Upload folder using huggingface_hub
Browse files- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/args.json +393 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/chat_template.jinja +154 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/config.json +113 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/generation_config.json +9 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/latest +1 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model-00001-of-00003.safetensors +3 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model-00002-of-00003.safetensors +3 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model-00003-of-00003.safetensors +3 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model.safetensors.index.json +740 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/preprocessor_config.json +21 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/processor_config.json +60 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/tokenizer.json +3 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/tokenizer_config.json +33 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/trainer_state.json +666 -0
- qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/zero_to_fp32.py +760 -0
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/args.json
ADDED
|
@@ -0,0 +1,393 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"output_dir": "/NHNHOME/data/sanghyeok/qwen-cua/runs/qwen35_4b_planA_wm_phase1-20260601-041141/v0-20260601-041256",
|
| 3 |
+
"per_device_train_batch_size": 1,
|
| 4 |
+
"num_train_epochs": 1.0,
|
| 5 |
+
"max_steps": -1,
|
| 6 |
+
"learning_rate": 1e-05,
|
| 7 |
+
"lr_scheduler_type": "cosine",
|
| 8 |
+
"lr_scheduler_kwargs": null,
|
| 9 |
+
"warmup_steps": 200.0,
|
| 10 |
+
"optim": "adamw_torch_fused",
|
| 11 |
+
"optim_args": null,
|
| 12 |
+
"weight_decay": 0.0,
|
| 13 |
+
"adam_beta1": 0.9,
|
| 14 |
+
"adam_beta2": 0.95,
|
| 15 |
+
"adam_epsilon": 1e-08,
|
| 16 |
+
"optim_target_modules": null,
|
| 17 |
+
"gradient_accumulation_steps": 16,
|
| 18 |
+
"average_tokens_across_devices": true,
|
| 19 |
+
"max_grad_norm": 1.0,
|
| 20 |
+
"label_smoothing_factor": 0.0,
|
| 21 |
+
"bf16": true,
|
| 22 |
+
"fp16": false,
|
| 23 |
+
"bf16_full_eval": false,
|
| 24 |
+
"fp16_full_eval": false,
|
| 25 |
+
"tf32": true,
|
| 26 |
+
"gradient_checkpointing": true,
|
| 27 |
+
"gradient_checkpointing_kwargs": null,
|
| 28 |
+
"torch_compile": false,
|
| 29 |
+
"torch_compile_backend": null,
|
| 30 |
+
"torch_compile_mode": null,
|
| 31 |
+
"use_liger_kernel": false,
|
| 32 |
+
"liger_kernel_config": null,
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"neftune_noise_alpha": null,
|
| 35 |
+
"torch_empty_cache_steps": null,
|
| 36 |
+
"auto_find_batch_size": false,
|
| 37 |
+
"logging_strategy": "steps",
|
| 38 |
+
"logging_steps": 10,
|
| 39 |
+
"logging_first_step": true,
|
| 40 |
+
"log_on_each_node": true,
|
| 41 |
+
"logging_nan_inf_filter": true,
|
| 42 |
+
"include_num_input_tokens_seen": false,
|
| 43 |
+
"log_level": "passive",
|
| 44 |
+
"log_level_replica": "warning",
|
| 45 |
+
"disable_tqdm": null,
|
| 46 |
+
"report_to": [
|
| 47 |
+
"wandb"
|
| 48 |
+
],
|
| 49 |
+
"run_name": "qwen35_4b_planA_wm_phase1-20260601-041141",
|
| 50 |
+
"project": "huggingface",
|
| 51 |
+
"trackio_space_id": null,
|
| 52 |
+
"trackio_bucket_id": null,
|
| 53 |
+
"trackio_static_space_id": null,
|
| 54 |
+
"eval_strategy": "steps",
|
| 55 |
+
"eval_steps": 500.0,
|
| 56 |
+
"eval_delay": 0,
|
| 57 |
+
"per_device_eval_batch_size": 1,
|
| 58 |
+
"prediction_loss_only": false,
|
| 59 |
+
"eval_on_start": false,
|
| 60 |
+
"eval_do_concat_batches": true,
|
| 61 |
+
"eval_use_gather_object": false,
|
| 62 |
+
"eval_accumulation_steps": null,
|
| 63 |
+
"include_for_metrics": [],
|
| 64 |
+
"batch_eval_metrics": false,
|
| 65 |
+
"save_only_model": false,
|
| 66 |
+
"save_strategy": "steps",
|
| 67 |
+
"save_steps": 100.0,
|
| 68 |
+
"save_on_each_node": false,
|
| 69 |
+
"save_total_limit": 3,
|
| 70 |
+
"enable_jit_checkpoint": false,
|
| 71 |
+
"push_to_hub": false,
|
| 72 |
+
"hub_token": null,
|
| 73 |
+
"hub_private_repo": null,
|
| 74 |
+
"hub_model_id": null,
|
| 75 |
+
"hub_strategy": "every_save",
|
| 76 |
+
"hub_always_push": false,
|
| 77 |
+
"hub_revision": null,
|
| 78 |
+
"load_best_model_at_end": false,
|
| 79 |
+
"metric_for_best_model": "loss",
|
| 80 |
+
"greater_is_better": false,
|
| 81 |
+
"ignore_data_skip": false,
|
| 82 |
+
"restore_callback_states_from_checkpoint": false,
|
| 83 |
+
"full_determinism": false,
|
| 84 |
+
"seed": 42,
|
| 85 |
+
"data_seed": 42,
|
| 86 |
+
"use_cpu": false,
|
| 87 |
+
"accelerator_config": {
|
| 88 |
+
"dispatch_batches": false
|
| 89 |
+
},
|
| 90 |
+
"parallelism_config": null,
|
| 91 |
+
"dataloader_drop_last": false,
|
| 92 |
+
"dataloader_num_workers": 8,
|
| 93 |
+
"dataloader_pin_memory": true,
|
| 94 |
+
"dataloader_persistent_workers": true,
|
| 95 |
+
"dataloader_prefetch_factor": 4,
|
| 96 |
+
"remove_unused_columns": true,
|
| 97 |
+
"label_names": null,
|
| 98 |
+
"train_sampling_strategy": "random",
|
| 99 |
+
"length_column_name": "length",
|
| 100 |
+
"ddp_find_unused_parameters": null,
|
| 101 |
+
"ddp_bucket_cap_mb": null,
|
| 102 |
+
"ddp_broadcast_buffers": null,
|
| 103 |
+
"ddp_static_graph": null,
|
| 104 |
+
"ddp_backend": null,
|
| 105 |
+
"ddp_timeout": 18000000,
|
| 106 |
+
"fsdp": [],
|
| 107 |
+
"fsdp_config": null,
|
| 108 |
+
"deepspeed": {
|
| 109 |
+
"bf16": {
|
| 110 |
+
"enabled": true
|
| 111 |
+
},
|
| 112 |
+
"fp16": {
|
| 113 |
+
"enabled": false
|
| 114 |
+
},
|
| 115 |
+
"zero_optimization": {
|
| 116 |
+
"stage": 2,
|
| 117 |
+
"overlap_comm": true,
|
| 118 |
+
"contiguous_gradients": true,
|
| 119 |
+
"reduce_bucket_size": 500000000.0,
|
| 120 |
+
"allgather_bucket_size": 500000000.0,
|
| 121 |
+
"reduce_scatter": true,
|
| 122 |
+
"round_robin_gradients": true
|
| 123 |
+
},
|
| 124 |
+
"gradient_clipping": "auto",
|
| 125 |
+
"gradient_accumulation_steps": "auto",
|
| 126 |
+
"train_batch_size": "auto",
|
| 127 |
+
"train_micro_batch_size_per_gpu": "auto",
|
| 128 |
+
"steps_per_print": 100,
|
| 129 |
+
"wall_clock_breakdown": false,
|
| 130 |
+
"_comment_bf16_safety": "DeepSpeed 0.19 BF16_Optimizer keeps fp32 master grad partitions (fp32_groups_flat_partition + accumulate_hp_grads_and_remove_lp hook), so gradient accumulation across many micro-batches is done in fp32. bf16 is only used for the param master copy and forward/backward activations. This is safe at grad_accum=16+; no need to force fp32 collectives."
|
| 131 |
+
},
|
| 132 |
+
"debug": null,
|
| 133 |
+
"skip_memory_metrics": true,
|
| 134 |
+
"do_train": false,
|
| 135 |
+
"do_eval": false,
|
| 136 |
+
"do_predict": false,
|
| 137 |
+
"resume_from_checkpoint": null,
|
| 138 |
+
"warmup_ratio": null,
|
| 139 |
+
"logging_dir": "/NHNHOME/data/sanghyeok/qwen-cua/runs/qwen35_4b_planA_wm_phase1-20260601-041141/v0-20260601-041256/runs",
|
| 140 |
+
"local_rank": 0,
|
| 141 |
+
"sortish_sampler": false,
|
| 142 |
+
"predict_with_generate": false,
|
| 143 |
+
"generation_max_length": null,
|
| 144 |
+
"generation_num_beams": null,
|
| 145 |
+
"generation_config": null,
|
| 146 |
+
"tuner_backend": "peft",
|
| 147 |
+
"vit_gradient_checkpointing": false,
|
| 148 |
+
"router_aux_loss_coef": 0.0,
|
| 149 |
+
"enable_dft_loss": false,
|
| 150 |
+
"enable_channel_loss": false,
|
| 151 |
+
"safe_serialization": true,
|
| 152 |
+
"max_shard_size": "5GB",
|
| 153 |
+
"check_model": true,
|
| 154 |
+
"acc_strategy": "token",
|
| 155 |
+
"train_dataloader_shuffle": true,
|
| 156 |
+
"group_by_length": false,
|
| 157 |
+
"max_epochs": null,
|
| 158 |
+
"aligner_lr": null,
|
| 159 |
+
"vit_lr": null,
|
| 160 |
+
"use_logits_to_keep": null,
|
| 161 |
+
"ds3_gather_for_generation": true,
|
| 162 |
+
"resume_only_model": false,
|
| 163 |
+
"optimizer": null,
|
| 164 |
+
"loss_type": null,
|
| 165 |
+
"eval_metric": null,
|
| 166 |
+
"callbacks": [],
|
| 167 |
+
"early_stop_interval": null,
|
| 168 |
+
"eval_use_evalscope": false,
|
| 169 |
+
"eval_dataset": [],
|
| 170 |
+
"eval_dataset_args": null,
|
| 171 |
+
"eval_limit": null,
|
| 172 |
+
"eval_generation_config": null,
|
| 173 |
+
"extra_eval_args": null,
|
| 174 |
+
"tuner_type": "full",
|
| 175 |
+
"use_galore": false,
|
| 176 |
+
"galore_target_modules": null,
|
| 177 |
+
"galore_rank": 128,
|
| 178 |
+
"galore_update_proj_gap": 50,
|
| 179 |
+
"galore_scale": 1.0,
|
| 180 |
+
"galore_proj_type": "std",
|
| 181 |
+
"galore_optim_per_parameter": false,
|
| 182 |
+
"galore_with_embedding": false,
|
| 183 |
+
"galore_quantization": false,
|
| 184 |
+
"galore_proj_quant": false,
|
| 185 |
+
"galore_proj_bits": 4,
|
| 186 |
+
"galore_proj_group_size": 256,
|
| 187 |
+
"galore_cos_threshold": 0.4,
|
| 188 |
+
"galore_gamma_proj": 2,
|
| 189 |
+
"galore_queue_size": 5,
|
| 190 |
+
"lisa_activated_layers": 0,
|
| 191 |
+
"lisa_step_interval": 20,
|
| 192 |
+
"use_flash_ckpt": false,
|
| 193 |
+
"use_ray": false,
|
| 194 |
+
"ray_exp_name": null,
|
| 195 |
+
"device_groups": null,
|
| 196 |
+
"model": "/NHNHOME/data/sanghyeok/qwen-cua/hf_cache/hub/models--Qwen--Qwen3.5-4B/snapshots/851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
|
| 197 |
+
"model_type": "qwen3_5",
|
| 198 |
+
"model_revision": null,
|
| 199 |
+
"task_type": "causal_lm",
|
| 200 |
+
"torch_dtype": "bfloat16",
|
| 201 |
+
"attn_impl": "flash_attention_2",
|
| 202 |
+
"experts_impl": null,
|
| 203 |
+
"new_special_tokens": [],
|
| 204 |
+
"num_labels": null,
|
| 205 |
+
"problem_type": null,
|
| 206 |
+
"rope_scaling": null,
|
| 207 |
+
"device_map": null,
|
| 208 |
+
"max_memory": {},
|
| 209 |
+
"max_model_len": null,
|
| 210 |
+
"local_repo_path": null,
|
| 211 |
+
"init_strategy": null,
|
| 212 |
+
"template": "qwen3_5",
|
| 213 |
+
"system": null,
|
| 214 |
+
"max_length": 32768,
|
| 215 |
+
"truncation_strategy": "delete",
|
| 216 |
+
"max_pixels": 16777216,
|
| 217 |
+
"agent_template": null,
|
| 218 |
+
"norm_bbox": null,
|
| 219 |
+
"use_chat_template": true,
|
| 220 |
+
"padding_side": "right",
|
| 221 |
+
"padding_free": true,
|
| 222 |
+
"loss_scale": "default",
|
| 223 |
+
"sequence_parallel_size": 1,
|
| 224 |
+
"template_backend": "swift",
|
| 225 |
+
"response_prefix": null,
|
| 226 |
+
"enable_thinking": null,
|
| 227 |
+
"add_non_thinking_prefix": true,
|
| 228 |
+
"dataset": [
|
| 229 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l1/ubuntu/train_wm.jsonl#30000",
|
| 230 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l2/ubuntu/train_wm.jsonl#30000",
|
| 231 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l3/ubuntu/train_wm.jsonl#30000",
|
| 232 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l1/win_mac/train_wm.jsonl#100000",
|
| 233 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l2/win_mac/train_wm.jsonl#100000",
|
| 234 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l3/win_mac/train_wm.jsonl#100000",
|
| 235 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/GUI-360/l1/train_wm.jsonl#30000",
|
| 236 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/GUI-360/l2/train_wm.jsonl#30000",
|
| 237 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/GUI-360/l3/train_wm.jsonl#30000",
|
| 238 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/OS-Atlas/linux/train.jsonl",
|
| 239 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/OS-Atlas/macos/train.jsonl",
|
| 240 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/OS-Atlas/windows/train.jsonl#50000",
|
| 241 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/ShowUI/desktop/train.jsonl",
|
| 242 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/ShowUI-web/train.jsonl#50000",
|
| 243 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/GroundCUA/train.jsonl#50000",
|
| 244 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/ScreenSpot/train.jsonl",
|
| 245 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/UGround/web_direct_150k_description_filtered/train.jsonl",
|
| 246 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/UGround/web_direct_258k_function_filtered/train.jsonl#50000",
|
| 247 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/UGround/web_hybrid_773k_max_25qa_filtered/train.jsonl#80000",
|
| 248 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/Multimodal-Mind2Web/train.jsonl"
|
| 249 |
+
],
|
| 250 |
+
"val_dataset": [
|
| 251 |
+
"/NHNHOME/data/sanghyeok/qwen-cua/data/AgentNet/l2/ubuntu/val_wm.jsonl"
|
| 252 |
+
],
|
| 253 |
+
"cached_dataset": [],
|
| 254 |
+
"cached_val_dataset": [],
|
| 255 |
+
"split_dataset_ratio": 0.0,
|
| 256 |
+
"dataset_num_proc": 4,
|
| 257 |
+
"load_from_cache_file": false,
|
| 258 |
+
"dataset_shuffle": true,
|
| 259 |
+
"val_dataset_shuffle": false,
|
| 260 |
+
"streaming": false,
|
| 261 |
+
"interleave_prob": null,
|
| 262 |
+
"stopping_strategy": "first_exhausted",
|
| 263 |
+
"shuffle_buffer_size": 1000,
|
| 264 |
+
"download_mode": "reuse_dataset_if_exists",
|
| 265 |
+
"columns": {},
|
| 266 |
+
"strict": false,
|
| 267 |
+
"disable_auto_column_mapping": false,
|
| 268 |
+
"model_name": null,
|
| 269 |
+
"model_author": null,
|
| 270 |
+
"custom_dataset_info": [],
|
| 271 |
+
"quant_method": null,
|
| 272 |
+
"quant_bits": null,
|
| 273 |
+
"hqq_axis": null,
|
| 274 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 275 |
+
"bnb_4bit_quant_type": "nf4",
|
| 276 |
+
"bnb_4bit_use_double_quant": true,
|
| 277 |
+
"bnb_4bit_quant_storage": null,
|
| 278 |
+
"max_new_tokens": 64,
|
| 279 |
+
"temperature": 0.0,
|
| 280 |
+
"top_k": null,
|
| 281 |
+
"top_p": null,
|
| 282 |
+
"repetition_penalty": null,
|
| 283 |
+
"num_beams": 1,
|
| 284 |
+
"stream": false,
|
| 285 |
+
"stop_words": [],
|
| 286 |
+
"logprobs": false,
|
| 287 |
+
"top_logprobs": null,
|
| 288 |
+
"structured_outputs_regex": null,
|
| 289 |
+
"adapters": [],
|
| 290 |
+
"external_plugins": [],
|
| 291 |
+
"custom_register_path": [],
|
| 292 |
+
"model_kwargs": {},
|
| 293 |
+
"enable_npu_model_patch": true,
|
| 294 |
+
"load_args": false,
|
| 295 |
+
"load_data_args": false,
|
| 296 |
+
"packing": true,
|
| 297 |
+
"packing_length": 32768,
|
| 298 |
+
"packing_num_proc": 1,
|
| 299 |
+
"lazy_tokenize": false,
|
| 300 |
+
"use_hf": false,
|
| 301 |
+
"ignore_args_error": false,
|
| 302 |
+
"use_swift_lora": false,
|
| 303 |
+
"freeze_parameters": [
|
| 304 |
+
"model.visual",
|
| 305 |
+
"model.visual.merger"
|
| 306 |
+
],
|
| 307 |
+
"freeze_parameters_regex": null,
|
| 308 |
+
"freeze_parameters_ratio": 0.0,
|
| 309 |
+
"trainable_parameters": [],
|
| 310 |
+
"trainable_parameters_regex": null,
|
| 311 |
+
"freeze_llm": false,
|
| 312 |
+
"freeze_vit": true,
|
| 313 |
+
"freeze_aligner": true,
|
| 314 |
+
"target_modules": [
|
| 315 |
+
"all-linear"
|
| 316 |
+
],
|
| 317 |
+
"target_regex": null,
|
| 318 |
+
"target_parameters": null,
|
| 319 |
+
"modules_to_save": [],
|
| 320 |
+
"lora_rank": 8,
|
| 321 |
+
"lora_alpha": 32,
|
| 322 |
+
"lora_dropout": 0.05,
|
| 323 |
+
"lora_bias": "none",
|
| 324 |
+
"lora_dtype": null,
|
| 325 |
+
"lorap_lr_ratio": null,
|
| 326 |
+
"use_rslora": false,
|
| 327 |
+
"use_dora": false,
|
| 328 |
+
"lora_ga_batch_size": 2,
|
| 329 |
+
"lora_ga_iters": 2,
|
| 330 |
+
"lora_ga_max_length": 1024,
|
| 331 |
+
"lora_ga_direction": "ArB2r",
|
| 332 |
+
"lora_ga_scale": "stable",
|
| 333 |
+
"lora_ga_stable_gamma": 16,
|
| 334 |
+
"init_weights": true,
|
| 335 |
+
"fourier_n_frequency": 2000,
|
| 336 |
+
"fourier_scaling": 300.0,
|
| 337 |
+
"boft_block_size": 4,
|
| 338 |
+
"boft_block_num": 0,
|
| 339 |
+
"boft_n_butterfly_factor": 1,
|
| 340 |
+
"boft_dropout": 0.0,
|
| 341 |
+
"vera_rank": 256,
|
| 342 |
+
"vera_projection_prng_key": 0,
|
| 343 |
+
"vera_dropout": 0.0,
|
| 344 |
+
"vera_d_initial": 0.1,
|
| 345 |
+
"adapter_act": "gelu",
|
| 346 |
+
"adapter_length": 128,
|
| 347 |
+
"adalora_target_r": 8,
|
| 348 |
+
"adalora_init_r": 12,
|
| 349 |
+
"adalora_tinit": 0,
|
| 350 |
+
"adalora_tfinal": 0,
|
| 351 |
+
"adalora_deltaT": 1,
|
| 352 |
+
"adalora_beta1": 0.85,
|
| 353 |
+
"adalora_beta2": 0.85,
|
| 354 |
+
"adalora_orth_reg_weight": 0.5,
|
| 355 |
+
"llamapro_num_new_blocks": 4,
|
| 356 |
+
"llamapro_num_groups": null,
|
| 357 |
+
"reft_layer_key": null,
|
| 358 |
+
"reft_layers": null,
|
| 359 |
+
"reft_rank": 4,
|
| 360 |
+
"reft_intervention_type": "LoreftIntervention",
|
| 361 |
+
"reft_args": null,
|
| 362 |
+
"swanlab_token": null,
|
| 363 |
+
"swanlab_project": "ms-swift",
|
| 364 |
+
"swanlab_workspace": null,
|
| 365 |
+
"swanlab_exp_name": null,
|
| 366 |
+
"swanlab_notification_method": null,
|
| 367 |
+
"swanlab_webhook_url": null,
|
| 368 |
+
"swanlab_secret": null,
|
| 369 |
+
"swanlab_sender_email": null,
|
| 370 |
+
"swanlab_receiver_email": null,
|
| 371 |
+
"swanlab_smtp_server": null,
|
| 372 |
+
"swanlab_smtp_port": null,
|
| 373 |
+
"swanlab_email_language": "zh",
|
| 374 |
+
"swanlab_mode": "cloud",
|
| 375 |
+
"add_version": true,
|
| 376 |
+
"create_checkpoint_symlink": false,
|
| 377 |
+
"zero_hpz_partition_size": null,
|
| 378 |
+
"deepspeed_autotp_size": null,
|
| 379 |
+
"swift_version": "4.2.1",
|
| 380 |
+
"ckpt_dir": null,
|
| 381 |
+
"rank": 0,
|
| 382 |
+
"global_world_size": 8,
|
| 383 |
+
"local_world_size": 8,
|
| 384 |
+
"model_suffix": "Qwen3.5-4B",
|
| 385 |
+
"model_info": "ModelInfo(model_type='qwen3_5', model_dir='/NHNHOME/data/sanghyeok/qwen-cua/hf_cache/hub/models--Qwen--Qwen3.5-4B/snapshots/851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a', torch_dtype=torch.bfloat16, max_model_len=262144, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
|
| 386 |
+
"model_meta": "ModelMeta(model_type='qwen3_5', model_groups=[ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3.5-0.8B', hf_model_id='Qwen/Qwen3.5-0.8B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-2B', hf_model_id='Qwen/Qwen3.5-2B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-4B', hf_model_id='Qwen/Qwen3.5-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-9B', hf_model_id='Qwen/Qwen3.5-9B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-27B', hf_model_id='Qwen/Qwen3.5-27B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-27B-FP8', hf_model_id='Qwen/Qwen3.5-27B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-0.8B-Base', hf_model_id='Qwen/Qwen3.5-0.8B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-2B-Base', hf_model_id='Qwen/Qwen3.5-2B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-4B-Base', hf_model_id='Qwen/Qwen3.5-4B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-9B-Base', hf_model_id='Qwen/Qwen3.5-9B-Base', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_5', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3.6-27B', hf_model_id='Qwen/Qwen3.6-27B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.6-27B-FP8', hf_model_id='Qwen/Qwen3.6-27B-FP8', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_5', ignore_patterns=None, requires=None, tags=[])], loader=<class 'swift.model.models.qwen.Qwen3_5Loader'>, template='qwen3_5', model_arch=MultiModelKeys(arch_name='qwen2_vl', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model.language_model', 'lm_head'], aligner=['model.visual.merger'], vision_tower=['model.visual'], generator=[]), mcore_model_type=None, architectures=['Qwen3_5ForConditionalGeneration'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, task_type=None, ignore_patterns=None, requires=['transformers>=5.0.0.dev', 'qwen_vl_utils>=0.0.14', 'decord'], tags=[])",
|
| 387 |
+
"model_dir": "/NHNHOME/data/sanghyeok/qwen-cua/hf_cache/hub/models--Qwen--Qwen3.5-4B/snapshots/851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
|
| 388 |
+
"template_meta": "QwenTemplateMeta(template_type='qwen3_5', prefix=[], prompt=['<|im_start|>user\\n{{QUERY}}<|im_end|>\\n<|im_start|>assistant\\n'], chat_sep=['<|im_end|>\\n'], suffix=['<|im_end|>\\n'], template_cls=<class 'swift.template.templates.qwen.Qwen3_5Template'>, system_prefix=['<|im_start|>system\\n{{SYSTEM}}<|im_end|>\\n'], default_system=None, auto_add_bos=False, stop_words=['<|endoftext|>'], agent_template='qwen3_5', is_thinking=True, thinking_prefix='<think>\\n', non_thinking_prefix='<think>\\n\\n</think>\\n\\n', history_thinking_prefix='')",
|
| 389 |
+
"_val_dataset_exists": true,
|
| 390 |
+
"hub": "<class 'swift.hub.hub.MSHub'>",
|
| 391 |
+
"evaluation_strategy": "steps",
|
| 392 |
+
"training_args": "Seq2SeqTrainingArguments(output_dir='/NHNHOME/data/sanghyeok/qwen-cua/runs/qwen35_4b_planA_wm_phase1-20260601-041141/v0-20260601-041256', per_device_train_batch_size=1, num_train_epochs=1.0, max_steps=-1, learning_rate=1e-05, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_steps=200.0, optim=<OptimizerNames.ADAMW_TORCH_FUSED: 'adamw_torch_fused'>, optim_args=None, weight_decay=0.0, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, optim_target_modules=None, gradient_accumulation_steps=16, average_tokens_across_devices=None, max_grad_norm=1.0, label_smoothing_factor=0.0, bf16=True, fp16=False, bf16_full_eval=False, fp16_full_eval=False, tf32=True, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, use_liger_kernel=False, liger_kernel_config=None, use_cache=False, neftune_noise_alpha=None, torch_empty_cache_steps=None, auto_find_batch_size=False, logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_steps=10, logging_first_step=True, log_on_each_node=True, logging_nan_inf_filter=True, include_num_input_tokens_seen=None, log_level='passive', log_level_replica='warning', disable_tqdm=False, report_to=['wandb'], run_name='qwen35_4b_planA_wm_phase1-20260601-041141', project='huggingface', trackio_space_id=None, trackio_bucket_id=None, trackio_static_space_id=None, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, eval_steps=500, eval_delay=0, per_device_eval_batch_size=1, prediction_loss_only=False, eval_on_start=False, eval_do_concat_batches=True, eval_use_gather_object=False, eval_accumulation_steps=None, include_for_metrics=[], batch_eval_metrics=False, save_only_model=False, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=100, save_on_each_node=False, save_total_limit=3, enable_jit_checkpoint=False, push_to_hub=False, hub_token=None, hub_private_repo=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_always_push=False, hub_revision=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, restore_callback_states_from_checkpoint=False, full_determinism=False, seed=42, data_seed=42, use_cpu=False, 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, dataloader_drop_last=False, dataloader_num_workers=8, dataloader_pin_memory=True, dataloader_persistent_workers=True, dataloader_prefetch_factor=4, remove_unused_columns=False, label_names=None, train_sampling_strategy='random', length_column_name='length', ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, ddp_static_graph=None, ddp_backend=None, ddp_timeout=18000000, fsdp=[], fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, deepspeed={'bf16': {'enabled': True}, 'fp16': {'enabled': False}, 'zero_optimization': {'stage': 2, 'overlap_comm': True, 'contiguous_gradients': True, 'reduce_bucket_size': 500000000.0, 'allgather_bucket_size': 500000000.0, 'reduce_scatter': True, 'round_robin_gradients': True}, 'gradient_clipping': 'auto', 'gradient_accumulation_steps': 'auto', 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'steps_per_print': 100, 'wall_clock_breakdown': False, '_comment_bf16_safety': 'DeepSpeed 0.19 BF16_Optimizer keeps fp32 master grad partitions (fp32_groups_flat_partition + accumulate_hp_grads_and_remove_lp hook), so gradient accumulation across many micro-batches is done in fp32. bf16 is only used for the param master copy and forward/backward activations. This is safe at grad_accum=16+; no need to force fp32 collectives.'}, debug=[], skip_memory_metrics=True, do_train=False, do_eval=True, do_predict=False, resume_from_checkpoint=None, warmup_ratio=None, logging_dir='/NHNHOME/data/sanghyeok/qwen-cua/runs/qwen35_4b_planA_wm_phase1-20260601-041141/v0-20260601-041256/runs', local_rank=0, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=False, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, safe_serialization=True, max_shard_size='5GB', check_model=True, acc_strategy='token', train_dataloader_shuffle=True, group_by_length=False, max_epochs=None, aligner_lr=None, vit_lr=None, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer=None, loss_type=None, eval_metric=None, callbacks=[], early_stop_interval=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, tuner_type='full', use_galore=False, galore_target_modules=None, galore_rank=128, galore_update_proj_gap=50, galore_scale=1.0, galore_proj_type='std', galore_optim_per_parameter=False, galore_with_embedding=False, galore_quantization=False, galore_proj_quant=False, galore_proj_bits=4, galore_proj_group_size=256, galore_cos_threshold=0.4, galore_gamma_proj=2, galore_queue_size=5, lisa_activated_layers=0, lisa_step_interval=20, use_flash_ckpt=False)"
|
| 393 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"eos_token_id": 248046,
|
| 7 |
+
"hidden_size": 2560,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 2560,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 9216,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": 248044,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": true,
|
| 87 |
+
"use_cache": false,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": true,
|
| 91 |
+
"transformers_version": "5.8.1",
|
| 92 |
+
"use_cache": false,
|
| 93 |
+
"video_token_id": 248057,
|
| 94 |
+
"vision_config": {
|
| 95 |
+
"deepstack_visual_indexes": [],
|
| 96 |
+
"depth": 24,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1024,
|
| 100 |
+
"in_channels": 3,
|
| 101 |
+
"initializer_range": 0.02,
|
| 102 |
+
"intermediate_size": 4096,
|
| 103 |
+
"model_type": "qwen3_5_vision",
|
| 104 |
+
"num_heads": 16,
|
| 105 |
+
"num_position_embeddings": 2304,
|
| 106 |
+
"out_hidden_size": 2560,
|
| 107 |
+
"patch_size": 16,
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.8.1",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step607
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model-00001-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ee2f63ab706ea5c603749884667c1f0a94262ee5d697d3cee15bd4ab0194270
|
| 3 |
+
size 4987577312
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model-00002-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:048b8b30744c19a9fcfeafc7527e9a9a9af73a4a2163da3e5410eea0f721b1f5
|
| 3 |
+
size 4993498424
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model-00003-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9132529e66cac3079a828b70ff720888e735b62dcc8a839f7c42730ef896d42c
|
| 3 |
+
size 398327248
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,740 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 4553957376,
|
| 4 |
+
"total_size": 10379313152
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"_wm_projector.net.0.bias": "model-00001-of-00003.safetensors",
|
| 8 |
+
"_wm_projector.net.0.weight": "model-00001-of-00003.safetensors",
|
| 9 |
+
"_wm_projector.net.2.bias": "model-00001-of-00003.safetensors",
|
| 10 |
+
"_wm_projector.net.2.weight": "model-00001-of-00003.safetensors",
|
| 11 |
+
"_wm_projector.net.4.bias": "model-00001-of-00003.safetensors",
|
| 12 |
+
"_wm_projector.net.4.weight": "model-00001-of-00003.safetensors",
|
| 13 |
+
"_wm_projector.norm.bias": "model-00001-of-00003.safetensors",
|
| 14 |
+
"_wm_projector.norm.weight": "model-00001-of-00003.safetensors",
|
| 15 |
+
"lm_head.weight": "model-00001-of-00003.safetensors",
|
| 16 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00003.safetensors",
|
| 17 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 18 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 19 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 20 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 21 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 22 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 23 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 24 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 25 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 26 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 27 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 28 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 29 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 30 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 31 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 32 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 33 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 34 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 35 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 36 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 37 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 38 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 39 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 40 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 41 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 42 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 43 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 44 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 45 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 46 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 47 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 48 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 49 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 50 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 51 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 52 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 53 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 54 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 55 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 56 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 57 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 58 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 59 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 60 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 61 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 62 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 63 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 64 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 65 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 66 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 67 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 68 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 69 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 70 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 71 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 72 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 73 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 74 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 75 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 76 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 77 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 78 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 79 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 80 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 81 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 82 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 83 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 84 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 85 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 86 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 87 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 88 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 89 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 90 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 91 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 92 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 93 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 94 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 95 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 96 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 97 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 98 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 99 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 100 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 101 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 102 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 103 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 104 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 105 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 106 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 107 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 108 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 109 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 110 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 111 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 112 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 113 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 114 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 115 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 116 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 117 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 118 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 119 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 120 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 121 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 122 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 123 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 124 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 125 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 126 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 127 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 128 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 129 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 130 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 131 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 132 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 133 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 134 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 135 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 136 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 137 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 138 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 139 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 140 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 141 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 142 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 143 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 144 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 145 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 146 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 147 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 148 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 149 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 150 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 151 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 152 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 153 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 154 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 155 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 156 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 157 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 158 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 159 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 160 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 161 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 162 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 163 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 164 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 165 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 166 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 167 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 168 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 169 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 170 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 171 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 172 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 173 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 174 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 175 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 176 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 177 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 178 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 179 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 180 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 181 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 182 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 183 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 184 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 185 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 186 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 187 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 188 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 189 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 190 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 191 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 192 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 193 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 194 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 195 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 196 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 197 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 198 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 199 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 200 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 201 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 202 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 203 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 204 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 205 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 206 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 207 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 208 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 209 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 210 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 211 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 212 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 213 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 214 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 215 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 216 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 217 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 218 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 219 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 220 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 221 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 222 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 223 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 224 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 225 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 226 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 227 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 228 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 229 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 230 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 231 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 232 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 233 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 234 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 235 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 236 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 237 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 238 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 239 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 240 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 241 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 242 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 243 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 244 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 245 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 246 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 247 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 248 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 249 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 250 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 251 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 252 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 253 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 254 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 255 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 256 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 257 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 258 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 259 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 260 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 261 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 262 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 263 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 264 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 265 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 266 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 267 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 268 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 269 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 270 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 271 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 272 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 273 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 274 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 275 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 276 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 277 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 278 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 279 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 280 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 281 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 282 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 283 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 284 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 285 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 286 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 287 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 288 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 289 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 290 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 291 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 292 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 293 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 294 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 295 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 296 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 297 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 298 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 299 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 300 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 301 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 302 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 303 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 304 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 305 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 306 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 307 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 308 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 309 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 310 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 311 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 312 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 313 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 314 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 315 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 316 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 317 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 318 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 319 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 320 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 321 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 322 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 323 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 324 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 325 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 326 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 327 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 328 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 329 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 330 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 331 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 332 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 333 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 334 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 335 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 336 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 337 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 338 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 339 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 340 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 341 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 342 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 343 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 344 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 345 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 346 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 347 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 348 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 349 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 350 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 351 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 352 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 353 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 354 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 355 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 356 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 357 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 358 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 359 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 360 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 361 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 362 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 363 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 364 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 365 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 366 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 367 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 368 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 369 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 370 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 371 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 372 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 373 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 374 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 375 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 376 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 377 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 378 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 379 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 380 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 381 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 382 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 383 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 384 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 385 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 386 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 387 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 388 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 389 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 390 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 391 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 392 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 393 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 394 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 395 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 396 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 397 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 398 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 399 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 400 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 401 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 402 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 403 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 404 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 405 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 406 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 407 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 408 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 409 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 410 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 411 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 412 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 413 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 414 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 415 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 416 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 417 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 418 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 419 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 420 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 421 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 422 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 423 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 424 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 425 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 426 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 427 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 428 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 429 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 430 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 431 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 432 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 433 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 434 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 435 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 436 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 437 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 438 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 439 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 440 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 441 |
+
"model.language_model.norm.weight": "model-00002-of-00003.safetensors",
|
| 442 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 443 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 444 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 445 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 446 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 447 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 448 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 449 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 450 |
+
"model.visual.blocks.0.norm1.bias": "model-00002-of-00003.safetensors",
|
| 451 |
+
"model.visual.blocks.0.norm1.weight": "model-00002-of-00003.safetensors",
|
| 452 |
+
"model.visual.blocks.0.norm2.bias": "model-00002-of-00003.safetensors",
|
| 453 |
+
"model.visual.blocks.0.norm2.weight": "model-00002-of-00003.safetensors",
|
| 454 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 455 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 456 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 457 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 458 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 459 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 460 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 461 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 462 |
+
"model.visual.blocks.1.norm1.bias": "model-00002-of-00003.safetensors",
|
| 463 |
+
"model.visual.blocks.1.norm1.weight": "model-00002-of-00003.safetensors",
|
| 464 |
+
"model.visual.blocks.1.norm2.bias": "model-00002-of-00003.safetensors",
|
| 465 |
+
"model.visual.blocks.1.norm2.weight": "model-00002-of-00003.safetensors",
|
| 466 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 467 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 468 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 469 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 470 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 471 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 472 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 473 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 474 |
+
"model.visual.blocks.10.norm1.bias": "model-00003-of-00003.safetensors",
|
| 475 |
+
"model.visual.blocks.10.norm1.weight": "model-00003-of-00003.safetensors",
|
| 476 |
+
"model.visual.blocks.10.norm2.bias": "model-00003-of-00003.safetensors",
|
| 477 |
+
"model.visual.blocks.10.norm2.weight": "model-00003-of-00003.safetensors",
|
| 478 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 479 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 480 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 481 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 482 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 483 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 484 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 485 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 486 |
+
"model.visual.blocks.11.norm1.bias": "model-00003-of-00003.safetensors",
|
| 487 |
+
"model.visual.blocks.11.norm1.weight": "model-00003-of-00003.safetensors",
|
| 488 |
+
"model.visual.blocks.11.norm2.bias": "model-00003-of-00003.safetensors",
|
| 489 |
+
"model.visual.blocks.11.norm2.weight": "model-00003-of-00003.safetensors",
|
| 490 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 491 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 492 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 493 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 494 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 495 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 496 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 497 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 498 |
+
"model.visual.blocks.12.norm1.bias": "model-00003-of-00003.safetensors",
|
| 499 |
+
"model.visual.blocks.12.norm1.weight": "model-00003-of-00003.safetensors",
|
| 500 |
+
"model.visual.blocks.12.norm2.bias": "model-00003-of-00003.safetensors",
|
| 501 |
+
"model.visual.blocks.12.norm2.weight": "model-00003-of-00003.safetensors",
|
| 502 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 503 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 504 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 505 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 506 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 507 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 508 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 509 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 510 |
+
"model.visual.blocks.13.norm1.bias": "model-00003-of-00003.safetensors",
|
| 511 |
+
"model.visual.blocks.13.norm1.weight": "model-00003-of-00003.safetensors",
|
| 512 |
+
"model.visual.blocks.13.norm2.bias": "model-00003-of-00003.safetensors",
|
| 513 |
+
"model.visual.blocks.13.norm2.weight": "model-00003-of-00003.safetensors",
|
| 514 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 515 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 516 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 517 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 518 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 519 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 520 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 521 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 522 |
+
"model.visual.blocks.14.norm1.bias": "model-00003-of-00003.safetensors",
|
| 523 |
+
"model.visual.blocks.14.norm1.weight": "model-00003-of-00003.safetensors",
|
| 524 |
+
"model.visual.blocks.14.norm2.bias": "model-00003-of-00003.safetensors",
|
| 525 |
+
"model.visual.blocks.14.norm2.weight": "model-00003-of-00003.safetensors",
|
| 526 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 527 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 528 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 529 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 530 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 531 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 532 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 533 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 534 |
+
"model.visual.blocks.15.norm1.bias": "model-00003-of-00003.safetensors",
|
| 535 |
+
"model.visual.blocks.15.norm1.weight": "model-00003-of-00003.safetensors",
|
| 536 |
+
"model.visual.blocks.15.norm2.bias": "model-00003-of-00003.safetensors",
|
| 537 |
+
"model.visual.blocks.15.norm2.weight": "model-00003-of-00003.safetensors",
|
| 538 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 539 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 540 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 541 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 542 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 543 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 544 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 545 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 546 |
+
"model.visual.blocks.16.norm1.bias": "model-00003-of-00003.safetensors",
|
| 547 |
+
"model.visual.blocks.16.norm1.weight": "model-00003-of-00003.safetensors",
|
| 548 |
+
"model.visual.blocks.16.norm2.bias": "model-00003-of-00003.safetensors",
|
| 549 |
+
"model.visual.blocks.16.norm2.weight": "model-00003-of-00003.safetensors",
|
| 550 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 551 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 552 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 553 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 554 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 555 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 556 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 557 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 558 |
+
"model.visual.blocks.17.norm1.bias": "model-00003-of-00003.safetensors",
|
| 559 |
+
"model.visual.blocks.17.norm1.weight": "model-00003-of-00003.safetensors",
|
| 560 |
+
"model.visual.blocks.17.norm2.bias": "model-00003-of-00003.safetensors",
|
| 561 |
+
"model.visual.blocks.17.norm2.weight": "model-00003-of-00003.safetensors",
|
| 562 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 563 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 564 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 565 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 566 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 567 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 568 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 569 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 570 |
+
"model.visual.blocks.18.norm1.bias": "model-00003-of-00003.safetensors",
|
| 571 |
+
"model.visual.blocks.18.norm1.weight": "model-00003-of-00003.safetensors",
|
| 572 |
+
"model.visual.blocks.18.norm2.bias": "model-00003-of-00003.safetensors",
|
| 573 |
+
"model.visual.blocks.18.norm2.weight": "model-00003-of-00003.safetensors",
|
| 574 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 575 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 576 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 577 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 578 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 579 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 580 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 581 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 582 |
+
"model.visual.blocks.19.norm1.bias": "model-00003-of-00003.safetensors",
|
| 583 |
+
"model.visual.blocks.19.norm1.weight": "model-00003-of-00003.safetensors",
|
| 584 |
+
"model.visual.blocks.19.norm2.bias": "model-00003-of-00003.safetensors",
|
| 585 |
+
"model.visual.blocks.19.norm2.weight": "model-00003-of-00003.safetensors",
|
| 586 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 587 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 588 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 589 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 590 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 591 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 592 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 593 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 594 |
+
"model.visual.blocks.2.norm1.bias": "model-00002-of-00003.safetensors",
|
| 595 |
+
"model.visual.blocks.2.norm1.weight": "model-00002-of-00003.safetensors",
|
| 596 |
+
"model.visual.blocks.2.norm2.bias": "model-00002-of-00003.safetensors",
|
| 597 |
+
"model.visual.blocks.2.norm2.weight": "model-00002-of-00003.safetensors",
|
| 598 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 599 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 600 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 601 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 602 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 603 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 604 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 605 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 606 |
+
"model.visual.blocks.20.norm1.bias": "model-00003-of-00003.safetensors",
|
| 607 |
+
"model.visual.blocks.20.norm1.weight": "model-00003-of-00003.safetensors",
|
| 608 |
+
"model.visual.blocks.20.norm2.bias": "model-00003-of-00003.safetensors",
|
| 609 |
+
"model.visual.blocks.20.norm2.weight": "model-00003-of-00003.safetensors",
|
| 610 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 611 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 612 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 613 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 614 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 615 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 616 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 617 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 618 |
+
"model.visual.blocks.21.norm1.bias": "model-00003-of-00003.safetensors",
|
| 619 |
+
"model.visual.blocks.21.norm1.weight": "model-00003-of-00003.safetensors",
|
| 620 |
+
"model.visual.blocks.21.norm2.bias": "model-00003-of-00003.safetensors",
|
| 621 |
+
"model.visual.blocks.21.norm2.weight": "model-00003-of-00003.safetensors",
|
| 622 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 623 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 624 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 625 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 626 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 627 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 628 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 629 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 630 |
+
"model.visual.blocks.22.norm1.bias": "model-00003-of-00003.safetensors",
|
| 631 |
+
"model.visual.blocks.22.norm1.weight": "model-00003-of-00003.safetensors",
|
| 632 |
+
"model.visual.blocks.22.norm2.bias": "model-00003-of-00003.safetensors",
|
| 633 |
+
"model.visual.blocks.22.norm2.weight": "model-00003-of-00003.safetensors",
|
| 634 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 635 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 636 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 637 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 638 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 639 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 640 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 641 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 642 |
+
"model.visual.blocks.23.norm1.bias": "model-00003-of-00003.safetensors",
|
| 643 |
+
"model.visual.blocks.23.norm1.weight": "model-00003-of-00003.safetensors",
|
| 644 |
+
"model.visual.blocks.23.norm2.bias": "model-00003-of-00003.safetensors",
|
| 645 |
+
"model.visual.blocks.23.norm2.weight": "model-00003-of-00003.safetensors",
|
| 646 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 647 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 648 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 649 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 650 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 651 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 652 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 653 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 654 |
+
"model.visual.blocks.3.norm1.bias": "model-00002-of-00003.safetensors",
|
| 655 |
+
"model.visual.blocks.3.norm1.weight": "model-00002-of-00003.safetensors",
|
| 656 |
+
"model.visual.blocks.3.norm2.bias": "model-00002-of-00003.safetensors",
|
| 657 |
+
"model.visual.blocks.3.norm2.weight": "model-00002-of-00003.safetensors",
|
| 658 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 659 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 660 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 661 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 662 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 663 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 664 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 665 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 666 |
+
"model.visual.blocks.4.norm1.bias": "model-00002-of-00003.safetensors",
|
| 667 |
+
"model.visual.blocks.4.norm1.weight": "model-00002-of-00003.safetensors",
|
| 668 |
+
"model.visual.blocks.4.norm2.bias": "model-00002-of-00003.safetensors",
|
| 669 |
+
"model.visual.blocks.4.norm2.weight": "model-00002-of-00003.safetensors",
|
| 670 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 671 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 672 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 673 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 674 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 675 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 676 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 677 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 678 |
+
"model.visual.blocks.5.norm1.bias": "model-00002-of-00003.safetensors",
|
| 679 |
+
"model.visual.blocks.5.norm1.weight": "model-00002-of-00003.safetensors",
|
| 680 |
+
"model.visual.blocks.5.norm2.bias": "model-00002-of-00003.safetensors",
|
| 681 |
+
"model.visual.blocks.5.norm2.weight": "model-00002-of-00003.safetensors",
|
| 682 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 683 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 684 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 685 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 686 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 687 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 688 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 689 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 690 |
+
"model.visual.blocks.6.norm1.bias": "model-00002-of-00003.safetensors",
|
| 691 |
+
"model.visual.blocks.6.norm1.weight": "model-00002-of-00003.safetensors",
|
| 692 |
+
"model.visual.blocks.6.norm2.bias": "model-00002-of-00003.safetensors",
|
| 693 |
+
"model.visual.blocks.6.norm2.weight": "model-00002-of-00003.safetensors",
|
| 694 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 695 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 696 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 697 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 698 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 699 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 700 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 701 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 702 |
+
"model.visual.blocks.7.norm1.bias": "model-00002-of-00003.safetensors",
|
| 703 |
+
"model.visual.blocks.7.norm1.weight": "model-00002-of-00003.safetensors",
|
| 704 |
+
"model.visual.blocks.7.norm2.bias": "model-00002-of-00003.safetensors",
|
| 705 |
+
"model.visual.blocks.7.norm2.weight": "model-00002-of-00003.safetensors",
|
| 706 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 707 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 708 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 709 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 710 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 711 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 712 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 713 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 714 |
+
"model.visual.blocks.8.norm1.bias": "model-00002-of-00003.safetensors",
|
| 715 |
+
"model.visual.blocks.8.norm1.weight": "model-00002-of-00003.safetensors",
|
| 716 |
+
"model.visual.blocks.8.norm2.bias": "model-00002-of-00003.safetensors",
|
| 717 |
+
"model.visual.blocks.8.norm2.weight": "model-00002-of-00003.safetensors",
|
| 718 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00002-of-00003.safetensors",
|
| 719 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00002-of-00003.safetensors",
|
| 720 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00002-of-00003.safetensors",
|
| 721 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00002-of-00003.safetensors",
|
| 722 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00002-of-00003.safetensors",
|
| 723 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00002-of-00003.safetensors",
|
| 724 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00002-of-00003.safetensors",
|
| 725 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00002-of-00003.safetensors",
|
| 726 |
+
"model.visual.blocks.9.norm1.bias": "model-00002-of-00003.safetensors",
|
| 727 |
+
"model.visual.blocks.9.norm1.weight": "model-00002-of-00003.safetensors",
|
| 728 |
+
"model.visual.blocks.9.norm2.bias": "model-00002-of-00003.safetensors",
|
| 729 |
+
"model.visual.blocks.9.norm2.weight": "model-00002-of-00003.safetensors",
|
| 730 |
+
"model.visual.merger.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 731 |
+
"model.visual.merger.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 732 |
+
"model.visual.merger.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 733 |
+
"model.visual.merger.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 734 |
+
"model.visual.merger.norm.bias": "model-00003-of-00003.safetensors",
|
| 735 |
+
"model.visual.merger.norm.weight": "model-00003-of-00003.safetensors",
|
| 736 |
+
"model.visual.patch_embed.proj.bias": "model-00003-of-00003.safetensors",
|
| 737 |
+
"model.visual.patch_embed.proj.weight": "model-00003-of-00003.safetensors",
|
| 738 |
+
"model.visual.pos_embed.weight": "model-00003-of-00003.safetensors"
|
| 739 |
+
}
|
| 740 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/processor_config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"merge_size": 2,
|
| 48 |
+
"min_frames": 4,
|
| 49 |
+
"patch_size": 16,
|
| 50 |
+
"resample": 3,
|
| 51 |
+
"rescale_factor": 0.00392156862745098,
|
| 52 |
+
"return_metadata": false,
|
| 53 |
+
"size": {
|
| 54 |
+
"longest_edge": 25165824,
|
| 55 |
+
"shortest_edge": 4096
|
| 56 |
+
},
|
| 57 |
+
"temporal_patch_size": 2,
|
| 58 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 59 |
+
}
|
| 60 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
|
| 3 |
+
size 19989325
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": false,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/trainer_state.json
ADDED
|
@@ -0,0 +1,666 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 607,
|
| 3 |
+
"best_metric": 0.01024034,
|
| 4 |
+
"best_model_checkpoint": "/NHNHOME/data/sanghyeok/qwen-cua/runs/qwen35_4b_planA_wm_phase1-20260601-041141/v0-20260601-041256/checkpoint-607",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 607,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0016474464579901153,
|
| 14 |
+
"grad_norm": 67.77246856689453,
|
| 15 |
+
"learning_rate": 5.0000000000000004e-08,
|
| 16 |
+
"loss": 12.97900390625,
|
| 17 |
+
"loss_sft": 0.12126511335372925,
|
| 18 |
+
"loss_wm": 1.0078125,
|
| 19 |
+
"step": 1,
|
| 20 |
+
"token_acc": 0.7420708478136671
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.016474464579901153,
|
| 24 |
+
"grad_norm": 66.71990203857422,
|
| 25 |
+
"learning_rate": 5.000000000000001e-07,
|
| 26 |
+
"loss": 13.211805555555555,
|
| 27 |
+
"loss_sft": 0.020476022735238075,
|
| 28 |
+
"loss_wm": 0.9921875,
|
| 29 |
+
"step": 10,
|
| 30 |
+
"token_acc": 0.7415035375110337
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.032948929159802305,
|
| 34 |
+
"grad_norm": 66.3600082397461,
|
| 35 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 36 |
+
"loss": 12.337841796875,
|
| 37 |
+
"loss_sft": 0.03363444656133652,
|
| 38 |
+
"loss_wm": 0.82421875,
|
| 39 |
+
"step": 20,
|
| 40 |
+
"token_acc": 0.7436525508739965
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.04942339373970346,
|
| 44 |
+
"grad_norm": 31.513206481933594,
|
| 45 |
+
"learning_rate": 1.5e-06,
|
| 46 |
+
"loss": 9.913916015625,
|
| 47 |
+
"loss_sft": 0.1270032525062561,
|
| 48 |
+
"loss_wm": 0.65625,
|
| 49 |
+
"step": 30,
|
| 50 |
+
"token_acc": 0.7449547496756478
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.06589785831960461,
|
| 54 |
+
"grad_norm": 21.876819610595703,
|
| 55 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 56 |
+
"loss": 7.912451171875,
|
| 57 |
+
"loss_sft": 0.0386272557079792,
|
| 58 |
+
"loss_wm": 0.5546875,
|
| 59 |
+
"step": 40,
|
| 60 |
+
"token_acc": 0.7423125710842913
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.08237232289950576,
|
| 64 |
+
"grad_norm": 23.65790557861328,
|
| 65 |
+
"learning_rate": 2.5e-06,
|
| 66 |
+
"loss": 6.934912109375,
|
| 67 |
+
"loss_sft": 0.07173172384500504,
|
| 68 |
+
"loss_wm": 0.486328125,
|
| 69 |
+
"step": 50,
|
| 70 |
+
"token_acc": 0.7368241322538603
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.09884678747940692,
|
| 74 |
+
"grad_norm": 37.9365119934082,
|
| 75 |
+
"learning_rate": 3e-06,
|
| 76 |
+
"loss": 6.0048828125,
|
| 77 |
+
"loss_sft": 0.015908721834421158,
|
| 78 |
+
"loss_wm": 0.0,
|
| 79 |
+
"step": 60,
|
| 80 |
+
"token_acc": 0.7322204102696694
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11532125205930807,
|
| 84 |
+
"grad_norm": 25.693683624267578,
|
| 85 |
+
"learning_rate": 3.5e-06,
|
| 86 |
+
"loss": 5.1912353515625,
|
| 87 |
+
"loss_sft": 0.1040407046675682,
|
| 88 |
+
"loss_wm": 0.34765625,
|
| 89 |
+
"step": 70,
|
| 90 |
+
"token_acc": 0.7288802605577288
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.13179571663920922,
|
| 94 |
+
"grad_norm": 42.76861572265625,
|
| 95 |
+
"learning_rate": 4.000000000000001e-06,
|
| 96 |
+
"loss": 4.3314697265625,
|
| 97 |
+
"loss_sft": 0.038324933499097824,
|
| 98 |
+
"loss_wm": 0.302734375,
|
| 99 |
+
"step": 80,
|
| 100 |
+
"token_acc": 0.7251751001847648
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14827018121911037,
|
| 104 |
+
"grad_norm": 42.2775764465332,
|
| 105 |
+
"learning_rate": 4.5e-06,
|
| 106 |
+
"loss": 3.6370849609375,
|
| 107 |
+
"loss_sft": 0.04018238186836243,
|
| 108 |
+
"loss_wm": 0.251953125,
|
| 109 |
+
"step": 90,
|
| 110 |
+
"token_acc": 0.7257638857187575
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.16474464579901152,
|
| 114 |
+
"grad_norm": 48.08127975463867,
|
| 115 |
+
"learning_rate": 5e-06,
|
| 116 |
+
"loss": 2.82974853515625,
|
| 117 |
+
"loss_sft": 0.11553160846233368,
|
| 118 |
+
"loss_wm": 0.1806640625,
|
| 119 |
+
"step": 100,
|
| 120 |
+
"token_acc": 0.7243545236316892
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.1812191103789127,
|
| 124 |
+
"grad_norm": 44.53425979614258,
|
| 125 |
+
"learning_rate": 5.500000000000001e-06,
|
| 126 |
+
"loss": 2.14354248046875,
|
| 127 |
+
"loss_sft": 0.04376419261097908,
|
| 128 |
+
"loss_wm": 0.146484375,
|
| 129 |
+
"step": 110,
|
| 130 |
+
"token_acc": 0.7223867851792997
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.19769357495881384,
|
| 134 |
+
"grad_norm": 61.72502899169922,
|
| 135 |
+
"learning_rate": 6e-06,
|
| 136 |
+
"loss": 1.587005615234375,
|
| 137 |
+
"loss_sft": 0.07198437303304672,
|
| 138 |
+
"loss_wm": 0.0966796875,
|
| 139 |
+
"step": 120,
|
| 140 |
+
"token_acc": 0.7242483014090744
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.214168039538715,
|
| 144 |
+
"grad_norm": 55.525936126708984,
|
| 145 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 146 |
+
"loss": 1.11756591796875,
|
| 147 |
+
"loss_sft": 0.02939743734896183,
|
| 148 |
+
"loss_wm": 0.0,
|
| 149 |
+
"step": 130,
|
| 150 |
+
"token_acc": 0.7226163937980855
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.23064250411861614,
|
| 154 |
+
"grad_norm": 59.483978271484375,
|
| 155 |
+
"learning_rate": 7e-06,
|
| 156 |
+
"loss": 0.781060791015625,
|
| 157 |
+
"loss_sft": 0.04185626655817032,
|
| 158 |
+
"loss_wm": 0.049560546875,
|
| 159 |
+
"step": 140,
|
| 160 |
+
"token_acc": 0.7218057264614721
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.2471169686985173,
|
| 164 |
+
"grad_norm": 58.55827713012695,
|
| 165 |
+
"learning_rate": 7.500000000000001e-06,
|
| 166 |
+
"loss": 0.5791427612304687,
|
| 167 |
+
"loss_sft": 0.028999704867601395,
|
| 168 |
+
"loss_wm": 0.0,
|
| 169 |
+
"step": 150,
|
| 170 |
+
"token_acc": 0.7206720302136304
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.26359143327841844,
|
| 174 |
+
"grad_norm": 51.63848876953125,
|
| 175 |
+
"learning_rate": 8.000000000000001e-06,
|
| 176 |
+
"loss": 0.502850341796875,
|
| 177 |
+
"loss_sft": 0.027779903262853622,
|
| 178 |
+
"loss_wm": 0.032958984375,
|
| 179 |
+
"step": 160,
|
| 180 |
+
"token_acc": 0.7199326074701833
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.2800658978583196,
|
| 184 |
+
"grad_norm": 43.319480895996094,
|
| 185 |
+
"learning_rate": 8.5e-06,
|
| 186 |
+
"loss": 0.4743316650390625,
|
| 187 |
+
"loss_sft": 0.039876826107501984,
|
| 188 |
+
"loss_wm": 0.0390625,
|
| 189 |
+
"step": 170,
|
| 190 |
+
"token_acc": 0.7185668767479726
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.29654036243822074,
|
| 194 |
+
"grad_norm": 31.008390426635742,
|
| 195 |
+
"learning_rate": 9e-06,
|
| 196 |
+
"loss": 0.44373016357421874,
|
| 197 |
+
"loss_sft": 0.03223320096731186,
|
| 198 |
+
"loss_wm": 0.0,
|
| 199 |
+
"step": 180,
|
| 200 |
+
"token_acc": 0.7202119434119547
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.3130148270181219,
|
| 204 |
+
"grad_norm": 36.54234313964844,
|
| 205 |
+
"learning_rate": 9.5e-06,
|
| 206 |
+
"loss": 0.4494415283203125,
|
| 207 |
+
"loss_sft": 0.04214436188340187,
|
| 208 |
+
"loss_wm": 0.0296630859375,
|
| 209 |
+
"step": 190,
|
| 210 |
+
"token_acc": 0.7208116386521944
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.32948929159802304,
|
| 214 |
+
"grad_norm": 39.17905807495117,
|
| 215 |
+
"learning_rate": 1e-05,
|
| 216 |
+
"loss": 0.39757232666015624,
|
| 217 |
+
"loss_sft": 0.06208985298871994,
|
| 218 |
+
"loss_wm": 0.046142578125,
|
| 219 |
+
"step": 200,
|
| 220 |
+
"token_acc": 0.7207964079784009
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.34596375617792424,
|
| 224 |
+
"grad_norm": 27.340139389038086,
|
| 225 |
+
"learning_rate": 9.98511203659851e-06,
|
| 226 |
+
"loss": 0.39337158203125,
|
| 227 |
+
"loss_sft": 0.04444035142660141,
|
| 228 |
+
"loss_wm": 0.02490234375,
|
| 229 |
+
"step": 210,
|
| 230 |
+
"token_acc": 0.7187266104609756
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.3624382207578254,
|
| 234 |
+
"grad_norm": 22.59505844116211,
|
| 235 |
+
"learning_rate": 9.940536806975732e-06,
|
| 236 |
+
"loss": 0.3745513916015625,
|
| 237 |
+
"loss_sft": 0.14363454282283783,
|
| 238 |
+
"loss_wm": 0.0286865234375,
|
| 239 |
+
"step": 220,
|
| 240 |
+
"token_acc": 0.7169596409249033
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.37891268533772654,
|
| 244 |
+
"grad_norm": 27.243209838867188,
|
| 245 |
+
"learning_rate": 9.866539764886562e-06,
|
| 246 |
+
"loss": 0.35409088134765626,
|
| 247 |
+
"loss_sft": 0.05909347161650658,
|
| 248 |
+
"loss_wm": 0.032470703125,
|
| 249 |
+
"step": 230,
|
| 250 |
+
"token_acc": 0.7153547497154839
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.3953871499176277,
|
| 254 |
+
"grad_norm": 17.35493278503418,
|
| 255 |
+
"learning_rate": 9.763561576432781e-06,
|
| 256 |
+
"loss": 0.35498809814453125,
|
| 257 |
+
"loss_sft": 0.041506327688694,
|
| 258 |
+
"loss_wm": 0.02490234375,
|
| 259 |
+
"step": 240,
|
| 260 |
+
"token_acc": 0.7132838779769349
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.41186161449752884,
|
| 264 |
+
"grad_norm": 15.78779125213623,
|
| 265 |
+
"learning_rate": 9.632215495814724e-06,
|
| 266 |
+
"loss": 0.34316864013671877,
|
| 267 |
+
"loss_sft": 0.08028920739889145,
|
| 268 |
+
"loss_wm": 0.018798828125,
|
| 269 |
+
"step": 250,
|
| 270 |
+
"token_acc": 0.7074259600351923
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.42833607907743,
|
| 274 |
+
"grad_norm": 19.711910247802734,
|
| 275 |
+
"learning_rate": 9.473283713288862e-06,
|
| 276 |
+
"loss": 0.34627761840820315,
|
| 277 |
+
"loss_sft": 0.053114842623472214,
|
| 278 |
+
"loss_wm": 0.0247802734375,
|
| 279 |
+
"step": 260,
|
| 280 |
+
"token_acc": 0.7048950696646151
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.44481054365733114,
|
| 284 |
+
"grad_norm": 20.045223236083984,
|
| 285 |
+
"learning_rate": 9.287712697079827e-06,
|
| 286 |
+
"loss": 0.33700103759765626,
|
| 287 |
+
"loss_sft": 0.1380854696035385,
|
| 288 |
+
"loss_wm": 0.02685546875,
|
| 289 |
+
"step": 270,
|
| 290 |
+
"token_acc": 0.702730003801209
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.4612850082372323,
|
| 294 |
+
"grad_norm": 13.392139434814453,
|
| 295 |
+
"learning_rate": 9.076607556986699e-06,
|
| 296 |
+
"loss": 0.31904449462890627,
|
| 297 |
+
"loss_sft": 0.02688782848417759,
|
| 298 |
+
"loss_wm": 0.0,
|
| 299 |
+
"step": 280,
|
| 300 |
+
"token_acc": 0.6983074722450135
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.47775947281713343,
|
| 304 |
+
"grad_norm": 13.579081535339355,
|
| 305 |
+
"learning_rate": 8.841225463249305e-06,
|
| 306 |
+
"loss": 0.3242401123046875,
|
| 307 |
+
"loss_sft": 0.04501771926879883,
|
| 308 |
+
"loss_wm": 0.0220947265625,
|
| 309 |
+
"step": 290,
|
| 310 |
+
"token_acc": 0.6937485803207011
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4942339373970346,
|
| 314 |
+
"grad_norm": 12.500804901123047,
|
| 315 |
+
"learning_rate": 8.582968159866416e-06,
|
| 316 |
+
"loss": 0.31450653076171875,
|
| 317 |
+
"loss_sft": 0.028333961963653564,
|
| 318 |
+
"loss_wm": 0.0,
|
| 319 |
+
"step": 300,
|
| 320 |
+
"token_acc": 0.6939164096431044
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.5107084019769358,
|
| 324 |
+
"grad_norm": 8.997215270996094,
|
| 325 |
+
"learning_rate": 8.303373616950408e-06,
|
| 326 |
+
"loss": 0.29952468872070315,
|
| 327 |
+
"loss_sft": 0.16261766850948334,
|
| 328 |
+
"loss_wm": 0.021728515625,
|
| 329 |
+
"step": 310,
|
| 330 |
+
"token_acc": 0.6891285248747049
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.5271828665568369,
|
| 334 |
+
"grad_norm": 10.469232559204102,
|
| 335 |
+
"learning_rate": 8.004106871830155e-06,
|
| 336 |
+
"loss": 0.29231109619140627,
|
| 337 |
+
"loss_sft": 0.028151841834187508,
|
| 338 |
+
"loss_wm": 0.018798828125,
|
| 339 |
+
"step": 320,
|
| 340 |
+
"token_acc": 0.6860036474213919
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.5436573311367381,
|
| 344 |
+
"grad_norm": 8.854934692382812,
|
| 345 |
+
"learning_rate": 7.68695011344511e-06,
|
| 346 |
+
"loss": 0.28349761962890624,
|
| 347 |
+
"loss_sft": 0.045533645898103714,
|
| 348 |
+
"loss_wm": 0.0230712890625,
|
| 349 |
+
"step": 330,
|
| 350 |
+
"token_acc": 0.6843208429190192
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.5601317957166392,
|
| 354 |
+
"grad_norm": 10.360681533813477,
|
| 355 |
+
"learning_rate": 7.353792069079826e-06,
|
| 356 |
+
"loss": 0.27960891723632814,
|
| 357 |
+
"loss_sft": 0.15268206596374512,
|
| 358 |
+
"loss_wm": 0.01806640625,
|
| 359 |
+
"step": 340,
|
| 360 |
+
"token_acc": 0.6821943929731927
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5766062602965404,
|
| 364 |
+
"grad_norm": 11.015717506408691,
|
| 365 |
+
"learning_rate": 7.006616756642867e-06,
|
| 366 |
+
"loss": 0.26487884521484373,
|
| 367 |
+
"loss_sft": 0.06506884098052979,
|
| 368 |
+
"loss_wm": 0.0,
|
| 369 |
+
"step": 350,
|
| 370 |
+
"token_acc": 0.6778631514224018
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.5930807248764415,
|
| 374 |
+
"grad_norm": 9.47918701171875,
|
| 375 |
+
"learning_rate": 6.647491669472421e-06,
|
| 376 |
+
"loss": 0.26204757690429686,
|
| 377 |
+
"loss_sft": 0.04625123366713524,
|
| 378 |
+
"loss_wm": 0.0206298828125,
|
| 379 |
+
"step": 360,
|
| 380 |
+
"token_acc": 0.6772995701376829
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6095551894563427,
|
| 384 |
+
"grad_norm": 7.8423638343811035,
|
| 385 |
+
"learning_rate": 6.278555464030228e-06,
|
| 386 |
+
"loss": 0.2632942199707031,
|
| 387 |
+
"loss_sft": 0.044745832681655884,
|
| 388 |
+
"loss_wm": 0.01708984375,
|
| 389 |
+
"step": 370,
|
| 390 |
+
"token_acc": 0.6763580278517305
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.6260296540362438,
|
| 394 |
+
"grad_norm": 8.983335494995117,
|
| 395 |
+
"learning_rate": 5.902005223805931e-06,
|
| 396 |
+
"loss": 0.248980712890625,
|
| 397 |
+
"loss_sft": 0.048628680408000946,
|
| 398 |
+
"loss_wm": 0.01806640625,
|
| 399 |
+
"step": 380,
|
| 400 |
+
"token_acc": 0.6740036652563094
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.642504118616145,
|
| 404 |
+
"grad_norm": 7.505780220031738,
|
| 405 |
+
"learning_rate": 5.520083375277644e-06,
|
| 406 |
+
"loss": 0.24360427856445313,
|
| 407 |
+
"loss_sft": 0.14316882193088531,
|
| 408 |
+
"loss_wm": 0.01318359375,
|
| 409 |
+
"step": 390,
|
| 410 |
+
"token_acc": 0.671775135798147
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.6589785831960461,
|
| 414 |
+
"grad_norm": 6.753406047821045,
|
| 415 |
+
"learning_rate": 5.135064333846612e-06,
|
| 416 |
+
"loss": 0.2411224365234375,
|
| 417 |
+
"loss_sft": 0.08399983495473862,
|
| 418 |
+
"loss_wm": 0.02587890625,
|
| 419 |
+
"step": 400,
|
| 420 |
+
"token_acc": 0.6710974688492609
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.6754530477759473,
|
| 424 |
+
"grad_norm": 5.655179977416992,
|
| 425 |
+
"learning_rate": 4.749240959271918e-06,
|
| 426 |
+
"loss": 0.23962631225585937,
|
| 427 |
+
"loss_sft": 0.07269106805324554,
|
| 428 |
+
"loss_wm": 0.0220947265625,
|
| 429 |
+
"step": 410,
|
| 430 |
+
"token_acc": 0.6694721741395608
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.6919275123558485,
|
| 434 |
+
"grad_norm": 4.641175746917725,
|
| 435 |
+
"learning_rate": 4.364910901265607e-06,
|
| 436 |
+
"loss": 0.22925224304199218,
|
| 437 |
+
"loss_sft": 0.07298626750707626,
|
| 438 |
+
"loss_wm": 0.0159912109375,
|
| 439 |
+
"step": 420,
|
| 440 |
+
"token_acc": 0.6646786023889786
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.7084019769357496,
|
| 444 |
+
"grad_norm": 4.991263389587402,
|
| 445 |
+
"learning_rate": 3.984362916562753e-06,
|
| 446 |
+
"loss": 0.22557373046875,
|
| 447 |
+
"loss_sft": 0.049899011850357056,
|
| 448 |
+
"loss_wm": 0.0,
|
| 449 |
+
"step": 430,
|
| 450 |
+
"token_acc": 0.6628667790221485
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.7248764415156508,
|
| 454 |
+
"grad_norm": 4.784546375274658,
|
| 455 |
+
"learning_rate": 3.6098632389508637e-06,
|
| 456 |
+
"loss": 0.2221851348876953,
|
| 457 |
+
"loss_sft": 0.025372540578246117,
|
| 458 |
+
"loss_wm": 0.0107421875,
|
| 459 |
+
"step": 440,
|
| 460 |
+
"token_acc": 0.6636077319878697
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.7413509060955519,
|
| 464 |
+
"grad_norm": 3.8330953121185303,
|
| 465 |
+
"learning_rate": 3.2436420834276013e-06,
|
| 466 |
+
"loss": 0.22017364501953124,
|
| 467 |
+
"loss_sft": 0.0922170877456665,
|
| 468 |
+
"loss_wm": 0.0140380859375,
|
| 469 |
+
"step": 450,
|
| 470 |
+
"token_acc": 0.660493189775746
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.7578253706754531,
|
| 474 |
+
"grad_norm": 3.9884116649627686,
|
| 475 |
+
"learning_rate": 2.8878803648570773e-06,
|
| 476 |
+
"loss": 0.21807899475097656,
|
| 477 |
+
"loss_sft": 0.06469698995351791,
|
| 478 |
+
"loss_wm": 0.01300048828125,
|
| 479 |
+
"step": 460,
|
| 480 |
+
"token_acc": 0.6596398709867601
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.7742998352553542,
|
| 484 |
+
"grad_norm": 3.1394307613372803,
|
| 485 |
+
"learning_rate": 2.544696710217588e-06,
|
| 486 |
+
"loss": 0.2091053009033203,
|
| 487 |
+
"loss_sft": 0.05961334705352783,
|
| 488 |
+
"loss_wm": 0.0,
|
| 489 |
+
"step": 470,
|
| 490 |
+
"token_acc": 0.6546295784853023
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.7907742998352554,
|
| 494 |
+
"grad_norm": 2.9814352989196777,
|
| 495 |
+
"learning_rate": 2.2161348417852346e-06,
|
| 496 |
+
"loss": 0.20496063232421874,
|
| 497 |
+
"loss_sft": 0.18844947218894958,
|
| 498 |
+
"loss_wm": 0.0166015625,
|
| 499 |
+
"step": 480,
|
| 500 |
+
"token_acc": 0.6561429387133613
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.8072487644151565,
|
| 504 |
+
"grad_norm": 2.502312183380127,
|
| 505 |
+
"learning_rate": 1.9041514063889571e-06,
|
| 506 |
+
"loss": 0.20882492065429686,
|
| 507 |
+
"loss_sft": 0.056482382118701935,
|
| 508 |
+
"loss_wm": 0.0181884765625,
|
| 509 |
+
"step": 490,
|
| 510 |
+
"token_acc": 0.6562061824656142
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.8237232289950577,
|
| 514 |
+
"grad_norm": 2.4276952743530273,
|
| 515 |
+
"learning_rate": 1.6106043232159745e-06,
|
| 516 |
+
"loss": 0.2047943115234375,
|
| 517 |
+
"loss_sft": 0.09242410957813263,
|
| 518 |
+
"loss_wm": 0.00909423828125,
|
| 519 |
+
"step": 500,
|
| 520 |
+
"token_acc": 0.6539098745876972
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.8237232289950577,
|
| 524 |
+
"eval_loss": 0.01082339696586132,
|
| 525 |
+
"eval_runtime": 83.8749,
|
| 526 |
+
"eval_samples_per_second": 4.96,
|
| 527 |
+
"eval_steps_per_second": 0.62,
|
| 528 |
+
"eval_token_acc": 0.6893272922535332,
|
| 529 |
+
"loss_sft": 1.3956140279769897,
|
| 530 |
+
"loss_wm": 0.0133056640625,
|
| 531 |
+
"step": 500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 0.8401976935749588,
|
| 535 |
+
"grad_norm": 2.082486867904663,
|
| 536 |
+
"learning_rate": 1.337241719558648e-06,
|
| 537 |
+
"loss": 0.19888381958007811,
|
| 538 |
+
"loss_sft": 0.18075883388519287,
|
| 539 |
+
"loss_wm": 0.0169677734375,
|
| 540 |
+
"step": 510,
|
| 541 |
+
"token_acc": 0.6512219449659321
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.85667215815486,
|
| 545 |
+
"grad_norm": 1.9151747226715088,
|
| 546 |
+
"learning_rate": 1.0856915203924096e-06,
|
| 547 |
+
"loss": 0.20433616638183594,
|
| 548 |
+
"loss_sft": 0.07122018933296204,
|
| 549 |
+
"loss_wm": 0.0185546875,
|
| 550 |
+
"step": 520,
|
| 551 |
+
"token_acc": 0.6526840483886955
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.8731466227347611,
|
| 555 |
+
"grad_norm": 1.5124331712722778,
|
| 556 |
+
"learning_rate": 8.574517537807897e-07,
|
| 557 |
+
"loss": 0.19777755737304686,
|
| 558 |
+
"loss_sft": 0.055201489478349686,
|
| 559 |
+
"loss_wm": 0.0159912109375,
|
| 560 |
+
"step": 530,
|
| 561 |
+
"token_acc": 0.6494442326212959
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.8896210873146623,
|
| 565 |
+
"grad_norm": 1.307037353515625,
|
| 566 |
+
"learning_rate": 6.538816298406203e-07,
|
| 567 |
+
"loss": 0.19518165588378905,
|
| 568 |
+
"loss_sft": 0.07271558046340942,
|
| 569 |
+
"loss_wm": 0.0,
|
| 570 |
+
"step": 540,
|
| 571 |
+
"token_acc": 0.6501236063288529
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.9060955518945635,
|
| 575 |
+
"grad_norm": 1.517959475517273,
|
| 576 |
+
"learning_rate": 4.7619344639384447e-07,
|
| 577 |
+
"loss": 0.19379692077636718,
|
| 578 |
+
"loss_sft": 0.08473915606737137,
|
| 579 |
+
"loss_wm": 0.009765625,
|
| 580 |
+
"step": 550,
|
| 581 |
+
"token_acc": 0.6503523424659791
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.9225700164744646,
|
| 585 |
+
"grad_norm": 1.1579397916793823,
|
| 586 |
+
"learning_rate": 3.254453695092752e-07,
|
| 587 |
+
"loss": 0.19301300048828124,
|
| 588 |
+
"loss_sft": 0.07396426796913147,
|
| 589 |
+
"loss_wm": 0.0162353515625,
|
| 590 |
+
"step": 560,
|
| 591 |
+
"token_acc": 0.6483781498384551
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.9390444810543658,
|
| 595 |
+
"grad_norm": 1.0250858068466187,
|
| 596 |
+
"learning_rate": 2.0253513192751374e-07,
|
| 597 |
+
"loss": 0.19713287353515624,
|
| 598 |
+
"loss_sft": 0.06145636364817619,
|
| 599 |
+
"loss_wm": 0.0,
|
| 600 |
+
"step": 570,
|
| 601 |
+
"token_acc": 0.6508906181283831
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.9555189456342669,
|
| 605 |
+
"grad_norm": 0.9135099053382874,
|
| 606 |
+
"learning_rate": 1.0819468689607426e-07,
|
| 607 |
+
"loss": 0.1919933319091797,
|
| 608 |
+
"loss_sft": 0.05189112201333046,
|
| 609 |
+
"loss_wm": 0.01348876953125,
|
| 610 |
+
"step": 580,
|
| 611 |
+
"token_acc": 0.6479599766135538
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.9719934102141681,
|
| 615 |
+
"grad_norm": 0.6780261397361755,
|
| 616 |
+
"learning_rate": 4.298584925212068e-08,
|
| 617 |
+
"loss": 0.1974506378173828,
|
| 618 |
+
"loss_sft": 0.035562317818403244,
|
| 619 |
+
"loss_wm": 0.0,
|
| 620 |
+
"step": 590,
|
| 621 |
+
"token_acc": 0.6500147874279253
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.9884678747940692,
|
| 625 |
+
"grad_norm": 0.4821934700012207,
|
| 626 |
+
"learning_rate": 7.2969497109715016e-09,
|
| 627 |
+
"loss": 0.1888751983642578,
|
| 628 |
+
"loss_sft": 0.17050720751285553,
|
| 629 |
+
"loss_wm": 0.013671875,
|
| 630 |
+
"step": 600,
|
| 631 |
+
"token_acc": 0.6502708682238647
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.0,
|
| 635 |
+
"eval_loss": 0.010240335017442703,
|
| 636 |
+
"eval_runtime": 69.1743,
|
| 637 |
+
"eval_samples_per_second": 6.014,
|
| 638 |
+
"eval_steps_per_second": 0.752,
|
| 639 |
+
"eval_token_acc": 0.6850229371176432,
|
| 640 |
+
"loss_sft": 1.432279109954834,
|
| 641 |
+
"loss_wm": 0.01251220703125,
|
| 642 |
+
"step": 607
|
| 643 |
+
}
|
| 644 |
+
],
|
| 645 |
+
"logging_steps": 10,
|
| 646 |
+
"max_steps": 607,
|
| 647 |
+
"num_input_tokens_seen": 0,
|
| 648 |
+
"num_train_epochs": 1,
|
| 649 |
+
"save_steps": 100,
|
| 650 |
+
"stateful_callbacks": {
|
| 651 |
+
"TrainerControl": {
|
| 652 |
+
"args": {
|
| 653 |
+
"should_epoch_stop": false,
|
| 654 |
+
"should_evaluate": false,
|
| 655 |
+
"should_log": false,
|
| 656 |
+
"should_save": true,
|
| 657 |
+
"should_training_stop": true
|
| 658 |
+
},
|
| 659 |
+
"attributes": {}
|
| 660 |
+
}
|
| 661 |
+
},
|
| 662 |
+
"total_flos": 5.941383373300695e+19,
|
| 663 |
+
"train_batch_size": 1,
|
| 664 |
+
"trial_name": null,
|
| 665 |
+
"trial_params": null
|
| 666 |
+
}
|
qwen-cua__runs__qwen35_4b_planA_wm_phase1-20260601-041141__v0-20260601-041256__checkpoint-607__evalweights/zero_to_fp32.py
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python
|
| 2 |
+
|
| 3 |
+
# Copyright (c) Microsoft Corporation.
|
| 4 |
+
# SPDX-License-Identifier: Apache-2.0
|
| 5 |
+
|
| 6 |
+
# DeepSpeed Team
|
| 7 |
+
|
| 8 |
+
# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
|
| 9 |
+
# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
|
| 10 |
+
# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
|
| 11 |
+
# application.
|
| 12 |
+
#
|
| 13 |
+
# example:
|
| 14 |
+
# python zero_to_fp32.py . output_dir/
|
| 15 |
+
# or
|
| 16 |
+
# python zero_to_fp32.py . output_dir/ --safe_serialization
|
| 17 |
+
|
| 18 |
+
import argparse
|
| 19 |
+
import torch
|
| 20 |
+
import glob
|
| 21 |
+
import math
|
| 22 |
+
import os
|
| 23 |
+
import re
|
| 24 |
+
import gc
|
| 25 |
+
import json
|
| 26 |
+
import numpy as np
|
| 27 |
+
from tqdm import tqdm
|
| 28 |
+
from collections import OrderedDict
|
| 29 |
+
from dataclasses import dataclass
|
| 30 |
+
|
| 31 |
+
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
| 32 |
+
# DeepSpeed data structures it has to be available in the current python environment.
|
| 33 |
+
from deepspeed.utils import logger
|
| 34 |
+
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
| 35 |
+
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
| 36 |
+
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
@dataclass
|
| 40 |
+
class zero_model_state:
|
| 41 |
+
buffers: dict()
|
| 42 |
+
param_shapes: dict()
|
| 43 |
+
shared_params: list
|
| 44 |
+
ds_version: int
|
| 45 |
+
frozen_param_shapes: dict()
|
| 46 |
+
frozen_param_fragments: dict()
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
debug = 0
|
| 50 |
+
|
| 51 |
+
# load to cpu
|
| 52 |
+
device = torch.device('cpu')
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
def atoi(text):
|
| 56 |
+
return int(text) if text.isdigit() else text
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
def natural_keys(text):
|
| 60 |
+
'''
|
| 61 |
+
alist.sort(key=natural_keys) sorts in human order
|
| 62 |
+
http://nedbatchelder.com/blog/200712/human_sorting.html
|
| 63 |
+
(See Toothy's implementation in the comments)
|
| 64 |
+
'''
|
| 65 |
+
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
def get_model_state_file(checkpoint_dir, zero_stage):
|
| 69 |
+
if not os.path.isdir(checkpoint_dir):
|
| 70 |
+
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
| 71 |
+
|
| 72 |
+
# there should be only one file
|
| 73 |
+
if zero_stage <= 2:
|
| 74 |
+
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
| 75 |
+
elif zero_stage == 3:
|
| 76 |
+
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
| 77 |
+
|
| 78 |
+
if not os.path.exists(file):
|
| 79 |
+
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
| 80 |
+
|
| 81 |
+
return file
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
| 85 |
+
# XXX: need to test that this simple glob rule works for multi-node setup too
|
| 86 |
+
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
| 87 |
+
|
| 88 |
+
if len(ckpt_files) == 0:
|
| 89 |
+
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
| 90 |
+
|
| 91 |
+
return ckpt_files
|
| 92 |
+
|
| 93 |
+
|
| 94 |
+
def get_optim_files(checkpoint_dir):
|
| 95 |
+
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
def get_model_state_files(checkpoint_dir):
|
| 99 |
+
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
def parse_model_states(files):
|
| 103 |
+
zero_model_states = []
|
| 104 |
+
for file in files:
|
| 105 |
+
state_dict = torch.load(file, map_location=device, weights_only=False)
|
| 106 |
+
|
| 107 |
+
if BUFFER_NAMES not in state_dict:
|
| 108 |
+
raise ValueError(f"{file} is not a model state checkpoint")
|
| 109 |
+
buffer_names = state_dict[BUFFER_NAMES]
|
| 110 |
+
if debug:
|
| 111 |
+
print("Found buffers:", buffer_names)
|
| 112 |
+
|
| 113 |
+
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
| 114 |
+
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
| 115 |
+
param_shapes = state_dict[PARAM_SHAPES]
|
| 116 |
+
|
| 117 |
+
# collect parameters that are included in param_shapes
|
| 118 |
+
param_names = []
|
| 119 |
+
for s in param_shapes:
|
| 120 |
+
for name in s.keys():
|
| 121 |
+
param_names.append(name)
|
| 122 |
+
|
| 123 |
+
# update with frozen parameters
|
| 124 |
+
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
| 125 |
+
if frozen_param_shapes is not None:
|
| 126 |
+
if debug:
|
| 127 |
+
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
| 128 |
+
param_names += list(frozen_param_shapes.keys())
|
| 129 |
+
|
| 130 |
+
# handle shared params
|
| 131 |
+
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
| 132 |
+
|
| 133 |
+
ds_version = state_dict.get(DS_VERSION, None)
|
| 134 |
+
|
| 135 |
+
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
| 136 |
+
|
| 137 |
+
z_model_state = zero_model_state(buffers=buffers,
|
| 138 |
+
param_shapes=param_shapes,
|
| 139 |
+
shared_params=shared_params,
|
| 140 |
+
ds_version=ds_version,
|
| 141 |
+
frozen_param_shapes=frozen_param_shapes,
|
| 142 |
+
frozen_param_fragments=frozen_param_fragments)
|
| 143 |
+
zero_model_states.append(z_model_state)
|
| 144 |
+
|
| 145 |
+
return zero_model_states
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
def parse_optim_states(files, ds_checkpoint_dir):
|
| 149 |
+
total_files = len(files)
|
| 150 |
+
state_dicts = []
|
| 151 |
+
for f in tqdm(files, desc='Loading checkpoint shards'):
|
| 152 |
+
state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
|
| 153 |
+
# immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
|
| 154 |
+
# and also handle the case where it was already removed by another helper script
|
| 155 |
+
state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
|
| 156 |
+
state_dicts.append(state_dict)
|
| 157 |
+
|
| 158 |
+
if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
| 159 |
+
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
| 160 |
+
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
| 161 |
+
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
| 162 |
+
|
| 163 |
+
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
| 164 |
+
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
| 165 |
+
# use the max of the partition_count to get the dp world_size.
|
| 166 |
+
|
| 167 |
+
if type(world_size) is list:
|
| 168 |
+
world_size = max(world_size)
|
| 169 |
+
|
| 170 |
+
if world_size != total_files:
|
| 171 |
+
raise ValueError(
|
| 172 |
+
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
| 173 |
+
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
| 174 |
+
)
|
| 175 |
+
|
| 176 |
+
# the groups are named differently in each stage
|
| 177 |
+
if zero_stage <= 2:
|
| 178 |
+
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
| 179 |
+
elif zero_stage == 3:
|
| 180 |
+
fp32_groups_key = FP32_FLAT_GROUPS
|
| 181 |
+
else:
|
| 182 |
+
raise ValueError(f"unknown zero stage {zero_stage}")
|
| 183 |
+
|
| 184 |
+
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
| 185 |
+
return zero_stage, world_size, fp32_flat_groups
|
| 186 |
+
|
| 187 |
+
|
| 188 |
+
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
|
| 189 |
+
"""
|
| 190 |
+
Returns fp32 state_dict reconstructed from ds checkpoint
|
| 191 |
+
|
| 192 |
+
Args:
|
| 193 |
+
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
| 194 |
+
|
| 195 |
+
"""
|
| 196 |
+
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
| 197 |
+
|
| 198 |
+
optim_files = get_optim_files(ds_checkpoint_dir)
|
| 199 |
+
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
| 200 |
+
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
| 201 |
+
|
| 202 |
+
model_files = get_model_state_files(ds_checkpoint_dir)
|
| 203 |
+
|
| 204 |
+
zero_model_states = parse_model_states(model_files)
|
| 205 |
+
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
| 206 |
+
|
| 207 |
+
if zero_stage <= 2:
|
| 208 |
+
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 209 |
+
exclude_frozen_parameters)
|
| 210 |
+
elif zero_stage == 3:
|
| 211 |
+
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 212 |
+
exclude_frozen_parameters)
|
| 213 |
+
|
| 214 |
+
|
| 215 |
+
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
| 216 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 217 |
+
return
|
| 218 |
+
|
| 219 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 220 |
+
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
| 221 |
+
|
| 222 |
+
if debug:
|
| 223 |
+
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
| 224 |
+
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 225 |
+
|
| 226 |
+
wanted_params = len(frozen_param_shapes)
|
| 227 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 228 |
+
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
| 229 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 230 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 231 |
+
|
| 232 |
+
total_params = 0
|
| 233 |
+
total_numel = 0
|
| 234 |
+
for name, shape in frozen_param_shapes.items():
|
| 235 |
+
total_params += 1
|
| 236 |
+
unpartitioned_numel = shape.numel()
|
| 237 |
+
total_numel += unpartitioned_numel
|
| 238 |
+
|
| 239 |
+
state_dict[name] = frozen_param_fragments[name]
|
| 240 |
+
|
| 241 |
+
if debug:
|
| 242 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 243 |
+
|
| 244 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 245 |
+
|
| 246 |
+
|
| 247 |
+
def _has_callable(obj, fn):
|
| 248 |
+
attr = getattr(obj, fn, None)
|
| 249 |
+
return callable(attr)
|
| 250 |
+
|
| 251 |
+
|
| 252 |
+
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 253 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 254 |
+
|
| 255 |
+
# Reconstruction protocol:
|
| 256 |
+
#
|
| 257 |
+
# XXX: document this
|
| 258 |
+
|
| 259 |
+
if debug:
|
| 260 |
+
for i in range(world_size):
|
| 261 |
+
for j in range(len(fp32_flat_groups[0])):
|
| 262 |
+
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
| 263 |
+
|
| 264 |
+
# XXX: memory usage doubles here (zero2)
|
| 265 |
+
num_param_groups = len(fp32_flat_groups[0])
|
| 266 |
+
merged_single_partition_of_fp32_groups = []
|
| 267 |
+
for i in range(num_param_groups):
|
| 268 |
+
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
| 269 |
+
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
| 270 |
+
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
| 271 |
+
avail_numel = sum(
|
| 272 |
+
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
| 273 |
+
|
| 274 |
+
if debug:
|
| 275 |
+
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
| 276 |
+
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
| 277 |
+
# not asserting if there is a mismatch due to possible padding
|
| 278 |
+
print(f"Have {avail_numel} numels to process.")
|
| 279 |
+
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
| 280 |
+
|
| 281 |
+
# params
|
| 282 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 283 |
+
# out-of-core computing solution
|
| 284 |
+
total_numel = 0
|
| 285 |
+
total_params = 0
|
| 286 |
+
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
| 287 |
+
offset = 0
|
| 288 |
+
avail_numel = full_single_fp32_vector.numel()
|
| 289 |
+
for name, shape in shapes.items():
|
| 290 |
+
|
| 291 |
+
unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
|
| 292 |
+
total_numel += unpartitioned_numel
|
| 293 |
+
total_params += 1
|
| 294 |
+
|
| 295 |
+
if debug:
|
| 296 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 297 |
+
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
| 298 |
+
offset += unpartitioned_numel
|
| 299 |
+
|
| 300 |
+
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
| 301 |
+
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
| 302 |
+
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
| 303 |
+
# live optimizer object, so we are checking that the numbers are within the right range
|
| 304 |
+
align_to = 2 * world_size
|
| 305 |
+
|
| 306 |
+
def zero2_align(x):
|
| 307 |
+
return align_to * math.ceil(x / align_to)
|
| 308 |
+
|
| 309 |
+
if debug:
|
| 310 |
+
print(f"original offset={offset}, avail_numel={avail_numel}")
|
| 311 |
+
|
| 312 |
+
offset = zero2_align(offset)
|
| 313 |
+
avail_numel = zero2_align(avail_numel)
|
| 314 |
+
|
| 315 |
+
if debug:
|
| 316 |
+
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
| 317 |
+
|
| 318 |
+
# Sanity check
|
| 319 |
+
if offset != avail_numel:
|
| 320 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 321 |
+
|
| 322 |
+
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
| 323 |
+
|
| 324 |
+
|
| 325 |
+
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 326 |
+
exclude_frozen_parameters):
|
| 327 |
+
state_dict = OrderedDict()
|
| 328 |
+
|
| 329 |
+
# buffers
|
| 330 |
+
buffers = zero_model_states[0].buffers
|
| 331 |
+
state_dict.update(buffers)
|
| 332 |
+
if debug:
|
| 333 |
+
print(f"added {len(buffers)} buffers")
|
| 334 |
+
|
| 335 |
+
if not exclude_frozen_parameters:
|
| 336 |
+
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
| 337 |
+
|
| 338 |
+
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 339 |
+
|
| 340 |
+
# recover shared parameters
|
| 341 |
+
for pair in zero_model_states[0].shared_params:
|
| 342 |
+
if pair[1] in state_dict:
|
| 343 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 344 |
+
|
| 345 |
+
return state_dict
|
| 346 |
+
|
| 347 |
+
|
| 348 |
+
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
| 349 |
+
remainder = unpartitioned_numel % world_size
|
| 350 |
+
padding_numel = (world_size - remainder) if remainder else 0
|
| 351 |
+
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
| 352 |
+
return partitioned_numel, padding_numel
|
| 353 |
+
|
| 354 |
+
|
| 355 |
+
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
| 356 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 357 |
+
return
|
| 358 |
+
|
| 359 |
+
if debug:
|
| 360 |
+
for i in range(world_size):
|
| 361 |
+
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
| 362 |
+
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 363 |
+
|
| 364 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 365 |
+
wanted_params = len(frozen_param_shapes)
|
| 366 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 367 |
+
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
| 368 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 369 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 370 |
+
|
| 371 |
+
total_params = 0
|
| 372 |
+
total_numel = 0
|
| 373 |
+
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
| 374 |
+
total_params += 1
|
| 375 |
+
unpartitioned_numel = shape.numel()
|
| 376 |
+
total_numel += unpartitioned_numel
|
| 377 |
+
|
| 378 |
+
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
| 379 |
+
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 380 |
+
|
| 381 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 382 |
+
|
| 383 |
+
if debug:
|
| 384 |
+
print(
|
| 385 |
+
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 386 |
+
)
|
| 387 |
+
|
| 388 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 389 |
+
|
| 390 |
+
|
| 391 |
+
class GatheredTensor:
|
| 392 |
+
"""
|
| 393 |
+
A pseudo tensor that collects partitioned weights.
|
| 394 |
+
It is more memory efficient when there are multiple groups.
|
| 395 |
+
"""
|
| 396 |
+
|
| 397 |
+
def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
|
| 398 |
+
self.flat_groups = flat_groups
|
| 399 |
+
self.flat_groups_offset = flat_groups_offset
|
| 400 |
+
self.offset = offset
|
| 401 |
+
self.partitioned_numel = partitioned_numel
|
| 402 |
+
self.shape = shape
|
| 403 |
+
self.dtype = self.flat_groups[0][0].dtype
|
| 404 |
+
|
| 405 |
+
def contiguous(self):
|
| 406 |
+
"""
|
| 407 |
+
Merge partitioned weights from flat_groups into a single tensor.
|
| 408 |
+
"""
|
| 409 |
+
end_idx = self.offset + self.partitioned_numel
|
| 410 |
+
world_size = len(self.flat_groups)
|
| 411 |
+
pad_flat_param_chunks = []
|
| 412 |
+
|
| 413 |
+
for rank_i in range(world_size):
|
| 414 |
+
# for each rank, we need to collect weights from related group/groups
|
| 415 |
+
flat_groups_at_rank_i = self.flat_groups[rank_i]
|
| 416 |
+
start_group_id = None
|
| 417 |
+
end_group_id = None
|
| 418 |
+
for group_id in range(len(self.flat_groups_offset)):
|
| 419 |
+
if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
|
| 420 |
+
start_group_id = group_id
|
| 421 |
+
if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
|
| 422 |
+
end_group_id = group_id
|
| 423 |
+
break
|
| 424 |
+
# collect weights from related group/groups
|
| 425 |
+
for group_id in range(start_group_id, end_group_id + 1):
|
| 426 |
+
flat_tensor = flat_groups_at_rank_i[group_id]
|
| 427 |
+
start_offset = self.offset - self.flat_groups_offset[group_id]
|
| 428 |
+
end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
|
| 429 |
+
pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
|
| 430 |
+
|
| 431 |
+
# collect weights from all ranks
|
| 432 |
+
pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
|
| 433 |
+
param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
|
| 434 |
+
return param
|
| 435 |
+
|
| 436 |
+
|
| 437 |
+
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 438 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 439 |
+
avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
|
| 440 |
+
|
| 441 |
+
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
| 442 |
+
# param, re-consolidating each param, while dealing with padding if any
|
| 443 |
+
|
| 444 |
+
# merge list of dicts, preserving order
|
| 445 |
+
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
| 446 |
+
|
| 447 |
+
if debug:
|
| 448 |
+
for i in range(world_size):
|
| 449 |
+
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
| 450 |
+
|
| 451 |
+
wanted_params = len(param_shapes)
|
| 452 |
+
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
| 453 |
+
# not asserting if there is a mismatch due to possible padding
|
| 454 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 455 |
+
print(f"Trainable params: Have {avail_numel} numels to process.")
|
| 456 |
+
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
| 457 |
+
|
| 458 |
+
# params
|
| 459 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 460 |
+
# out-of-core computing solution
|
| 461 |
+
offset = 0
|
| 462 |
+
total_numel = 0
|
| 463 |
+
total_params = 0
|
| 464 |
+
flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
|
| 465 |
+
for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
|
| 466 |
+
unpartitioned_numel = shape.numel()
|
| 467 |
+
total_numel += unpartitioned_numel
|
| 468 |
+
total_params += 1
|
| 469 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 470 |
+
|
| 471 |
+
if debug:
|
| 472 |
+
print(
|
| 473 |
+
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 474 |
+
)
|
| 475 |
+
|
| 476 |
+
# memory efficient tensor
|
| 477 |
+
tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
|
| 478 |
+
state_dict[name] = tensor
|
| 479 |
+
offset += partitioned_numel
|
| 480 |
+
|
| 481 |
+
offset *= world_size
|
| 482 |
+
|
| 483 |
+
# Sanity check
|
| 484 |
+
if offset != avail_numel:
|
| 485 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 486 |
+
|
| 487 |
+
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
| 488 |
+
|
| 489 |
+
|
| 490 |
+
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 491 |
+
exclude_frozen_parameters):
|
| 492 |
+
state_dict = OrderedDict()
|
| 493 |
+
|
| 494 |
+
# buffers
|
| 495 |
+
buffers = zero_model_states[0].buffers
|
| 496 |
+
state_dict.update(buffers)
|
| 497 |
+
if debug:
|
| 498 |
+
print(f"added {len(buffers)} buffers")
|
| 499 |
+
|
| 500 |
+
if not exclude_frozen_parameters:
|
| 501 |
+
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
| 502 |
+
|
| 503 |
+
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 504 |
+
|
| 505 |
+
# recover shared parameters
|
| 506 |
+
for pair in zero_model_states[0].shared_params:
|
| 507 |
+
if pair[1] in state_dict:
|
| 508 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 509 |
+
|
| 510 |
+
return state_dict
|
| 511 |
+
|
| 512 |
+
|
| 513 |
+
def to_torch_tensor(state_dict, return_empty_tensor=False):
|
| 514 |
+
"""
|
| 515 |
+
Convert state_dict of GatheredTensor to torch tensor
|
| 516 |
+
"""
|
| 517 |
+
torch_state_dict = {}
|
| 518 |
+
converted_tensors = {}
|
| 519 |
+
for name, tensor in state_dict.items():
|
| 520 |
+
tensor_id = id(tensor)
|
| 521 |
+
if tensor_id in converted_tensors: # shared tensors
|
| 522 |
+
shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
|
| 523 |
+
torch_state_dict[name] = shared_tensor
|
| 524 |
+
else:
|
| 525 |
+
converted_tensors[tensor_id] = name
|
| 526 |
+
if return_empty_tensor:
|
| 527 |
+
torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
|
| 528 |
+
else:
|
| 529 |
+
torch_state_dict[name] = tensor.contiguous()
|
| 530 |
+
return torch_state_dict
|
| 531 |
+
|
| 532 |
+
|
| 533 |
+
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 534 |
+
tag=None,
|
| 535 |
+
exclude_frozen_parameters=False,
|
| 536 |
+
lazy_mode=False):
|
| 537 |
+
"""
|
| 538 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
| 539 |
+
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
| 540 |
+
via a model hub.
|
| 541 |
+
|
| 542 |
+
Args:
|
| 543 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder
|
| 544 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
|
| 545 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 546 |
+
- ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
|
| 547 |
+
Convert the pesduo tensor to torch tensor by ``.contiguous()``
|
| 548 |
+
|
| 549 |
+
Returns:
|
| 550 |
+
- pytorch ``state_dict``
|
| 551 |
+
|
| 552 |
+
A typical usage might be ::
|
| 553 |
+
|
| 554 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 555 |
+
# do the training and checkpoint saving
|
| 556 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
| 557 |
+
model = model.cpu() # move to cpu
|
| 558 |
+
model.load_state_dict(state_dict)
|
| 559 |
+
# submit to model hub or save the model to share with others
|
| 560 |
+
|
| 561 |
+
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
| 562 |
+
application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 563 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 564 |
+
|
| 565 |
+
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
| 566 |
+
|
| 567 |
+
Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
|
| 568 |
+
You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
| 569 |
+
the checkpoint. Or you can load state_dict in lazy mode ::
|
| 570 |
+
|
| 571 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 572 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
|
| 573 |
+
for name, lazy_tensor in state_dict.item():
|
| 574 |
+
tensor = lazy_tensor.contiguous() # to cpu
|
| 575 |
+
print(name, tensor)
|
| 576 |
+
# del tensor to release memory if it no longer in use
|
| 577 |
+
"""
|
| 578 |
+
if tag is None:
|
| 579 |
+
latest_path = os.path.join(checkpoint_dir, 'latest')
|
| 580 |
+
if os.path.isfile(latest_path):
|
| 581 |
+
with open(latest_path, 'r') as fd:
|
| 582 |
+
tag = fd.read().strip()
|
| 583 |
+
else:
|
| 584 |
+
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
| 585 |
+
|
| 586 |
+
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
| 587 |
+
|
| 588 |
+
if not os.path.isdir(ds_checkpoint_dir):
|
| 589 |
+
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
| 590 |
+
|
| 591 |
+
state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
|
| 592 |
+
if lazy_mode:
|
| 593 |
+
return state_dict
|
| 594 |
+
else:
|
| 595 |
+
return to_torch_tensor(state_dict)
|
| 596 |
+
|
| 597 |
+
|
| 598 |
+
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
|
| 599 |
+
output_dir,
|
| 600 |
+
max_shard_size="5GB",
|
| 601 |
+
safe_serialization=False,
|
| 602 |
+
tag=None,
|
| 603 |
+
exclude_frozen_parameters=False):
|
| 604 |
+
"""
|
| 605 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
| 606 |
+
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
| 607 |
+
|
| 608 |
+
Args:
|
| 609 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 610 |
+
- ``output_dir``: directory to the pytorch fp32 state_dict output files
|
| 611 |
+
- ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
|
| 612 |
+
- ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
|
| 613 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 614 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 615 |
+
"""
|
| 616 |
+
|
| 617 |
+
# Dependency pre-check
|
| 618 |
+
if safe_serialization:
|
| 619 |
+
try:
|
| 620 |
+
from safetensors.torch import save_file
|
| 621 |
+
except ImportError:
|
| 622 |
+
print('If you want to use `safe_serialization`, please `pip install safetensors`')
|
| 623 |
+
raise
|
| 624 |
+
if max_shard_size is not None:
|
| 625 |
+
try:
|
| 626 |
+
from huggingface_hub import split_torch_state_dict_into_shards
|
| 627 |
+
except ImportError:
|
| 628 |
+
print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
|
| 629 |
+
raise
|
| 630 |
+
|
| 631 |
+
# Convert zero checkpoint to state_dict
|
| 632 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 633 |
+
tag,
|
| 634 |
+
exclude_frozen_parameters,
|
| 635 |
+
lazy_mode=True)
|
| 636 |
+
|
| 637 |
+
# Shard the model if it is too big.
|
| 638 |
+
weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
|
| 639 |
+
if max_shard_size is not None:
|
| 640 |
+
filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
|
| 641 |
+
# an memory-efficient approach for sharding
|
| 642 |
+
empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
|
| 643 |
+
state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
|
| 644 |
+
filename_pattern=filename_pattern,
|
| 645 |
+
max_shard_size=max_shard_size)
|
| 646 |
+
else:
|
| 647 |
+
from collections import namedtuple
|
| 648 |
+
StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
|
| 649 |
+
state_dict_split = StateDictSplit(is_sharded=False,
|
| 650 |
+
filename_to_tensors={weights_name: list(state_dict.keys())})
|
| 651 |
+
|
| 652 |
+
# Save the model by shard
|
| 653 |
+
os.makedirs(output_dir, exist_ok=True)
|
| 654 |
+
filename_to_tensors = state_dict_split.filename_to_tensors.items()
|
| 655 |
+
for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
|
| 656 |
+
shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
|
| 657 |
+
shard_state_dict = to_torch_tensor(shard_state_dict)
|
| 658 |
+
output_path = os.path.join(output_dir, shard_file)
|
| 659 |
+
if safe_serialization:
|
| 660 |
+
save_file(shard_state_dict, output_path, metadata={"format": "pt"})
|
| 661 |
+
else:
|
| 662 |
+
torch.save(shard_state_dict, output_path)
|
| 663 |
+
# release the memory of current shard
|
| 664 |
+
for tensor_name in list(shard_state_dict.keys()):
|
| 665 |
+
del state_dict[tensor_name]
|
| 666 |
+
del shard_state_dict[tensor_name]
|
| 667 |
+
del shard_state_dict
|
| 668 |
+
gc.collect()
|
| 669 |
+
|
| 670 |
+
# Save index if sharded
|
| 671 |
+
if state_dict_split.is_sharded:
|
| 672 |
+
index = {
|
| 673 |
+
"metadata": state_dict_split.metadata,
|
| 674 |
+
"weight_map": state_dict_split.tensor_to_filename,
|
| 675 |
+
}
|
| 676 |
+
save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
|
| 677 |
+
save_index_file = os.path.join(output_dir, save_index_file)
|
| 678 |
+
with open(save_index_file, "w", encoding="utf-8") as f:
|
| 679 |
+
content = json.dumps(index, indent=2, sort_keys=True) + "\n"
|
| 680 |
+
f.write(content)
|
| 681 |
+
|
| 682 |
+
|
| 683 |
+
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
| 684 |
+
"""
|
| 685 |
+
1. Put the provided model to cpu
|
| 686 |
+
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
| 687 |
+
3. Load it into the provided model
|
| 688 |
+
|
| 689 |
+
Args:
|
| 690 |
+
- ``model``: the model object to update
|
| 691 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 692 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 693 |
+
|
| 694 |
+
Returns:
|
| 695 |
+
- ``model`: modified model
|
| 696 |
+
|
| 697 |
+
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
| 698 |
+
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
| 699 |
+
conveniently placed for you in the checkpoint folder.
|
| 700 |
+
|
| 701 |
+
A typical usage might be ::
|
| 702 |
+
|
| 703 |
+
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
| 704 |
+
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
| 705 |
+
# submit to model hub or save the model to share with others
|
| 706 |
+
|
| 707 |
+
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
| 708 |
+
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 709 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 710 |
+
|
| 711 |
+
"""
|
| 712 |
+
logger.info("Extracting fp32 weights")
|
| 713 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 714 |
+
|
| 715 |
+
logger.info("Overwriting model with fp32 weights")
|
| 716 |
+
model = model.cpu()
|
| 717 |
+
model.load_state_dict(state_dict, strict=False)
|
| 718 |
+
|
| 719 |
+
return model
|
| 720 |
+
|
| 721 |
+
|
| 722 |
+
if __name__ == "__main__":
|
| 723 |
+
parser = argparse.ArgumentParser()
|
| 724 |
+
parser.add_argument("checkpoint_dir",
|
| 725 |
+
type=str,
|
| 726 |
+
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
| 727 |
+
parser.add_argument("output_dir",
|
| 728 |
+
type=str,
|
| 729 |
+
help="directory to the pytorch fp32 state_dict output files"
|
| 730 |
+
"(e.g. path/checkpoint-12-output/)")
|
| 731 |
+
parser.add_argument(
|
| 732 |
+
"--max_shard_size",
|
| 733 |
+
type=str,
|
| 734 |
+
default="5GB",
|
| 735 |
+
help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
|
| 736 |
+
"lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
|
| 737 |
+
"We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
|
| 738 |
+
"without CPU OOM issues.")
|
| 739 |
+
parser.add_argument(
|
| 740 |
+
"--safe_serialization",
|
| 741 |
+
default=False,
|
| 742 |
+
action='store_true',
|
| 743 |
+
help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
|
| 744 |
+
parser.add_argument("-t",
|
| 745 |
+
"--tag",
|
| 746 |
+
type=str,
|
| 747 |
+
default=None,
|
| 748 |
+
help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
|
| 749 |
+
parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
|
| 750 |
+
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
| 751 |
+
args = parser.parse_args()
|
| 752 |
+
|
| 753 |
+
debug = args.debug
|
| 754 |
+
|
| 755 |
+
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
|
| 756 |
+
args.output_dir,
|
| 757 |
+
max_shard_size=args.max_shard_size,
|
| 758 |
+
safe_serialization=args.safe_serialization,
|
| 759 |
+
tag=args.tag,
|
| 760 |
+
exclude_frozen_parameters=args.exclude_frozen_parameters)
|