model_type": null,
492
+ "teacher_model_revision": null,
493
+ "teacher_deepspeed": null,
494
+ "rlhf_type": "grpo",
495
+ "ref_model": null,
496
+ "ref_adapters": [],
497
+ "ref_model_type": null,
498
+ "ref_model_revision": null,
499
+ "beta": 0.01,
500
+ "label_smoothing": 0,
501
+ "max_completion_length": 512,
502
+ "rpo_alpha": null,
503
+ "ld_alpha": null,
504
+ "discopop_tau": 0.05,
505
+ "loss_weights": null,
506
+ "cpo_alpha": 1.0,
507
+ "simpo_gamma": 1,
508
+ "desirable_weight": 1.0,
509
+ "undesirable_weight": 1.0,
510
+ "center_rewards_coefficient": null,
511
+ "lmbda": 0.5,
512
+ "seq_kd": false,
513
+ "offload_teacher_model": false,
514
+ "rank": 0,
515
+ "global_world_size": 4,
516
+ "local_world_size": 4,
517
+ "model_suffix": "Qwen3-0.6B",
518
+ "model_info": "ModelInfo(model_type='qwen3', model_dir='[USER_HOME]/.cache/modelscope/hub/models/Qwen/Qwen3-0.6B', torch_dtype=torch.bfloat16, max_model_len=40960, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=False, config=None, task_type='causal_lm', num_labels=None)",
519
+ "model_meta": "ModelMeta(model_type='qwen3', model_groups=[ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3-0.6B-Base', hf_model_id='Qwen/Qwen3-0.6B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B-Base', hf_model_id='Qwen/Qwen3-1.7B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-Base', hf_model_id='Qwen/Qwen3-4B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-Base', hf_model_id='Qwen/Qwen3-8B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-Base', hf_model_id='Qwen/Qwen3-14B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-0.6B', hf_model_id='Qwen/Qwen3-0.6B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B', hf_model_id='Qwen/Qwen3-1.7B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B', hf_model_id='Qwen/Qwen3-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B', hf_model_id='Qwen/Qwen3-8B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B', hf_model_id='Qwen/Qwen3-14B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B', hf_model_id='Qwen/Qwen3-32B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-0.6B-FP8', hf_model_id='Qwen/Qwen3-0.6B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B-FP8', hf_model_id='Qwen/Qwen3-1.7B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-FP8', hf_model_id='Qwen/Qwen3-4B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-FP8', hf_model_id='Qwen/Qwen3-8B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-FP8', hf_model_id='Qwen/Qwen3-14B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B-FP8', hf_model_id='Qwen/Qwen3-32B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-AWQ', hf_model_id='Qwen/Qwen3-4B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-AWQ', hf_model_id='Qwen/Qwen3-8B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-AWQ', hf_model_id='Qwen/Qwen3-14B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B-AWQ', hf_model_id='Qwen/Qwen3-32B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='swift/Qwen3-32B-AWQ', hf_model_id=None, model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='qwen3', get_function=<function get_model_tokenizer_with_flash_attn at 0x7f9ba23d84a0>, model_arch=ModelKeys(arch_name='llama', embedding='model.embed_tokens', module_list='model.layers', lm_head='lm_head', q_proj='model.layers.{}.self_attn.q_proj', k_proj='model.layers.{}.self_attn.k_proj', v_proj='model.layers.{}.self_attn.v_proj', o_proj='model.layers.{}.self_attn.o_proj', attention='model.layers.{}.self_attn', mlp='model.layers.{}.mlp', down_proj='model.layers.{}.mlp.down_proj', qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None), architectures=['Qwen3ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=False, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers>=4.51'], tags=[])",
520
+ "model_dir": "[USER_HOME]/.cache/modelscope/hub/models/Qwen/Qwen3-0.6B",
521
+ "_val_dataset_exists": [
522
+ "[USER_HOME]/AGMS/data/router_grpo_v7/val_grpo.jsonl"
523
+ ],
524
+ "hub": "<class 'swift.hub.hub.MSHub'>",
525
+ "evaluation_strategy": "steps",
526
+ "training_args": "GRPOConfig(output_dir='[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=24, per_device_eval_batch_size=24, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=1, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=5.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.05, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=1, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=50, save_total_limit=None, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=True, eval_steps=50, dataloader_num_workers=4, dataloader_prefetch_factor=2, past_index=-1, run_name='[USER_HOME]/AGMS/output/router_grpo_qwen3_0.6B/v26-20260115-031952', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), parallelism_config=None, deepspeed={'fp16': {'enabled': 'auto', 'loss_scale': 0, 'loss_scale_window': 1000, 'initial_scale_power': 16, 'hysteresis': 2, 'min_loss_scale': 1}, 'bf16': {'enabled': 'auto'}, 'zero_optimization': {'stage': 2, 'offload_optimizer': {'device': 'none', 'pin_memory': True}, 'allgather_partitions': True, 'allgather_bucket_size': 200000000.0, 'overlap_comm': False, 'reduce_scatter': True, 'reduce_bucket_size': 200000000.0, 'contiguous_gradients': True}, 'gradient_accumulation_steps': 'auto', 'gradient_clipping': 'auto', 'steps_per_print': 2000, 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'wall_clock_breakdown': False}, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH_FUSED: 'adamw_torch_fused'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], project='huggingface', trackio_space_id='trackio', ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, hub_revision=None, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, liger_kernel_config=None, eval_use_gather_object=False, average_tokens_across_devices=None, model_init_kwargs=None, disable_dropout=False, max_prompt_length=512, num_generations=12, max_completion_length=512, ds3_gather_for_generation=True, shuffle_dataset=True, generation_batch_size=96, steps_per_generation=1, temperature=0.9, top_p=0.9, top_k=50, min_p=None, generation_kwargs=None, repetition_penalty=1.0, use_transformers_paged=False, cache_implementation=None, use_vllm=False, vllm_mode=None, vllm_model_impl='vllm', vllm_enable_sleep_mode=False, vllm_guided_decoding_regex=None, vllm_server_base_url=None, vllm_server_host=None, vllm_server_port=[8000], vllm_server_timeout=240.0, vllm_gpu_memory_utilization=0.9, vllm_tensor_parallel_size=1, beta=0.01, num_iterations=1, epsilon=0.2, delta=None, epsilon_high=None, importance_sampling_level='token', reward_weights=[0.95, 0.05], scale_rewards='group', loss_type='grpo', mask_truncated_completions=False, sync_ref_model=False, ref_model_mixup_alpha=0.6, ref_model_sync_steps=512, top_entropy_quantile=1.0, use_liger_loss=False, vllm_importance_sampling_correction=True, vllm_importance_sampling_cap=2.0, log_completions=True, num_completions_to_print=None, wandb_log_unique_prompts=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=None, vit_lr=None, use_logits_to_keep=None, resume_only_model=False, optimizer=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, task_type='causal_lm', problem_type=None, vllm_pipeline_parallel_size=1, vllm_enable_expert_parallel=False, vllm_max_num_seqs=None, vllm_max_model_len=None, vllm_disable_custom_all_reduce=True, vllm_enforce_eager=False, vllm_limit_mm_per_prompt=None, vllm_max_lora_rank=16, vllm_enable_prefix_caching=True, vllm_use_async_engine=False, vllm_quantization=None, vllm_reasoning_parser=None, vllm_disable_cascade_attn=False, vllm_mm_processor_cache_gb=None, vllm_speculative_config=None, vllm_engine_kwargs={}, vllm_data_parallel_size=1, stop_words=[], vllm_enable_lora=False, lora_rank=16, vllm_server_group_port=None, enable_flattened_weight_sync=True, async_generate=False, structured_outputs_regex=None, sleep_level=0, move_model_batches=None, offload_optimizer=False, offload_model=False, cosine_min_len_value_wrong=-0.5, cosine_max_len_value_wrong=0.0, cosine_min_len_value_correct=1.0, cosine_max_len_value_correct=0.5, cosine_max_len=512, repetition_n_grams=3, repetition_max_penalty=-1.0, reward_model=None, reward_model_plugin=None, multi_turn_scheduler=None, max_turns=None, completion_length_limit_scope='per_round', vllm_server_pass_dataset=False, dynamic_sample=False, max_resample_times=3, overlong_filter=False, soft_max_length=None, soft_cache_length=None, log_entropy=False, tau_pos=1.0, tau_neg=1.05, advantage_estimator='grpo', kl_in_reward=False, num_generations_eval=None, dataset_shuffle=True, rollout_importance_sampling_mode=None, rollout_importance_sampling_threshold=2.0, log_rollout_offpolicy_metrics=False, off_policy_sequence_mask_delta=None)"
527
+ }
model/qwen3_0.6b_lora/chat_template.jinja ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
46
+ {%- else %}
47
+ {{- '<|im_start|>' + message.role + '\n' + content }}
48
+ {%- endif %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' + content }}
51
+ {%- endif %}
52
+ {%- if message.tool_calls %}
53
+ {%- for tool_call in message.tool_calls %}
54
+ {%- if (loop.first and content) or (not loop.first) %}
55
+ {{- '\n' }}
56
+ {%- endif %}
57
+ {%- if tool_call.function %}
58
+ {%- set tool_call = tool_call.function %}
59
+ {%- endif %}
60
+ {{- '<tool_call>\n{"name": "' }}
61
+ {{- tool_call.name }}
62
+ {{- '", "arguments": ' }}
63
+ {%- if tool_call.arguments is string %}
64
+ {{- tool_call.arguments }}
65
+ {%- else %}
66
+ {{- tool_call.arguments | tojson }}
67
+ {%- endif %}
68
+ {{- '}\n</tool_call>' }}
69
+ {%- endfor %}
70
+ {%- endif %}
71
+ {{- '<|im_end|>\n' }}
72
+ {%- elif message.role == "tool" %}
73
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
74
+ {{- '<|im_start|>user' }}
75
+ {%- endif %}
76
+ {{- '\n<tool_response>\n' }}
77
+ {{- content }}
78
+ {{- '\n</tool_response>' }}
79
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
+ {{- '<|im_end|>\n' }}
81
+ {%- endif %}
82
+ {%- endif %}
83
+ {%- endfor %}
84
+ {%- if add_generation_prompt %}
85
+ {{- '<|im_start|>assistant\n' }}
86
+ {%- if enable_thinking is defined and enable_thinking is false %}
87
+ {{- '<think>\n\n</think>\n\n' }}
88
+ {%- endif %}
89
+ {%- endif %}
model/qwen3_0.6b_lora/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention"
44
+ ],
45
+ "max_position_embeddings": 40960,
46
+ "max_window_layers": 28,
47
+ "model_type": "qwen3",
48
+ "num_attention_heads": 16,
49
+ "num_hidden_layers": 28,
50
+ "num_key_value_heads": 8,
51
+ "pad_token_id": 151643,
52
+ "rms_norm_eps": 1e-06,
53
+ "rope_scaling": null,
54
+ "rope_theta": 1000000,
55
+ "sliding_window": null,
56
+ "tie_word_embeddings": true,
57
+ "transformers_version": "4.57.3",
58
+ "use_cache": true,
59
+ "use_sliding_window": false,
60
+ "vocab_size": 151936
61
+ }
model/qwen3_0.6b_lora/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "temperature": 0.6,
10
+ "top_k": 20,
11
+ "top_p": 0.95,
12
+ "transformers_version": "4.57.3"
13
+ }
model/qwen3_0.6b_lora/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model/qwen3_0.6b_lora/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bddf58450b039eedac5867e240ce5c75901d71a6fb7ce785d6ad302432cc0c6
3
+ size 1192135096
model/qwen3_0.6b_lora/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
model/qwen3_0.6b_lora/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
model/qwen3_0.6b_lora/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 131072,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
model/qwen3_0.6b_lora/vocab.json ADDED
The diff for this file is too large to render. See raw diff