Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

args.json

@@ -0,0 +1,403 @@
+{
+  "output_dir": "/data/songshuo.lu/ms-swift/output/v10-20260426-233812",
+  "per_device_train_batch_size": 1,
+  "num_train_epochs": 3.0,
+  "max_steps": -1,
+  "learning_rate": 1e-05,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_steps": 50.0,
+  "optim": "adamw_torch_fused",
+  "optim_args": null,
+  "weight_decay": 0.1,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "optim_target_modules": null,
+  "gradient_accumulation_steps": 16,
+  "average_tokens_across_devices": true,
+  "max_grad_norm": 1.0,
+  "label_smoothing_factor": 0.0,
+  "bf16": true,
+  "fp16": false,
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": null,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "use_liger_kernel": false,
+  "liger_kernel_config": null,
+  "use_cache": false,
+  "neftune_noise_alpha": null,
+  "torch_empty_cache_steps": null,
+  "auto_find_batch_size": false,
+  "logging_strategy": "steps",
+  "logging_steps": 1,
+  "logging_first_step": true,
+  "log_on_each_node": true,
+  "logging_nan_inf_filter": true,
+  "include_num_input_tokens_seen": false,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "disable_tqdm": null,
+  "report_to": [
+    "tensorboard"
+  ],
+  "run_name": "/data/songshuo.lu/ms-swift/output/v10-20260426-233812",
+  "project": "huggingface",
+  "trackio_space_id": "trackio",
+  "eval_strategy": "steps",
+  "eval_steps": 2000.0,
+  "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": null,
+  "include_for_metrics": [],
+  "batch_eval_metrics": false,
+  "save_only_model": true,
+  "save_strategy": "steps",
+  "save_steps": 50.0,
+  "save_on_each_node": false,
+  "save_total_limit": null,
+  "enable_jit_checkpoint": false,
+  "push_to_hub": false,
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_always_push": false,
+  "hub_revision": null,
+  "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": {
+    "dispatch_batches": false
+  },
+  "parallelism_config": null,
+  "dataloader_drop_last": false,
+  "dataloader_num_workers": 32,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "dataloader_prefetch_factor": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "train_sampling_strategy": "random",
+  "length_column_name": "length",
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "ddp_backend": null,
+  "ddp_timeout": 18000000,
+  "fsdp": [],
+  "fsdp_config": null,
+  "deepspeed": {
+    "fp16": {
+      "enabled": "auto",
+      "loss_scale": 0,
+      "loss_scale_window": 1000,
+      "initial_scale_power": 16,
+      "hysteresis": 2,
+      "min_loss_scale": 1
+    },
+    "bf16": {
+      "enabled": "auto"
+    },
+    "zero_optimization": {
+      "stage": 3,
+      "offload_optimizer": {
+        "device": "cpu",
+        "pin_memory": true
+      },
+      "offload_param": {
+        "device": "cpu",
+        "pin_memory": true
+      },
+      "overlap_comm": false,
+      "contiguous_gradients": true,
+      "sub_group_size": 1000000000.0,
+      "reduce_bucket_size": "auto",
+      "stage3_prefetch_bucket_size": "auto",
+      "stage3_param_persistence_threshold": "auto",
+      "stage3_max_live_parameters": 1000000000.0,
+      "stage3_max_reuse_distance": 1000000000.0,
+      "stage3_gather_16bit_weights_on_model_save": true
+    },
+    "gradient_accumulation_steps": "auto",
+    "gradient_clipping": "auto",
+    "steps_per_print": 2000,
+    "train_batch_size": "auto",
+    "train_micro_batch_size_per_gpu": "auto",
+    "wall_clock_breakdown": false
+  },
+  "debug": null,
+  "skip_memory_metrics": true,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "resume_from_checkpoint": null,
+  "warmup_ratio": null,
+  "logging_dir": "/data/songshuo.lu/ms-swift/output/v10-20260426-233812/runs",
+  "local_rank": 0,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "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": null,
+  "aligner_lr": null,
+  "vit_lr": null,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "eval_metric": null,
+  "callbacks": [],
+  "early_stop_interval": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "tuner_type": "full",
+  "use_galore": false,
+  "galore_target_modules": null,
+  "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,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/models/Qwen3.6-27B",
+  "model_type": "qwen3_5",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": "flash_attention_2",
+  "experts_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "qwen3_5",
+  "system": null,
+  "max_length": 36000,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_side": "right",
+  "padding_free": false,
+  "loss_scale": "last_round+ignore_empty_think",
+  "sequence_parallel_size": 1,
+  "template_backend": "swift",
+  "response_prefix": null,
+  "enable_thinking": null,
+  "add_non_thinking_prefix": true,
+  "dataset": [
+    "/data/songshuo.lu/datas/sft-v26/general_thinking_5k.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_kernelbench_l2l3_variants.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_l3_v1.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_l3_v2.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_l3_v3.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_l3_v4.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_l3.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_no_bias.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_v1.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_v3.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_short_prompt_mt_v1.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_short_prompt_mt_v2.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_short_prompt_no_bias_st_v2.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_qwen3_235b_distill_robust_strict_v2.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/cuda_data_ds_v3_distill_torch_op_v1.jsonl",
+    "/data/songshuo.lu/datas/sft-v26/operator_hard_related_qa.jsonl"
+  ],
+  "val_dataset": [
+    "/data/songshuo.lu/datas/sft-v26/cuda_agent_6k_k2.5_distill_robust_strict_mt_l3_v1.jsonl#50"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 32,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": [
+    "MusaChat-9B-v1"
+  ],
+  "model_author": [
+    "DashLuuu"
+  ],
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "structured_outputs_regex": null,
+  "adapters": [],
+  "external_plugins": [],
+  "custom_register_path": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [
+    "model.visual",
+    "model.visual.merger"
+  ],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": false,
+  "freeze_vit": true,
+  "freeze_aligner": true,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": "ms-swift",
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_notification_method": null,
+  "swanlab_webhook_url": null,
+  "swanlab_secret": null,
+  "swanlab_sender_email": null,
+  "swanlab_receiver_email": null,
+  "swanlab_smtp_server": null,
+  "swanlab_smtp_port": null,
+  "swanlab_email_language": "zh",
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "swift_version": "4.2.0.dev0",
+  "ckpt_dir": null,
+  "rank": 0,
+  "global_world_size": 8,
+  "local_world_size": 8,
+  "model_suffix": "Qwen3.6-27B",
+  "model_info": "ModelInfo(model_type='qwen3_5', model_dir='/data/models/Qwen3.6-27B', 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)",
+  "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=[])",
+  "model_dir": "/data/models/Qwen3.6-27B",
+  "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='')",
+  "_val_dataset_exists": true,
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "evaluation_strategy": "steps",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/songshuo.lu/ms-swift/output/v10-20260426-233812', per_device_train_batch_size=1, num_train_epochs=3.0, max_steps=-1, learning_rate=1e-05, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_steps=50.0, optim=<OptimizerNames.ADAMW_TORCH_FUSED: 'adamw_torch_fused'>, optim_args=None, weight_decay=0.1, 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=None, 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=1, 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=['tensorboard'], run_name='/data/songshuo.lu/ms-swift/output/v10-20260426-233812', project='huggingface', trackio_space_id='trackio', eval_strategy=<IntervalStrategy.STEPS: 'steps'>, eval_steps=2000, 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=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=50, save_on_each_node=False, save_total_limit=None, 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=32, dataloader_pin_memory=True, dataloader_persistent_workers=False, dataloader_prefetch_factor=2, 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_backend=None, ddp_timeout=18000000, fsdp=[], fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, deepspeed={'fp16': {'enabled': 'auto', 'loss_scale': 0, 'loss_scale_window': 1000, 'initial_scale_power': 16, 'hysteresis': 2, 'min_loss_scale': 1}, 'bf16': {'enabled': 'auto'}, 'zero_optimization': {'stage': 3, 'offload_optimizer': {'device': 'cpu', 'pin_memory': True}, 'offload_param': {'device': 'cpu', 'pin_memory': True}, 'overlap_comm': False, 'contiguous_gradients': True, 'sub_group_size': 1000000000.0, 'reduce_bucket_size': 'auto', 'stage3_prefetch_bucket_size': 'auto', 'stage3_param_persistence_threshold': 'auto', 'stage3_max_live_parameters': 1000000000.0, 'stage3_max_reuse_distance': 1000000000.0, 'stage3_gather_16bit_weights_on_model_save': True}, 'gradient_accumulation_steps': 'auto', 'gradient_clipping': 'auto', 'steps_per_print': 2000, 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'wall_clock_breakdown': False}, debug=[], skip_memory_metrics=True, do_train=False, do_eval=True, do_predict=False, resume_from_checkpoint=None, warmup_ratio=None, logging_dir='/data/songshuo.lu/ms-swift/output/v10-20260426-233812/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)"
+}

chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\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>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if (preserve_thinking is defined and preserve_thinking is true) or (loop.index0 > ns.last_query_index) %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,147 @@
+{
+  "architectures": [
+    "Qwen3_5ForConditionalGeneration"
+  ],
+  "bos_token_id": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 248046,
+  "hidden_size": 5120,
+  "image_token_id": 248056,
+  "language_model_only": false,
+  "model_type": "qwen3_5",
+  "pad_token_id": 248044,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": 248044,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "initializer_range": 0.02,
+    "intermediate_size": 17408,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention"
+    ],
+    "linear_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 48,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 24,
+    "num_hidden_layers": 64,
+    "num_key_value_heads": 4,
+    "output_gate_type": "swish",
+    "pad_token_id": 248044,
+    "partial_rotary_factor": 0.25,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": false,
+    "vocab_size": 248320
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.2.0",
+  "use_cache": false,
+  "video_token_id": 248057,
+  "vision_config": {
+    "deepstack_visual_indexes": [],
+    "depth": 27,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4304,
+    "model_type": "qwen3_5",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 5120,
+    "patch_size": 16,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 248044,
+  "do_sample": true,
+  "eos_token_id": [
+    248046,
+    248044
+  ],
+  "pad_token_id": 248044,
+  "temperature": 1.0,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "5.2.0"
+}

model-00001-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a10361f0790772d7c613e59ca58e4f6837807edc21e59c3b8d1704837a701c97
+size 2542796928

model-00002-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53b5eef4108b5ca68e7acd3df22ae10579a7236957b02c223b9c14f82d60d68f
+size 4842451920

model-00003-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc370e2fa68866704bfd67ed730d2f7d47e694d21174a328e694061ac557dd1
+size 4965227944

model-00004-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14bebc6966ec03ad9bb665bf97be107d8f84b1983049a97923095000625f74a6
+size 4912819264

model-00005-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:474832f396cdc08c589cbc7a08b0ee602c994b78c6a2db71d45148e76268382b
+size 4986198544

model-00006-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6362465eaaf59d3aa5c0ad552b403161475bd14aaac1ce97412591968670ab4a
+size 4912819320

model-00007-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d63e5bbe8fb142884193564f37a8a43f174e7715e4b44e2acc4f62bd10ebdaa3
+size 4932703272

model-00008-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2683f0032247a35c5619215642019ffa22eca07a106c7ea57ac83ab8a3e2d49c
+size 4966314576

model-00009-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acc8090ea3d96a4221470e255a3075614d5ee87537c826ef5c3fdb7e0cf97f6b
+size 4964162248

model-00010-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d2d91120d86b2539d31db794750e02cf7de4536b5176783f6ac300971c6e32a
+size 4933789824

model-00011-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8055858d4e03d46afa05473eb25ef339b074543319922bab02bd81d7793c366
+size 4965228032

model-00012-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fd3e1364be23487c83d3fc435b051ab13e9015fcd341491386b24bf98e8f569
+size 2789094896

preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 16777216,
+        "shortest_edge": 65536
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "image_processor_type": "Qwen2VLImageProcessorFast"
+}

processor_config.json

@@ -0,0 +1,63 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessorFast",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "data_format": "channels_first",
+    "default_to_square": true,
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
+size 19989343

tokenizer_config.json

@@ -0,0 +1,32 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "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+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

trainer_state.json

@@ -0,0 +1,2434 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.1588377723970944,
+  "eval_steps": 2000,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00387409200968523,
+      "grad_norm": 1.598986029624939,
+      "learning_rate": 2.0000000000000002e-07,
+      "loss": 0.40103477239608765,
+      "step": 1,
+      "token_acc": 0.8705013179702646
+    },
+    {
+      "epoch": 0.00774818401937046,
+      "grad_norm": 1.988427758216858,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 0.4314175248146057,
+      "step": 2,
+      "token_acc": 0.8610088406262493
+    },
+    {
+      "epoch": 0.01162227602905569,
+      "grad_norm": 1.6525965929031372,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 0.41751521825790405,
+      "step": 3,
+      "token_acc": 0.8659394954574845
+    },
+    {
+      "epoch": 0.01549636803874092,
+      "grad_norm": 1.3594496250152588,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 0.39516761898994446,
+      "step": 4,
+      "token_acc": 0.8712739341656057
+    },
+    {
+      "epoch": 0.01937046004842615,
+      "grad_norm": 1.4459697008132935,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.41443824768066406,
+      "step": 5,
+      "token_acc": 0.8673064711013153
+    },
+    {
+      "epoch": 0.02324455205811138,
+      "grad_norm": 1.165871024131775,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 0.3951181471347809,
+      "step": 6,
+      "token_acc": 0.8717731277799119
+    },
+    {
+      "epoch": 0.02711864406779661,
+      "grad_norm": 1.150416374206543,
+      "learning_rate": 1.4000000000000001e-06,
+      "loss": 0.40562719106674194,
+      "step": 7,
+      "token_acc": 0.8683618627898853
+    },
+    {
+      "epoch": 0.03099273607748184,
+      "grad_norm": 0.7621377110481262,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 0.4054454565048218,
+      "step": 8,
+      "token_acc": 0.8672108063124587
+    },
+    {
+      "epoch": 0.03486682808716707,
+      "grad_norm": 0.588590681552887,
+      "learning_rate": 1.8000000000000001e-06,
+      "loss": 0.383542001247406,
+      "step": 9,
+      "token_acc": 0.8732824386699718
+    },
+    {
+      "epoch": 0.0387409200968523,
+      "grad_norm": 0.5067570805549622,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.3769374489784241,
+      "step": 10,
+      "token_acc": 0.8749500487669789
+    },
+    {
+      "epoch": 0.04261501210653753,
+      "grad_norm": 0.6109248995780945,
+      "learning_rate": 2.2e-06,
+      "loss": 0.3687226176261902,
+      "step": 11,
+      "token_acc": 0.8778429629931872
+    },
+    {
+      "epoch": 0.04648910411622276,
+      "grad_norm": 0.6168301701545715,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 0.3631238639354706,
+      "step": 12,
+      "token_acc": 0.8792909317747671
+    },
+    {
+      "epoch": 0.05036319612590799,
+      "grad_norm": 0.5205990076065063,
+      "learning_rate": 2.6e-06,
+      "loss": 0.37530872225761414,
+      "step": 13,
+      "token_acc": 0.8747995859550826
+    },
+    {
+      "epoch": 0.05423728813559322,
+      "grad_norm": 0.4970836639404297,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 0.33857205510139465,
+      "step": 14,
+      "token_acc": 0.8863650931395268
+    },
+    {
+      "epoch": 0.05811138014527845,
+      "grad_norm": 0.4103075861930847,
+      "learning_rate": 3e-06,
+      "loss": 0.38399845361709595,
+      "step": 15,
+      "token_acc": 0.8722473100295478
+    },
+    {
+      "epoch": 0.06198547215496368,
+      "grad_norm": 0.505113959312439,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 0.37927311658859253,
+      "step": 16,
+      "token_acc": 0.8732506907722828
+    },
+    {
+      "epoch": 0.06585956416464891,
+      "grad_norm": 0.4578634202480316,
+      "learning_rate": 3.4000000000000005e-06,
+      "loss": 0.388744592666626,
+      "step": 17,
+      "token_acc": 0.8707925977418891
+    },
+    {
+      "epoch": 0.06973365617433414,
+      "grad_norm": 0.40881460905075073,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 0.37862884998321533,
+      "step": 18,
+      "token_acc": 0.8738148420049672
+    },
+    {
+      "epoch": 0.07360774818401937,
+      "grad_norm": 0.3267415165901184,
+      "learning_rate": 3.8000000000000005e-06,
+      "loss": 0.3523765206336975,
+      "step": 19,
+      "token_acc": 0.8821479488850912
+    },
+    {
+      "epoch": 0.0774818401937046,
+      "grad_norm": 0.3520510196685791,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.37575048208236694,
+      "step": 20,
+      "token_acc": 0.8746806805808569
+    },
+    {
+      "epoch": 0.08135593220338982,
+      "grad_norm": 0.3177695870399475,
+      "learning_rate": 4.2000000000000004e-06,
+      "loss": 0.3877210021018982,
+      "step": 21,
+      "token_acc": 0.8709381583839385
+    },
+    {
+      "epoch": 0.08523002421307506,
+      "grad_norm": 0.3101595640182495,
+      "learning_rate": 4.4e-06,
+      "loss": 0.35647860169410706,
+      "step": 22,
+      "token_acc": 0.8802609194999448
+    },
+    {
+      "epoch": 0.0891041162227603,
+      "grad_norm": 0.42295873165130615,
+      "learning_rate": 4.600000000000001e-06,
+      "loss": 0.34535130858421326,
+      "step": 23,
+      "token_acc": 0.8842312960154491
+    },
+    {
+      "epoch": 0.09297820823244551,
+      "grad_norm": 0.38459983468055725,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 0.3480440676212311,
+      "step": 24,
+      "token_acc": 0.8830844934941354
+    },
+    {
+      "epoch": 0.09685230024213075,
+      "grad_norm": 0.3167020082473755,
+      "learning_rate": 5e-06,
+      "loss": 0.3617573082447052,
+      "step": 25,
+      "token_acc": 0.8794729562611736
+    },
+    {
+      "epoch": 0.10072639225181598,
+      "grad_norm": 0.3235217332839966,
+      "learning_rate": 5.2e-06,
+      "loss": 0.34485846757888794,
+      "step": 26,
+      "token_acc": 0.8849654381719892
+    },
+    {
+      "epoch": 0.10460048426150122,
+      "grad_norm": 0.33688801527023315,
+      "learning_rate": 5.400000000000001e-06,
+      "loss": 0.325369268655777,
+      "step": 27,
+      "token_acc": 0.8904570911619978
+    },
+    {
+      "epoch": 0.10847457627118644,
+      "grad_norm": 0.28384602069854736,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 0.3820268213748932,
+      "step": 28,
+      "token_acc": 0.8722670041260794
+    },
+    {
+      "epoch": 0.11234866828087167,
+      "grad_norm": 0.2726050019264221,
+      "learning_rate": 5.8e-06,
+      "loss": 0.34821516275405884,
+      "step": 29,
+      "token_acc": 0.8829695430808375
+    },
+    {
+      "epoch": 0.1162227602905569,
+      "grad_norm": 0.2613418698310852,
+      "learning_rate": 6e-06,
+      "loss": 0.3505156636238098,
+      "step": 30,
+      "token_acc": 0.8820213661332177
+    },
+    {
+      "epoch": 0.12009685230024213,
+      "grad_norm": 0.27066054940223694,
+      "learning_rate": 6.200000000000001e-06,
+      "loss": 0.3500295877456665,
+      "step": 31,
+      "token_acc": 0.8819775128328553
+    },
+    {
+      "epoch": 0.12397094430992736,
+      "grad_norm": 0.2605418562889099,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 0.32833147048950195,
+      "step": 32,
+      "token_acc": 0.8892601629599358
+    },
+    {
+      "epoch": 0.12784503631961258,
+      "grad_norm": 0.2576088607311249,
+      "learning_rate": 6.600000000000001e-06,
+      "loss": 0.3447936475276947,
+      "step": 33,
+      "token_acc": 0.8835445537223737
+    },
+    {
+      "epoch": 0.13171912832929783,
+      "grad_norm": 0.2707255482673645,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 0.352622389793396,
+      "step": 34,
+      "token_acc": 0.8808578896779464
+    },
+    {
+      "epoch": 0.13559322033898305,
+      "grad_norm": 0.23704984784126282,
+      "learning_rate": 7e-06,
+      "loss": 0.34251606464385986,
+      "step": 35,
+      "token_acc": 0.8839590527934595
+    },
+    {
+      "epoch": 0.13946731234866827,
+      "grad_norm": 0.2552218735218048,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 0.36937713623046875,
+      "step": 36,
+      "token_acc": 0.8746555562093041
+    },
+    {
+      "epoch": 0.14334140435835352,
+      "grad_norm": 0.25926339626312256,
+      "learning_rate": 7.4e-06,
+      "loss": 0.37243181467056274,
+      "step": 37,
+      "token_acc": 0.8742657147624016
+    },
+    {
+      "epoch": 0.14721549636803874,
+      "grad_norm": 0.25272250175476074,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 0.3371140956878662,
+      "step": 38,
+      "token_acc": 0.8851879286597788
+    },
+    {
+      "epoch": 0.15108958837772396,
+      "grad_norm": 0.2262120097875595,
+      "learning_rate": 7.800000000000002e-06,
+      "loss": 0.32758837938308716,
+      "step": 39,
+      "token_acc": 0.8883092864316684
+    },
+    {
+      "epoch": 0.1549636803874092,
+      "grad_norm": 0.26067835092544556,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.32051679491996765,
+      "step": 40,
+      "token_acc": 0.8908219532219895
+    },
+    {
+      "epoch": 0.15883777239709443,
+      "grad_norm": 0.22696885466575623,
+      "learning_rate": 8.2e-06,
+      "loss": 0.34018558263778687,
+      "step": 41,
+      "token_acc": 0.8843387459744694
+    },
+    {
+      "epoch": 0.16271186440677965,
+      "grad_norm": 0.2458319216966629,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 0.3157382607460022,
+      "step": 42,
+      "token_acc": 0.8923707458363505
+    },
+    {
+      "epoch": 0.1665859564164649,
+      "grad_norm": 0.23234310746192932,
+      "learning_rate": 8.6e-06,
+      "loss": 0.32486584782600403,
+      "step": 43,
+      "token_acc": 0.8898832391328527
+    },
+    {
+      "epoch": 0.17046004842615012,
+      "grad_norm": 0.24149972200393677,
+      "learning_rate": 8.8e-06,
+      "loss": 0.3565906286239624,
+      "step": 44,
+      "token_acc": 0.8786636478836652
+    },
+    {
+      "epoch": 0.17433414043583534,
+      "grad_norm": 0.23454472422599792,
+      "learning_rate": 9e-06,
+      "loss": 0.34243613481521606,
+      "step": 45,
+      "token_acc": 0.8836981353220466
+    },
+    {
+      "epoch": 0.1782082324455206,
+      "grad_norm": 0.22611235082149506,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 0.3169807493686676,
+      "step": 46,
+      "token_acc": 0.8913642111117898
+    },
+    {
+      "epoch": 0.1820823244552058,
+      "grad_norm": 0.2332201898097992,
+      "learning_rate": 9.4e-06,
+      "loss": 0.3335682153701782,
+      "step": 47,
+      "token_acc": 0.8853106607331619
+    },
+    {
+      "epoch": 0.18595641646489103,
+      "grad_norm": 0.26498886942863464,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 0.3396722674369812,
+      "step": 48,
+      "token_acc": 0.884526213547764
+    },
+    {
+      "epoch": 0.18983050847457628,
+      "grad_norm": 0.29751622676849365,
+      "learning_rate": 9.800000000000001e-06,
+      "loss": 0.3544740676879883,
+      "step": 49,
+      "token_acc": 0.8790830507178009
+    },
+    {
+      "epoch": 0.1937046004842615,
+      "grad_norm": 0.24125243723392487,
+      "learning_rate": 1e-05,
+      "loss": 0.3444434702396393,
+      "step": 50,
+      "token_acc": 0.8833375152943368
+    },
+    {
+      "epoch": 0.19757869249394674,
+      "grad_norm": 0.23450158536434174,
+      "learning_rate": 9.999953315763929e-06,
+      "loss": 0.34759777784347534,
+      "step": 51,
+      "token_acc": 0.8809645656414854
+    },
+    {
+      "epoch": 0.20145278450363197,
+      "grad_norm": 0.24415536224842072,
+      "learning_rate": 9.999813263927483e-06,
+      "loss": 0.3302762508392334,
+      "step": 52,
+      "token_acc": 0.8872595593874244
+    },
+    {
+      "epoch": 0.20532687651331719,
+      "grad_norm": 0.23792694509029388,
+      "learning_rate": 9.999579847105947e-06,
+      "loss": 0.3057291805744171,
+      "step": 53,
+      "token_acc": 0.8958096559669801
+    },
+    {
+      "epoch": 0.20920096852300243,
+      "grad_norm": 0.24918483197689056,
+      "learning_rate": 9.999253069658074e-06,
+      "loss": 0.3550814390182495,
+      "step": 54,
+      "token_acc": 0.8789014457104403
+    },
+    {
+      "epoch": 0.21307506053268765,
+      "grad_norm": 0.24681781232357025,
+      "learning_rate": 9.99883293768601e-06,
+      "loss": 0.329832524061203,
+      "step": 55,
+      "token_acc": 0.8862846605616155
+    },
+    {
+      "epoch": 0.21694915254237288,
+      "grad_norm": 0.25197944045066833,
+      "learning_rate": 9.998319459035168e-06,
+      "loss": 0.3133784532546997,
+      "step": 56,
+      "token_acc": 0.8929686000759445
+    },
+    {
+      "epoch": 0.22082324455205812,
+      "grad_norm": 0.29595333337783813,
+      "learning_rate": 9.997712643294093e-06,
+      "loss": 0.3238765597343445,
+      "step": 57,
+      "token_acc": 0.8900121095092376
+    },
+    {
+      "epoch": 0.22469733656174334,
+      "grad_norm": 0.2436024248600006,
+      "learning_rate": 9.997012501794273e-06,
+      "loss": 0.3283236622810364,
+      "step": 58,
+      "token_acc": 0.887762605178964
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.23041026294231415,
+      "learning_rate": 9.996219047609943e-06,
+      "loss": 0.3104722797870636,
+      "step": 59,
+      "token_acc": 0.8931121325749851
+    },
+    {
+      "epoch": 0.2324455205811138,
+      "grad_norm": 0.237432062625885,
+      "learning_rate": 9.995332295557818e-06,
+      "loss": 0.30940210819244385,
+      "step": 60,
+      "token_acc": 0.8942070394423697
+    },
+    {
+      "epoch": 0.23631961259079903,
+      "grad_norm": 0.23901380598545074,
+      "learning_rate": 9.994352262196839e-06,
+      "loss": 0.32523292303085327,
+      "step": 61,
+      "token_acc": 0.8885503611348168
+    },
+    {
+      "epoch": 0.24019370460048425,
+      "grad_norm": 0.27438339591026306,
+      "learning_rate": 9.993278965827844e-06,
+      "loss": 0.3501031994819641,
+      "step": 62,
+      "token_acc": 0.8796217252529412
+    },
+    {
+      "epoch": 0.2440677966101695,
+      "grad_norm": 0.23662753403186798,
+      "learning_rate": 9.992112426493247e-06,
+      "loss": 0.32605987787246704,
+      "step": 63,
+      "token_acc": 0.8890396653634925
+    },
+    {
+      "epoch": 0.24794188861985472,
+      "grad_norm": 0.2232031375169754,
+      "learning_rate": 9.990852665976648e-06,
+      "loss": 0.3196948170661926,
+      "step": 64,
+      "token_acc": 0.8907072739748918
+    },
+    {
+      "epoch": 0.25181598062953997,
+      "grad_norm": 0.2665523886680603,
+      "learning_rate": 9.989499707802424e-06,
+      "loss": 0.33278700709342957,
+      "step": 65,
+      "token_acc": 0.8863953116150797
+    },
+    {
+      "epoch": 0.25569007263922516,
+      "grad_norm": 0.23870785534381866,
+      "learning_rate": 9.988053577235306e-06,
+      "loss": 0.351688951253891,
+      "step": 66,
+      "token_acc": 0.879823584223047
+    },
+    {
+      "epoch": 0.2595641646489104,
+      "grad_norm": 0.24755656719207764,
+      "learning_rate": 9.986514301279894e-06,
+      "loss": 0.31553030014038086,
+      "step": 67,
+      "token_acc": 0.8921622627267041
+    },
+    {
+      "epoch": 0.26343825665859566,
+      "grad_norm": 0.23198164999485016,
+      "learning_rate": 9.984881908680157e-06,
+      "loss": 0.3355843424797058,
+      "step": 68,
+      "token_acc": 0.8848336232927391
+    },
+    {
+      "epoch": 0.26731234866828085,
+      "grad_norm": 0.2461438924074173,
+      "learning_rate": 9.983156429918895e-06,
+      "loss": 0.3341342508792877,
+      "step": 69,
+      "token_acc": 0.8856444439357114
+    },
+    {
+      "epoch": 0.2711864406779661,
+      "grad_norm": 0.22579748928546906,
+      "learning_rate": 9.981337897217171e-06,
+      "loss": 0.3188900947570801,
+      "step": 70,
+      "token_acc": 0.8906781387812132
+    },
+    {
+      "epoch": 0.27506053268765135,
+      "grad_norm": 0.24103567004203796,
+      "learning_rate": 9.979426344533712e-06,
+      "loss": 0.3240354061126709,
+      "step": 71,
+      "token_acc": 0.8889305949367731
+    },
+    {
+      "epoch": 0.27893462469733654,
+      "grad_norm": 0.23146985471248627,
+      "learning_rate": 9.977421807564264e-06,
+      "loss": 0.3256258964538574,
+      "step": 72,
+      "token_acc": 0.8886470476040884
+    },
+    {
+      "epoch": 0.2828087167070218,
+      "grad_norm": 0.2992205023765564,
+      "learning_rate": 9.97532432374094e-06,
+      "loss": 0.3162704110145569,
+      "step": 73,
+      "token_acc": 0.8912486582241803
+    },
+    {
+      "epoch": 0.28668280871670704,
+      "grad_norm": 0.2314625382423401,
+      "learning_rate": 9.973133932231514e-06,
+      "loss": 0.33748123049736023,
+      "step": 74,
+      "token_acc": 0.8834313251000246
+    },
+    {
+      "epoch": 0.29055690072639223,
+      "grad_norm": 0.23197512328624725,
+      "learning_rate": 9.970850673938684e-06,
+      "loss": 0.3105667233467102,
+      "step": 75,
+      "token_acc": 0.8935043208256486
+    },
+    {
+      "epoch": 0.2944309927360775,
+      "grad_norm": 0.2275022268295288,
+      "learning_rate": 9.96847459149932e-06,
+      "loss": 0.3327932357788086,
+      "step": 76,
+      "token_acc": 0.8861917159302386
+    },
+    {
+      "epoch": 0.2983050847457627,
+      "grad_norm": 0.2508430778980255,
+      "learning_rate": 9.966005729283658e-06,
+      "loss": 0.32548677921295166,
+      "step": 77,
+      "token_acc": 0.8882381273480396
+    },
+    {
+      "epoch": 0.3021791767554479,
+      "grad_norm": 0.5134550333023071,
+      "learning_rate": 9.963444133394478e-06,
+      "loss": 0.3120523691177368,
+      "step": 78,
+      "token_acc": 0.8919503736696569
+    },
+    {
+      "epoch": 0.30605326876513317,
+      "grad_norm": 0.21315379440784454,
+      "learning_rate": 9.960789851666237e-06,
+      "loss": 0.3215460181236267,
+      "step": 79,
+      "token_acc": 0.8896002985397907
+    },
+    {
+      "epoch": 0.3099273607748184,
+      "grad_norm": 0.23902781307697296,
+      "learning_rate": 9.958042933664186e-06,
+      "loss": 0.33162713050842285,
+      "step": 80,
+      "token_acc": 0.8866171518838251
+    },
+    {
+      "epoch": 0.3138014527845036,
+      "grad_norm": 0.24128590524196625,
+      "learning_rate": 9.955203430683425e-06,
+      "loss": 0.3268725574016571,
+      "step": 81,
+      "token_acc": 0.8882163748841388
+    },
+    {
+      "epoch": 0.31767554479418886,
+      "grad_norm": 0.24751782417297363,
+      "learning_rate": 9.952271395747969e-06,
+      "loss": 0.3100839853286743,
+      "step": 82,
+      "token_acc": 0.893085253361785
+    },
+    {
+      "epoch": 0.3215496368038741,
+      "grad_norm": 0.23644764721393585,
+      "learning_rate": 9.949246883609743e-06,
+      "loss": 0.32995104789733887,
+      "step": 83,
+      "token_acc": 0.8866222032237766
+    },
+    {
+      "epoch": 0.3254237288135593,
+      "grad_norm": 0.232451930642128,
+      "learning_rate": 9.94612995074756e-06,
+      "loss": 0.31272488832473755,
+      "step": 84,
+      "token_acc": 0.8926265473810503
+    },
+    {
+      "epoch": 0.32929782082324455,
+      "grad_norm": 0.21610639989376068,
+      "learning_rate": 9.942920655366075e-06,
+      "loss": 0.302722692489624,
+      "step": 85,
+      "token_acc": 0.8952372082627079
+    },
+    {
+      "epoch": 0.3331719128329298,
+      "grad_norm": 0.24474947154521942,
+      "learning_rate": 9.939619057394687e-06,
+      "loss": 0.31238657236099243,
+      "step": 86,
+      "token_acc": 0.8932181956136864
+    },
+    {
+      "epoch": 0.337046004842615,
+      "grad_norm": 0.22313052415847778,
+      "learning_rate": 9.936225218486428e-06,
+      "loss": 0.30595749616622925,
+      "step": 87,
+      "token_acc": 0.8942476419229949
+    },
+    {
+      "epoch": 0.34092009685230024,
+      "grad_norm": 0.25018593668937683,
+      "learning_rate": 9.93273920201681e-06,
+      "loss": 0.34218600392341614,
+      "step": 88,
+      "token_acc": 0.8826220754003523
+    },
+    {
+      "epoch": 0.3447941888619855,
+      "grad_norm": 0.21603761613368988,
+      "learning_rate": 9.929161073082636e-06,
+      "loss": 0.26845768094062805,
+      "step": 89,
+      "token_acc": 0.9068716054841073
+    },
+    {
+      "epoch": 0.3486682808716707,
+      "grad_norm": 0.22996748983860016,
+      "learning_rate": 9.925490898500796e-06,
+      "loss": 0.32508569955825806,
+      "step": 90,
+      "token_acc": 0.8884358725254423
+    },
+    {
+      "epoch": 0.3525423728813559,
+      "grad_norm": 0.3635949194431305,
+      "learning_rate": 9.921728746807008e-06,
+      "loss": 0.34217730164527893,
+      "step": 91,
+      "token_acc": 0.8833008019688547
+    },
+    {
+      "epoch": 0.3564164648910412,
+      "grad_norm": 0.22128325700759888,
+      "learning_rate": 9.917874688254542e-06,
+      "loss": 0.32345396280288696,
+      "step": 92,
+      "token_acc": 0.8889643834760571
+    },
+    {
+      "epoch": 0.36029055690072637,
+      "grad_norm": 0.24601417779922485,
+      "learning_rate": 9.913928794812909e-06,
+      "loss": 0.3252776265144348,
+      "step": 93,
+      "token_acc": 0.8881070006006884
+    },
+    {
+      "epoch": 0.3641646489104116,
+      "grad_norm": 0.23473182320594788,
+      "learning_rate": 9.90989114016652e-06,
+      "loss": 0.33626118302345276,
+      "step": 94,
+      "token_acc": 0.8841867411739727
+    },
+    {
+      "epoch": 0.36803874092009686,
+      "grad_norm": 0.22333025932312012,
+      "learning_rate": 9.905761799713302e-06,
+      "loss": 0.34545931220054626,
+      "step": 95,
+      "token_acc": 0.8803537032594166
+    },
+    {
+      "epoch": 0.37191283292978206,
+      "grad_norm": 0.21172457933425903,
+      "learning_rate": 9.901540850563295e-06,
+      "loss": 0.3074107766151428,
+      "step": 96,
+      "token_acc": 0.8944196156632918
+    },
+    {
+      "epoch": 0.3757869249394673,
+      "grad_norm": 0.2134028971195221,
+      "learning_rate": 9.89722837153722e-06,
+      "loss": 0.2957490086555481,
+      "step": 97,
+      "token_acc": 0.8978778618134635
+    },
+    {
+      "epoch": 0.37966101694915255,
+      "grad_norm": 0.2610202729701996,
+      "learning_rate": 9.892824443164987e-06,
+      "loss": 0.3412560224533081,
+      "step": 98,
+      "token_acc": 0.8829380073969748
+    },
+    {
+      "epoch": 0.38353510895883774,
+      "grad_norm": 0.25488367676734924,
+      "learning_rate": 9.88832914768421e-06,
+      "loss": 0.3430347442626953,
+      "step": 99,
+      "token_acc": 0.8815459290145207
+    },
+    {
+      "epoch": 0.387409200968523,
+      "grad_norm": 0.22882606089115143,
+      "learning_rate": 9.883742569038663e-06,
+      "loss": 0.33350762724876404,
+      "step": 100,
+      "token_acc": 0.8861422500817198
+    },
+    {
+      "epoch": 0.39128329297820824,
+      "grad_norm": 0.304647833108902,
+      "learning_rate": 9.879064792876717e-06,
+      "loss": 0.31420135498046875,
+      "step": 101,
+      "token_acc": 0.8915588172822687
+    },
+    {
+      "epoch": 0.3951573849878935,
+      "grad_norm": 0.22871072590351105,
+      "learning_rate": 9.874295906549728e-06,
+      "loss": 0.3116581439971924,
+      "step": 102,
+      "token_acc": 0.8917020548921253
+    },
+    {
+      "epoch": 0.3990314769975787,
+      "grad_norm": 0.2979466915130615,
+      "learning_rate": 9.869435999110428e-06,
+      "loss": 0.3145788013935089,
+      "step": 103,
+      "token_acc": 0.8916011830301528
+    },
+    {
+      "epoch": 0.40290556900726393,
+      "grad_norm": 0.20779502391815186,
+      "learning_rate": 9.864485161311242e-06,
+      "loss": 0.3070036768913269,
+      "step": 104,
+      "token_acc": 0.8938107647266995
+    },
+    {
+      "epoch": 0.4067796610169492,
+      "grad_norm": 0.2354535311460495,
+      "learning_rate": 9.859443485602603e-06,
+      "loss": 0.32298558950424194,
+      "step": 105,
+      "token_acc": 0.8882189451059107
+    },
+    {
+      "epoch": 0.41065375302663437,
+      "grad_norm": 0.22240500152111053,
+      "learning_rate": 9.85431106613122e-06,
+      "loss": 0.3104989528656006,
+      "step": 106,
+      "token_acc": 0.8923007628162216
+    },
+    {
+      "epoch": 0.4145278450363196,
+      "grad_norm": 0.21981710195541382,
+      "learning_rate": 9.849087998738328e-06,
+      "loss": 0.3237101435661316,
+      "step": 107,
+      "token_acc": 0.8879955719309623
+    },
+    {
+      "epoch": 0.41840193704600487,
+      "grad_norm": 0.2649724781513214,
+      "learning_rate": 9.84377438095789e-06,
+      "loss": 0.323306679725647,
+      "step": 108,
+      "token_acc": 0.8889382382835521
+    },
+    {
+      "epoch": 0.42227602905569006,
+      "grad_norm": 0.2193301022052765,
+      "learning_rate": 9.838370312014783e-06,
+      "loss": 0.31488102674484253,
+      "step": 109,
+      "token_acc": 0.8910646836196473
+    },
+    {
+      "epoch": 0.4261501210653753,
+      "grad_norm": 0.21842491626739502,
+      "learning_rate": 9.832875892822937e-06,
+      "loss": 0.3206183910369873,
+      "step": 110,
+      "token_acc": 0.8890832728771944
+    },
+    {
+      "epoch": 0.43002421307506056,
+      "grad_norm": 0.2456243336200714,
+      "learning_rate": 9.827291225983458e-06,
+      "loss": 0.3201240301132202,
+      "step": 111,
+      "token_acc": 0.8904148288428204
+    },
+    {
+      "epoch": 0.43389830508474575,
+      "grad_norm": 0.21340763568878174,
+      "learning_rate": 9.821616415782708e-06,
+      "loss": 0.29961007833480835,
+      "step": 112,
+      "token_acc": 0.8965660205577574
+    },
+    {
+      "epoch": 0.437772397094431,
+      "grad_norm": 0.2308902144432068,
+      "learning_rate": 9.815851568190358e-06,
+      "loss": 0.3107410669326782,
+      "step": 113,
+      "token_acc": 0.8927536025516888
+    },
+    {
+      "epoch": 0.44164648910411625,
+      "grad_norm": 0.2292374223470688,
+      "learning_rate": 9.80999679085741e-06,
+      "loss": 0.3277205228805542,
+      "step": 114,
+      "token_acc": 0.886787084498464
+    },
+    {
+      "epoch": 0.44552058111380144,
+      "grad_norm": 0.21509671211242676,
+      "learning_rate": 9.80405219311419e-06,
+      "loss": 0.3161908984184265,
+      "step": 115,
+      "token_acc": 0.8916077261448497
+    },
+    {
+      "epoch": 0.4493946731234867,
+      "grad_norm": 0.20529279112815857,
+      "learning_rate": 9.798017885968295e-06,
+      "loss": 0.29131007194519043,
+      "step": 116,
+      "token_acc": 0.8990066361086406
+    },
+    {
+      "epoch": 0.45326876513317194,
+      "grad_norm": 0.24888373911380768,
+      "learning_rate": 9.791893982102537e-06,
+      "loss": 0.31967025995254517,
+      "step": 117,
+      "token_acc": 0.8899925908756566
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.22014780342578888,
+      "learning_rate": 9.785680595872824e-06,
+      "loss": 0.31103435158729553,
+      "step": 118,
+      "token_acc": 0.8928936680571538
+    },
+    {
+      "epoch": 0.4610169491525424,
+      "grad_norm": 0.21783359348773956,
+      "learning_rate": 9.77937784330603e-06,
+      "loss": 0.307749480009079,
+      "step": 119,
+      "token_acc": 0.8931600584652736
+    },
+    {
+      "epoch": 0.4648910411622276,
+      "grad_norm": 0.2104286551475525,
+      "learning_rate": 9.772985842097832e-06,
+      "loss": 0.31199365854263306,
+      "step": 120,
+      "token_acc": 0.8926850259294361
+    },
+    {
+      "epoch": 0.4687651331719128,
+      "grad_norm": 0.21124128997325897,
+      "learning_rate": 9.766504711610507e-06,
+      "loss": 0.3170176148414612,
+      "step": 121,
+      "token_acc": 0.8906264477918435
+    },
+    {
+      "epoch": 0.47263922518159807,
+      "grad_norm": 0.23777632415294647,
+      "learning_rate": 9.759934572870706e-06,
+      "loss": 0.3052697777748108,
+      "step": 122,
+      "token_acc": 0.894442848003123
+    },
+    {
+      "epoch": 0.4765133171912833,
+      "grad_norm": 0.2527632713317871,
+      "learning_rate": 9.753275548567192e-06,
+      "loss": 0.3045836091041565,
+      "step": 123,
+      "token_acc": 0.8951105518605069
+    },
+    {
+      "epoch": 0.4803874092009685,
+      "grad_norm": 0.20530211925506592,
+      "learning_rate": 9.74652776304855e-06,
+      "loss": 0.3366113305091858,
+      "step": 124,
+      "token_acc": 0.8836434912892324
+    },
+    {
+      "epoch": 0.48426150121065376,
+      "grad_norm": 0.26673150062561035,
+      "learning_rate": 9.739691342320866e-06,
+      "loss": 0.311764121055603,
+      "step": 125,
+      "token_acc": 0.8910826454277961
+    },
+    {
+      "epoch": 0.488135593220339,
+      "grad_norm": 0.2245185822248459,
+      "learning_rate": 9.732766414045368e-06,
+      "loss": 0.31055164337158203,
+      "step": 126,
+      "token_acc": 0.8926098098046538
+    },
+    {
+      "epoch": 0.4920096852300242,
+      "grad_norm": 0.2143883854150772,
+      "learning_rate": 9.725753107536053e-06,
+      "loss": 0.33499595522880554,
+      "step": 127,
+      "token_acc": 0.8840534260641282
+    },
+    {
+      "epoch": 0.49588377723970944,
+      "grad_norm": 0.22163285315036774,
+      "learning_rate": 9.718651553757266e-06,
+      "loss": 0.31920328736305237,
+      "step": 128,
+      "token_acc": 0.8901271163419964
+    },
+    {
+      "epoch": 0.4997578692493947,
+      "grad_norm": 0.2143898904323578,
+      "learning_rate": 9.711461885321247e-06,
+      "loss": 0.3301286995410919,
+      "step": 129,
+      "token_acc": 0.8853363916795757
+    },
+    {
+      "epoch": 0.5036319612590799,
+      "grad_norm": 0.24990734457969666,
+      "learning_rate": 9.704184236485672e-06,
+      "loss": 0.3278159201145172,
+      "step": 130,
+      "token_acc": 0.8874620923082561
+    },
+    {
+      "epoch": 0.5075060532687651,
+      "grad_norm": 0.22136539220809937,
+      "learning_rate": 9.696818743151128e-06,
+      "loss": 0.3319326937198639,
+      "step": 131,
+      "token_acc": 0.885009570455441
+    },
+    {
+      "epoch": 0.5113801452784503,
+      "grad_norm": 0.2669275999069214,
+      "learning_rate": 9.68936554285859e-06,
+      "loss": 0.3023684620857239,
+      "step": 132,
+      "token_acc": 0.8951259709956582
+    },
+    {
+      "epoch": 0.5152542372881356,
+      "grad_norm": 0.21833708882331848,
+      "learning_rate": 9.68182477478684e-06,
+      "loss": 0.3089104890823364,
+      "step": 133,
+      "token_acc": 0.8930920187299416
+    },
+    {
+      "epoch": 0.5191283292978208,
+      "grad_norm": 0.21197167038917542,
+      "learning_rate": 9.67419657974988e-06,
+      "loss": 0.3144392967224121,
+      "step": 134,
+      "token_acc": 0.8910884224709107
+    },
+    {
+      "epoch": 0.5230024213075061,
+      "grad_norm": 0.21434499323368073,
+      "learning_rate": 9.66648110019429e-06,
+      "loss": 0.3246540427207947,
+      "step": 135,
+      "token_acc": 0.8876412650671648
+    },
+    {
+      "epoch": 0.5268765133171913,
+      "grad_norm": 0.20343148708343506,
+      "learning_rate": 9.658678480196579e-06,
+      "loss": 0.315585196018219,
+      "step": 136,
+      "token_acc": 0.8905443269970013
+    },
+    {
+      "epoch": 0.5307506053268766,
+      "grad_norm": 0.23613257706165314,
+      "learning_rate": 9.650788865460487e-06,
+      "loss": 0.3131225109100342,
+      "step": 137,
+      "token_acc": 0.8912192170846405
+    },
+    {
+      "epoch": 0.5346246973365617,
+      "grad_norm": 0.4212075471878052,
+      "learning_rate": 9.642812403314272e-06,
+      "loss": 0.29884475469589233,
+      "step": 138,
+      "token_acc": 0.8966553773404051
+    },
+    {
+      "epoch": 0.538498789346247,
+      "grad_norm": 0.20193685591220856,
+      "learning_rate": 9.634749242707948e-06,
+      "loss": 0.26036083698272705,
+      "step": 139,
+      "token_acc": 0.9091038865111504
+    },
+    {
+      "epoch": 0.5423728813559322,
+      "grad_norm": 0.2208104431629181,
+      "learning_rate": 9.626599534210514e-06,
+      "loss": 0.33184394240379333,
+      "step": 140,
+      "token_acc": 0.8853617134142299
+    },
+    {
+      "epoch": 0.5462469733656174,
+      "grad_norm": 0.22493727505207062,
+      "learning_rate": 9.618363430007134e-06,
+      "loss": 0.31208667159080505,
+      "step": 141,
+      "token_acc": 0.8917024215686027
+    },
+    {
+      "epoch": 0.5501210653753027,
+      "grad_norm": 0.23963193595409393,
+      "learning_rate": 9.610041083896304e-06,
+      "loss": 0.33588868379592896,
+      "step": 142,
+      "token_acc": 0.883973627021253
+    },
+    {
+      "epoch": 0.553995157384988,
+      "grad_norm": 0.21784453094005585,
+      "learning_rate": 9.60163265128697e-06,
+      "loss": 0.3231375813484192,
+      "step": 143,
+      "token_acc": 0.8887875239014834
+    },
+    {
+      "epoch": 0.5578692493946731,
+      "grad_norm": 0.22835847735404968,
+      "learning_rate": 9.593138289195634e-06,
+      "loss": 0.3210199773311615,
+      "step": 144,
+      "token_acc": 0.8890582816354493
+    },
+    {
+      "epoch": 0.5617433414043583,
+      "grad_norm": 0.2136555314064026,
+      "learning_rate": 9.584558156243418e-06,
+      "loss": 0.3372665047645569,
+      "step": 145,
+      "token_acc": 0.8839793357706921
+    },
+    {
+      "epoch": 0.5656174334140436,
+      "grad_norm": 0.20598500967025757,
+      "learning_rate": 9.575892412653102e-06,
+      "loss": 0.30844664573669434,
+      "step": 146,
+      "token_acc": 0.8926156654585412
+    },
+    {
+      "epoch": 0.5694915254237288,
+      "grad_norm": 0.2522714138031006,
+      "learning_rate": 9.567141220246136e-06,
+      "loss": 0.36702272295951843,
+      "step": 147,
+      "token_acc": 0.8734296301671142
+    },
+    {
+      "epoch": 0.5733656174334141,
+      "grad_norm": 0.21975038945674896,
+      "learning_rate": 9.55830474243961e-06,
+      "loss": 0.32784411311149597,
+      "step": 148,
+      "token_acc": 0.8871756189192851
+    },
+    {
+      "epoch": 0.5772397094430993,
+      "grad_norm": 0.21233901381492615,
+      "learning_rate": 9.549383144243213e-06,
+      "loss": 0.2944122850894928,
+      "step": 149,
+      "token_acc": 0.8987453672884691
+    },
+    {
+      "epoch": 0.5811138014527845,
+      "grad_norm": 0.2199799120426178,
+      "learning_rate": 9.540376592256142e-06,
+      "loss": 0.3299463987350464,
+      "step": 150,
+      "token_acc": 0.8859144839374592
+    },
+    {
+      "epoch": 0.5849878934624697,
+      "grad_norm": 0.19698019325733185,
+      "learning_rate": 9.531285254663997e-06,
+      "loss": 0.3030051589012146,
+      "step": 151,
+      "token_acc": 0.8951707294894029
+    },
+    {
+      "epoch": 0.588861985472155,
+      "grad_norm": 0.22306668758392334,
+      "learning_rate": 9.522109301235637e-06,
+      "loss": 0.29752516746520996,
+      "step": 152,
+      "token_acc": 0.8966012679857996
+    },
+    {
+      "epoch": 0.5927360774818402,
+      "grad_norm": 0.21317337453365326,
+      "learning_rate": 9.512848903320017e-06,
+      "loss": 0.3052118122577667,
+      "step": 153,
+      "token_acc": 0.8944324633814714
+    },
+    {
+      "epoch": 0.5966101694915255,
+      "grad_norm": 0.2120915800333023,
+      "learning_rate": 9.503504233842973e-06,
+      "loss": 0.29761528968811035,
+      "step": 154,
+      "token_acc": 0.8966406260468731
+    },
+    {
+      "epoch": 0.6004842615012107,
+      "grad_norm": 0.23525090515613556,
+      "learning_rate": 9.494075467304007e-06,
+      "loss": 0.3034532070159912,
+      "step": 155,
+      "token_acc": 0.8944926637860167
+    },
+    {
+      "epoch": 0.6043583535108958,
+      "grad_norm": 0.2095353752374649,
+      "learning_rate": 9.484562779773027e-06,
+      "loss": 0.2903788089752197,
+      "step": 156,
+      "token_acc": 0.8990560027078014
+    },
+    {
+      "epoch": 0.6082324455205811,
+      "grad_norm": 0.23741677403450012,
+      "learning_rate": 9.474966348887055e-06,
+      "loss": 0.31467512249946594,
+      "step": 157,
+      "token_acc": 0.8904583329757747
+    },
+    {
+      "epoch": 0.6121065375302663,
+      "grad_norm": 0.2259555608034134,
+      "learning_rate": 9.465286353846905e-06,
+      "loss": 0.3404577374458313,
+      "step": 158,
+      "token_acc": 0.8826165622063978
+    },
+    {
+      "epoch": 0.6159806295399516,
+      "grad_norm": 0.2183879017829895,
+      "learning_rate": 9.455522975413846e-06,
+      "loss": 0.2766571640968323,
+      "step": 159,
+      "token_acc": 0.9038809421418853
+    },
+    {
+      "epoch": 0.6198547215496368,
+      "grad_norm": 0.22651784121990204,
+      "learning_rate": 9.445676395906226e-06,
+      "loss": 0.29638129472732544,
+      "step": 160,
+      "token_acc": 0.8970113168662065
+    },
+    {
+      "epoch": 0.6237288135593221,
+      "grad_norm": 0.22088395059108734,
+      "learning_rate": 9.435746799196061e-06,
+      "loss": 0.3023075759410858,
+      "step": 161,
+      "token_acc": 0.8946665593674712
+    },
+    {
+      "epoch": 0.6276029055690072,
+      "grad_norm": 0.21526560187339783,
+      "learning_rate": 9.425734370705606e-06,
+      "loss": 0.28661438822746277,
+      "step": 162,
+      "token_acc": 0.9002787847728345
+    },
+    {
+      "epoch": 0.6314769975786925,
+      "grad_norm": 0.23334769904613495,
+      "learning_rate": 9.415639297403891e-06,
+      "loss": 0.31685301661491394,
+      "step": 163,
+      "token_acc": 0.890886748080584
+    },
+    {
+      "epoch": 0.6353510895883777,
+      "grad_norm": 0.200165793299675,
+      "learning_rate": 9.40546176780323e-06,
+      "loss": 0.30981898307800293,
+      "step": 164,
+      "token_acc": 0.8924871164982372
+    },
+    {
+      "epoch": 0.639225181598063,
+      "grad_norm": 0.20800836384296417,
+      "learning_rate": 9.395201971955701e-06,
+      "loss": 0.3162352740764618,
+      "step": 165,
+      "token_acc": 0.8910434805285766
+    },
+    {
+      "epoch": 0.6430992736077482,
+      "grad_norm": 0.20923736691474915,
+      "learning_rate": 9.384860101449598e-06,
+      "loss": 0.32208406925201416,
+      "step": 166,
+      "token_acc": 0.8880633815629819
+    },
+    {
+      "epoch": 0.6469733656174335,
+      "grad_norm": 0.1986808031797409,
+      "learning_rate": 9.374436349405847e-06,
+      "loss": 0.28397923707962036,
+      "step": 167,
+      "token_acc": 0.9012052212352475
+    },
+    {
+      "epoch": 0.6508474576271186,
+      "grad_norm": 0.21215273439884186,
+      "learning_rate": 9.36393091047441e-06,
+      "loss": 0.3066609799861908,
+      "step": 168,
+      "token_acc": 0.894593303584187
+    },
+    {
+      "epoch": 0.6547215496368038,
+      "grad_norm": 0.20804037153720856,
+      "learning_rate": 9.353343980830644e-06,
+      "loss": 0.3097017705440521,
+      "step": 169,
+      "token_acc": 0.8926308156125992
+    },
+    {
+      "epoch": 0.6585956416464891,
+      "grad_norm": 0.20328834652900696,
+      "learning_rate": 9.342675758171638e-06,
+      "loss": 0.3010105490684509,
+      "step": 170,
+      "token_acc": 0.8950560660129195
+    },
+    {
+      "epoch": 0.6624697336561743,
+      "grad_norm": 0.2051060050725937,
+      "learning_rate": 9.331926441712522e-06,
+      "loss": 0.3019353151321411,
+      "step": 171,
+      "token_acc": 0.8949745506999682
+    },
+    {
+      "epoch": 0.6663438256658596,
+      "grad_norm": 0.24043123424053192,
+      "learning_rate": 9.32109623218275e-06,
+      "loss": 0.3116442859172821,
+      "step": 172,
+      "token_acc": 0.8915558784861239
+    },
+    {
+      "epoch": 0.6702179176755448,
+      "grad_norm": 0.21520181000232697,
+      "learning_rate": 9.310185331822338e-06,
+      "loss": 0.31186142563819885,
+      "step": 173,
+      "token_acc": 0.8917585320277845
+    },
+    {
+      "epoch": 0.67409200968523,
+      "grad_norm": 0.21344298124313354,
+      "learning_rate": 9.299193944378112e-06,
+      "loss": 0.3273160755634308,
+      "step": 174,
+      "token_acc": 0.886418268420563
+    },
+    {
+      "epoch": 0.6779661016949152,
+      "grad_norm": 0.20224156975746155,
+      "learning_rate": 9.28812227509988e-06,
+      "loss": 0.31608837842941284,
+      "step": 175,
+      "token_acc": 0.8894536504933755
+    },
+    {
+      "epoch": 0.6818401937046005,
+      "grad_norm": 0.2154257595539093,
+      "learning_rate": 9.27697053073661e-06,
+      "loss": 0.34367692470550537,
+      "step": 176,
+      "token_acc": 0.8811017511710314
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.22003678977489471,
+      "learning_rate": 9.26573891953257e-06,
+      "loss": 0.3205263018608093,
+      "step": 177,
+      "token_acc": 0.8893210947921869
+    },
+    {
+      "epoch": 0.689588377723971,
+      "grad_norm": 0.21449677646160126,
+      "learning_rate": 9.254427651223434e-06,
+      "loss": 0.28666430711746216,
+      "step": 178,
+      "token_acc": 0.9003720788020833
+    },
+    {
+      "epoch": 0.6934624697336562,
+      "grad_norm": 0.22110596299171448,
+      "learning_rate": 9.243036937032373e-06,
+      "loss": 0.3156067728996277,
+      "step": 179,
+      "token_acc": 0.8902597783694092
+    },
+    {
+      "epoch": 0.6973365617433414,
+      "grad_norm": 0.19700580835342407,
+      "learning_rate": 9.2315669896661e-06,
+      "loss": 0.28897273540496826,
+      "step": 180,
+      "token_acc": 0.8994499889336349
+    },
+    {
+      "epoch": 0.7012106537530266,
+      "grad_norm": 0.21460606157779694,
+      "learning_rate": 9.220018023310908e-06,
+      "loss": 0.31268295645713806,
+      "step": 181,
+      "token_acc": 0.8918378520876847
+    },
+    {
+      "epoch": 0.7050847457627119,
+      "grad_norm": 0.21692436933517456,
+      "learning_rate": 9.208390253628667e-06,
+      "loss": 0.28844964504241943,
+      "step": 182,
+      "token_acc": 0.8997311485616448
+    },
+    {
+      "epoch": 0.7089588377723971,
+      "grad_norm": 0.201703280210495,
+      "learning_rate": 9.196683897752794e-06,
+      "loss": 0.32861441373825073,
+      "step": 183,
+      "token_acc": 0.8854774295445417
+    },
+    {
+      "epoch": 0.7128329297820823,
+      "grad_norm": 3.976747751235962,
+      "learning_rate": 9.184899174284201e-06,
+      "loss": 0.33475255966186523,
+      "step": 184,
+      "token_acc": 0.8836819705392365
+    },
+    {
+      "epoch": 0.7167070217917676,
+      "grad_norm": 0.24247053265571594,
+      "learning_rate": 9.173036303287215e-06,
+      "loss": 0.3366454243659973,
+      "step": 185,
+      "token_acc": 0.8833432089980459
+    },
+    {
+      "epoch": 0.7205811138014527,
+      "grad_norm": 0.2282845675945282,
+      "learning_rate": 9.16109550628546e-06,
+      "loss": 0.2812536656856537,
+      "step": 186,
+      "token_acc": 0.9027706860502607
+    },
+    {
+      "epoch": 0.724455205811138,
+      "grad_norm": 0.2282128632068634,
+      "learning_rate": 9.149077006257734e-06,
+      "loss": 0.3136906027793884,
+      "step": 187,
+      "token_acc": 0.8912536222754189
+    },
+    {
+      "epoch": 0.7283292978208232,
+      "grad_norm": 0.20751290023326874,
+      "learning_rate": 9.136981027633834e-06,
+      "loss": 0.29636135697364807,
+      "step": 188,
+      "token_acc": 0.8974463288547996
+    },
+    {
+      "epoch": 0.7322033898305085,
+      "grad_norm": 0.23192144930362701,
+      "learning_rate": 9.124807796290366e-06,
+      "loss": 0.3046882152557373,
+      "step": 189,
+      "token_acc": 0.8943812414560115
+    },
+    {
+      "epoch": 0.7360774818401937,
+      "grad_norm": 0.221333310008049,
+      "learning_rate": 9.112557539546535e-06,
+      "loss": 0.32960376143455505,
+      "step": 190,
+      "token_acc": 0.8860915000599271
+    },
+    {
+      "epoch": 0.739951573849879,
+      "grad_norm": 0.1981872171163559,
+      "learning_rate": 9.100230486159893e-06,
+      "loss": 0.32151421904563904,
+      "step": 191,
+      "token_acc": 0.888598638535205
+    },
+    {
+      "epoch": 0.7438256658595641,
+      "grad_norm": 0.2172573357820511,
+      "learning_rate": 9.087826866322065e-06,
+      "loss": 0.3255336880683899,
+      "step": 192,
+      "token_acc": 0.8864367509340579
+    },
+    {
+      "epoch": 0.7476997578692494,
+      "grad_norm": 0.21215571463108063,
+      "learning_rate": 9.075346911654456e-06,
+      "loss": 0.30505236983299255,
+      "step": 193,
+      "token_acc": 0.8936060377931436
+    },
+    {
+      "epoch": 0.7515738498789346,
+      "grad_norm": 0.21355277299880981,
+      "learning_rate": 9.062790855203932e-06,
+      "loss": 0.3349328637123108,
+      "step": 194,
+      "token_acc": 0.8847527625851099
+    },
+    {
+      "epoch": 0.7554479418886199,
+      "grad_norm": 0.20415301620960236,
+      "learning_rate": 9.050158931438451e-06,
+      "loss": 0.3010273873806,
+      "step": 195,
+      "token_acc": 0.8946901896914337
+    },
+    {
+      "epoch": 0.7593220338983051,
+      "grad_norm": 0.2100018560886383,
+      "learning_rate": 9.037451376242696e-06,
+      "loss": 0.3295148015022278,
+      "step": 196,
+      "token_acc": 0.8861214255925314
+    },
+    {
+      "epoch": 0.7631961259079904,
+      "grad_norm": 0.21248096227645874,
+      "learning_rate": 9.024668426913671e-06,
+      "loss": 0.2901475727558136,
+      "step": 197,
+      "token_acc": 0.8984891018269412
+    },
+    {
+      "epoch": 0.7670702179176755,
+      "grad_norm": 0.20735451579093933,
+      "learning_rate": 9.011810322156269e-06,
+      "loss": 0.3123668134212494,
+      "step": 198,
+      "token_acc": 0.8911118341790296
+    },
+    {
+      "epoch": 0.7709443099273607,
+      "grad_norm": 0.2119433879852295,
+      "learning_rate": 8.998877302078803e-06,
+      "loss": 0.30766892433166504,
+      "step": 199,
+      "token_acc": 0.8930650097673094
+    },
+    {
+      "epoch": 0.774818401937046,
+      "grad_norm": 0.20151817798614502,
+      "learning_rate": 8.985869608188545e-06,
+      "loss": 0.294528067111969,
+      "step": 200,
+      "token_acc": 0.8973507748438794
+    },
+    {
+      "epoch": 0.7786924939467312,
+      "grad_norm": 0.20979715883731842,
+      "learning_rate": 8.97278748338719e-06,
+      "loss": 0.3116077184677124,
+      "step": 201,
+      "token_acc": 0.8916578293780434
+    },
+    {
+      "epoch": 0.7825665859564165,
+      "grad_norm": 0.21114560961723328,
+      "learning_rate": 8.95963117196634e-06,
+      "loss": 0.31117022037506104,
+      "step": 202,
+      "token_acc": 0.8922739117136779
+    },
+    {
+      "epoch": 0.7864406779661017,
+      "grad_norm": 0.2028111070394516,
+      "learning_rate": 8.946400919602933e-06,
+      "loss": 0.2925041913986206,
+      "step": 203,
+      "token_acc": 0.8979599612123477
+    },
+    {
+      "epoch": 0.790314769975787,
+      "grad_norm": 0.19873376190662384,
+      "learning_rate": 8.933096973354665e-06,
+      "loss": 0.3335387706756592,
+      "step": 204,
+      "token_acc": 0.8845781124549695
+    },
+    {
+      "epoch": 0.7941888619854721,
+      "grad_norm": 0.20865830779075623,
+      "learning_rate": 8.919719581655357e-06,
+      "loss": 0.3048374652862549,
+      "step": 205,
+      "token_acc": 0.8941424666394205
+    },
+    {
+      "epoch": 0.7980629539951574,
+      "grad_norm": 0.21847450733184814,
+      "learning_rate": 8.906268994310339e-06,
+      "loss": 0.30148929357528687,
+      "step": 206,
+      "token_acc": 0.8948231645494126
+    },
+    {
+      "epoch": 0.8019370460048426,
+      "grad_norm": 0.23447921872138977,
+      "learning_rate": 8.892745462491763e-06,
+      "loss": 0.3076891005039215,
+      "step": 207,
+      "token_acc": 0.8940680143003497
+    },
+    {
+      "epoch": 0.8058111380145279,
+      "grad_norm": 0.2047218531370163,
+      "learning_rate": 8.879149238733932e-06,
+      "loss": 0.2903471291065216,
+      "step": 208,
+      "token_acc": 0.8996930000967329
+    },
+    {
+      "epoch": 0.8096852300242131,
+      "grad_norm": 0.3560882806777954,
+      "learning_rate": 8.865480576928578e-06,
+      "loss": 0.2734353840351105,
+      "step": 209,
+      "token_acc": 0.9038816908230364
+    },
+    {
+      "epoch": 0.8135593220338984,
+      "grad_norm": 0.22588837146759033,
+      "learning_rate": 8.851739732320109e-06,
+      "loss": 0.30820316076278687,
+      "step": 210,
+      "token_acc": 0.8928903081404425
+    },
+    {
+      "epoch": 0.8174334140435835,
+      "grad_norm": 0.19928814470767975,
+      "learning_rate": 8.83792696150086e-06,
+      "loss": 0.30705487728118896,
+      "step": 211,
+      "token_acc": 0.8931717351449738
+    },
+    {
+      "epoch": 0.8213075060532687,
+      "grad_norm": 0.23134565353393555,
+      "learning_rate": 8.824042522406295e-06,
+      "loss": 0.3144133687019348,
+      "step": 212,
+      "token_acc": 0.8904542748607169
+    },
+    {
+      "epoch": 0.825181598062954,
+      "grad_norm": 0.20952780544757843,
+      "learning_rate": 8.810086674310184e-06,
+      "loss": 0.3166520595550537,
+      "step": 213,
+      "token_acc": 0.8902617260259249
+    },
+    {
+      "epoch": 0.8290556900726392,
+      "grad_norm": 0.21133121848106384,
+      "learning_rate": 8.796059677819773e-06,
+      "loss": 0.31384018063545227,
+      "step": 214,
+      "token_acc": 0.8909493414116798
+    },
+    {
+      "epoch": 0.8329297820823245,
+      "grad_norm": 0.3206462264060974,
+      "learning_rate": 8.781961794870903e-06,
+      "loss": 0.30939990282058716,
+      "step": 215,
+      "token_acc": 0.8926290243396312
+    },
+    {
+      "epoch": 0.8368038740920097,
+      "grad_norm": 0.21380406618118286,
+      "learning_rate": 8.767793288723137e-06,
+      "loss": 0.3126541078090668,
+      "step": 216,
+      "token_acc": 0.8918149018414423
+    },
+    {
+      "epoch": 0.8406779661016949,
+      "grad_norm": 0.2241922914981842,
+      "learning_rate": 8.753554423954828e-06,
+      "loss": 0.32906076312065125,
+      "step": 217,
+      "token_acc": 0.8866828065863777
+    },
+    {
+      "epoch": 0.8445520581113801,
+      "grad_norm": 0.19776619970798492,
+      "learning_rate": 8.739245466458187e-06,
+      "loss": 0.28062158823013306,
+      "step": 218,
+      "token_acc": 0.9022684784065322
+    },
+    {
+      "epoch": 0.8484261501210654,
+      "grad_norm": 0.2141999900341034,
+      "learning_rate": 8.72486668343431e-06,
+      "loss": 0.3276277184486389,
+      "step": 219,
+      "token_acc": 0.8861141792995992
+    },
+    {
+      "epoch": 0.8523002421307506,
+      "grad_norm": 0.2332129180431366,
+      "learning_rate": 8.7104183433882e-06,
+      "loss": 0.3168509304523468,
+      "step": 220,
+      "token_acc": 0.8899989570826125
+    },
+    {
+      "epoch": 0.8561743341404359,
+      "grad_norm": 0.2141677886247635,
+      "learning_rate": 8.695900716123744e-06,
+      "loss": 0.3259914219379425,
+      "step": 221,
+      "token_acc": 0.8866733094194235
+    },
+    {
+      "epoch": 0.8600484261501211,
+      "grad_norm": 0.20929858088493347,
+      "learning_rate": 8.681314072738678e-06,
+      "loss": 0.2776751220226288,
+      "step": 222,
+      "token_acc": 0.9029569916163804
+    },
+    {
+      "epoch": 0.8639225181598063,
+      "grad_norm": 0.26802197098731995,
+      "learning_rate": 8.666658685619523e-06,
+      "loss": 0.3192378282546997,
+      "step": 223,
+      "token_acc": 0.8888524656782731
+    },
+    {
+      "epoch": 0.8677966101694915,
+      "grad_norm": 0.19303195178508759,
+      "learning_rate": 8.651934828436497e-06,
+      "loss": 0.2820873260498047,
+      "step": 224,
+      "token_acc": 0.9010663601046539
+    },
+    {
+      "epoch": 0.8716707021791767,
+      "grad_norm": 0.20784462988376617,
+      "learning_rate": 8.637142776138415e-06,
+      "loss": 0.2850268483161926,
+      "step": 225,
+      "token_acc": 0.9003609394726915
+    },
+    {
+      "epoch": 0.875544794188862,
+      "grad_norm": 0.2194257229566574,
+      "learning_rate": 8.622282804947537e-06,
+      "loss": 0.31484997272491455,
+      "step": 226,
+      "token_acc": 0.8909253202507496
+    },
+    {
+      "epoch": 0.8794188861985472,
+      "grad_norm": 0.21197804808616638,
+      "learning_rate": 8.607355192354425e-06,
+      "loss": 0.3072202801704407,
+      "step": 227,
+      "token_acc": 0.8929364556285221
+    },
+    {
+      "epoch": 0.8832929782082325,
+      "grad_norm": 0.19514977931976318,
+      "learning_rate": 8.592360217112759e-06,
+      "loss": 0.31343895196914673,
+      "step": 228,
+      "token_acc": 0.8909144611151198
+    },
+    {
+      "epoch": 0.8871670702179176,
+      "grad_norm": 0.2198445200920105,
+      "learning_rate": 8.57729815923412e-06,
+      "loss": 0.31176120042800903,
+      "step": 229,
+      "token_acc": 0.8916788161998124
+    },
+    {
+      "epoch": 0.8910411622276029,
+      "grad_norm": 0.20297633111476898,
+      "learning_rate": 8.562169299982776e-06,
+      "loss": 0.30840498208999634,
+      "step": 230,
+      "token_acc": 0.8921534903182912
+    },
+    {
+      "epoch": 0.8949152542372881,
+      "grad_norm": 0.21356205642223358,
+      "learning_rate": 8.546973921870421e-06,
+      "loss": 0.3210839629173279,
+      "step": 231,
+      "token_acc": 0.8882864775840541
+    },
+    {
+      "epoch": 0.8987893462469734,
+      "grad_norm": 0.21405935287475586,
+      "learning_rate": 8.531712308650904e-06,
+      "loss": 0.3006952702999115,
+      "step": 232,
+      "token_acc": 0.8953128142705267
+    },
+    {
+      "epoch": 0.9026634382566586,
+      "grad_norm": 0.21220295131206512,
+      "learning_rate": 8.516384745314926e-06,
+      "loss": 0.33272668719291687,
+      "step": 233,
+      "token_acc": 0.8845533899027282
+    },
+    {
+      "epoch": 0.9065375302663439,
+      "grad_norm": 0.19546008110046387,
+      "learning_rate": 8.50099151808472e-06,
+      "loss": 0.26581257581710815,
+      "step": 234,
+      "token_acc": 0.9067262813046539
+    },
+    {
+      "epoch": 0.910411622276029,
+      "grad_norm": 0.2057773917913437,
+      "learning_rate": 8.485532914408712e-06,
+      "loss": 0.2936754524707794,
+      "step": 235,
+      "token_acc": 0.8980145512690381
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.21968601644039154,
+      "learning_rate": 8.470009222956138e-06,
+      "loss": 0.2990136742591858,
+      "step": 236,
+      "token_acc": 0.8944779048351311
+    },
+    {
+      "epoch": 0.9181598062953995,
+      "grad_norm": 0.22149494290351868,
+      "learning_rate": 8.45442073361167e-06,
+      "loss": 0.29907599091529846,
+      "step": 237,
+      "token_acc": 0.8953804266415489
+    },
+    {
+      "epoch": 0.9220338983050848,
+      "grad_norm": 0.18807418644428253,
+      "learning_rate": 8.438767737469995e-06,
+      "loss": 0.2596169412136078,
+      "step": 238,
+      "token_acc": 0.9094668271985952
+    },
+    {
+      "epoch": 0.92590799031477,
+      "grad_norm": 0.2053857445716858,
+      "learning_rate": 8.42305052683038e-06,
+      "loss": 0.320443719625473,
+      "step": 239,
+      "token_acc": 0.8882472950063495
+    },
+    {
+      "epoch": 0.9297820823244553,
+      "grad_norm": 0.19474725425243378,
+      "learning_rate": 8.407269395191216e-06,
+      "loss": 0.29054853320121765,
+      "step": 240,
+      "token_acc": 0.8986681898213217
+    },
+    {
+      "epoch": 0.9336561743341404,
+      "grad_norm": 0.22415153682231903,
+      "learning_rate": 8.391424637244528e-06,
+      "loss": 0.29720863699913025,
+      "step": 241,
+      "token_acc": 0.8967865758573351
+    },
+    {
+      "epoch": 0.9375302663438256,
+      "grad_norm": 0.20295462012290955,
+      "learning_rate": 8.375516548870489e-06,
+      "loss": 0.3213497996330261,
+      "step": 242,
+      "token_acc": 0.8888211973402874
+    },
+    {
+      "epoch": 0.9414043583535109,
+      "grad_norm": 0.235239177942276,
+      "learning_rate": 8.359545427131876e-06,
+      "loss": 0.31140708923339844,
+      "step": 243,
+      "token_acc": 0.8917541696945803
+    },
+    {
+      "epoch": 0.9452784503631961,
+      "grad_norm": 0.21419954299926758,
+      "learning_rate": 8.343511570268541e-06,
+      "loss": 0.3142154812812805,
+      "step": 244,
+      "token_acc": 0.890589961402836
+    },
+    {
+      "epoch": 0.9491525423728814,
+      "grad_norm": 0.20498663187026978,
+      "learning_rate": 8.327415277691824e-06,
+      "loss": 0.3464815020561218,
+      "step": 245,
+      "token_acc": 0.8797665540392294
+    },
+    {
+      "epoch": 0.9530266343825666,
+      "grad_norm": 0.20611073076725006,
+      "learning_rate": 8.311256849978974e-06,
+      "loss": 0.31497207283973694,
+      "step": 246,
+      "token_acc": 0.889790752866034
+    },
+    {
+      "epoch": 0.9569007263922518,
+      "grad_norm": 0.21447882056236267,
+      "learning_rate": 8.295036588867533e-06,
+      "loss": 0.28588759899139404,
+      "step": 247,
+      "token_acc": 0.8993494375908707
+    },
+    {
+      "epoch": 0.960774818401937,
+      "grad_norm": 0.21430622041225433,
+      "learning_rate": 8.278754797249702e-06,
+      "loss": 0.3209206461906433,
+      "step": 248,
+      "token_acc": 0.8878057052632179
+    },
+    {
+      "epoch": 0.9646489104116223,
+      "grad_norm": 0.1971716433763504,
+      "learning_rate": 8.262411779166681e-06,
+      "loss": 0.29577910900115967,
+      "step": 249,
+      "token_acc": 0.8970768255184925
+    },
+    {
+      "epoch": 0.9685230024213075,
+      "grad_norm": 0.20728042721748352,
+      "learning_rate": 8.246007839802997e-06,
+      "loss": 0.3149109482765198,
+      "step": 250,
+      "token_acc": 0.8904120076852685
+    },
+    {
+      "epoch": 0.9723970944309928,
+      "grad_norm": 0.23157289624214172,
+      "learning_rate": 8.229543285480797e-06,
+      "loss": 0.3057391047477722,
+      "step": 251,
+      "token_acc": 0.8943966929583815
+    },
+    {
+      "epoch": 0.976271186440678,
+      "grad_norm": 0.21818409860134125,
+      "learning_rate": 8.213018423654144e-06,
+      "loss": 0.3090881109237671,
+      "step": 252,
+      "token_acc": 0.8931029437419457
+    },
+    {
+      "epoch": 0.9801452784503631,
+      "grad_norm": 0.20345434546470642,
+      "learning_rate": 8.196433562903252e-06,
+      "loss": 0.2966330051422119,
+      "step": 253,
+      "token_acc": 0.8959465166900704
+    },
+    {
+      "epoch": 0.9840193704600484,
+      "grad_norm": 0.203868567943573,
+      "learning_rate": 8.179789012928747e-06,
+      "loss": 0.2893424928188324,
+      "step": 254,
+      "token_acc": 0.8989887993032385
+    },
+    {
+      "epoch": 0.9878934624697336,
+      "grad_norm": 0.20835842192173004,
+      "learning_rate": 8.163085084545867e-06,
+      "loss": 0.29561957716941833,
+      "step": 255,
+      "token_acc": 0.897130295078995
+    },
+    {
+      "epoch": 0.9917675544794189,
+      "grad_norm": 0.2602974772453308,
+      "learning_rate": 8.146322089678668e-06,
+      "loss": 0.33309951424598694,
+      "step": 256,
+      "token_acc": 0.8842519179704944
+    },
+    {
+      "epoch": 0.9956416464891041,
+      "grad_norm": 0.1993730664253235,
+      "learning_rate": 8.129500341354192e-06,
+      "loss": 0.32513946294784546,
+      "step": 257,
+      "token_acc": 0.8869922494628838
+    },
+    {
+      "epoch": 0.9995157384987894,
+      "grad_norm": 0.2033330649137497,
+      "learning_rate": 8.11262015369663e-06,
+      "loss": 0.29512181878089905,
+      "step": 258,
+      "token_acc": 0.8968425014801387
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6673643589019775,
+      "learning_rate": 8.095681841921441e-06,
+      "loss": 0.28728920221328735,
+      "step": 259,
+      "token_acc": 0.9003083713758805
+    },
+    {
+      "epoch": 1.0038740920096851,
+      "grad_norm": 0.32744893431663513,
+      "learning_rate": 8.07868572232949e-06,
+      "loss": 0.269972562789917,
+      "step": 260,
+      "token_acc": 0.9038492097273063
+    },
+    {
+      "epoch": 1.0077481840193705,
+      "grad_norm": 0.2596898376941681,
+      "learning_rate": 8.061632112301122e-06,
+      "loss": 0.2655790150165558,
+      "step": 261,
+      "token_acc": 0.9053338855906853
+    },
+    {
+      "epoch": 1.0116222760290556,
+      "grad_norm": 0.2612839639186859,
+      "learning_rate": 8.044521330290235e-06,
+      "loss": 0.2887282967567444,
+      "step": 262,
+      "token_acc": 0.8971828029711167
+    },
+    {
+      "epoch": 1.015496368038741,
+      "grad_norm": 0.2769652009010315,
+      "learning_rate": 8.027353695818345e-06,
+      "loss": 0.26126527786254883,
+      "step": 263,
+      "token_acc": 0.9065780969019781
+    },
+    {
+      "epoch": 1.0193704600484261,
+      "grad_norm": 0.27929142117500305,
+      "learning_rate": 8.010129529468614e-06,
+      "loss": 0.27868735790252686,
+      "step": 264,
+      "token_acc": 0.9001419249114798
+    },
+    {
+      "epoch": 1.0232445520581113,
+      "grad_norm": 0.23997750878334045,
+      "learning_rate": 7.992849152879857e-06,
+      "loss": 0.2831759750843048,
+      "step": 265,
+      "token_acc": 0.899304001670737
+    },
+    {
+      "epoch": 1.0271186440677966,
+      "grad_norm": 0.25313815474510193,
+      "learning_rate": 7.97551288874055e-06,
+      "loss": 0.27934202551841736,
+      "step": 266,
+      "token_acc": 0.9004498805562496
+    },
+    {
+      "epoch": 1.0309927360774818,
+      "grad_norm": 0.23287494480609894,
+      "learning_rate": 7.95812106078279e-06,
+      "loss": 0.26112881302833557,
+      "step": 267,
+      "token_acc": 0.9065508038300509
+    },
+    {
+      "epoch": 1.0348668280871671,
+      "grad_norm": 0.22660091519355774,
+      "learning_rate": 7.940673993776258e-06,
+      "loss": 0.2504875063896179,
+      "step": 268,
+      "token_acc": 0.9097140867981872
+    },
+    {
+      "epoch": 1.0387409200968523,
+      "grad_norm": 0.2266615480184555,
+      "learning_rate": 7.923172013522153e-06,
+      "loss": 0.25760790705680847,
+      "step": 269,
+      "token_acc": 0.9073963735109954
+    },
+    {
+      "epoch": 1.0426150121065376,
+      "grad_norm": 0.22593924403190613,
+      "learning_rate": 7.905615446847107e-06,
+      "loss": 0.28686419129371643,
+      "step": 270,
+      "token_acc": 0.8976161305002275
+    },
+    {
+      "epoch": 1.0464891041162228,
+      "grad_norm": 0.2425071895122528,
+      "learning_rate": 7.888004621597079e-06,
+      "loss": 0.2573948800563812,
+      "step": 271,
+      "token_acc": 0.907380557815819
+    },
+    {
+      "epoch": 1.050363196125908,
+      "grad_norm": 0.23996935784816742,
+      "learning_rate": 7.87033986663124e-06,
+      "loss": 0.2808932065963745,
+      "step": 272,
+      "token_acc": 0.8994914728045711
+    },
+    {
+      "epoch": 1.0542372881355933,
+      "grad_norm": 0.25931164622306824,
+      "learning_rate": 7.852621511815825e-06,
+      "loss": 0.26375657320022583,
+      "step": 273,
+      "token_acc": 0.9051297163863579
+    },
+    {
+      "epoch": 1.0581113801452784,
+      "grad_norm": 0.20594951510429382,
+      "learning_rate": 7.834849888017979e-06,
+      "loss": 0.23789554834365845,
+      "step": 274,
+      "token_acc": 0.9142479611743739
+    },
+    {
+      "epoch": 1.0619854721549637,
+      "grad_norm": 0.23315519094467163,
+      "learning_rate": 7.817025327099574e-06,
+      "loss": 0.24684631824493408,
+      "step": 275,
+      "token_acc": 0.9110874200426439
+    },
+    {
+      "epoch": 1.0658595641646489,
+      "grad_norm": 0.2189839482307434,
+      "learning_rate": 7.799148161911013e-06,
+      "loss": 0.2684437334537506,
+      "step": 276,
+      "token_acc": 0.9041172254519392
+    },
+    {
+      "epoch": 1.0697336561743342,
+      "grad_norm": 0.21298226714134216,
+      "learning_rate": 7.781218726285014e-06,
+      "loss": 0.2720562815666199,
+      "step": 277,
+      "token_acc": 0.9027445373018297
+    },
+    {
+      "epoch": 1.0736077481840194,
+      "grad_norm": 0.21282611787319183,
+      "learning_rate": 7.763237355030384e-06,
+      "loss": 0.2579670548439026,
+      "step": 278,
+      "token_acc": 0.9080073119376767
+    },
+    {
+      "epoch": 1.0774818401937045,
+      "grad_norm": 0.21488887071609497,
+      "learning_rate": 7.745204383925753e-06,
+      "loss": 0.2742394804954529,
+      "step": 279,
+      "token_acc": 0.9015262545209174
+    },
+    {
+      "epoch": 1.0813559322033899,
+      "grad_norm": 0.19826629757881165,
+      "learning_rate": 7.727120149713313e-06,
+      "loss": 0.23731666803359985,
+      "step": 280,
+      "token_acc": 0.9146603883445988
+    },
+    {
+      "epoch": 1.085230024213075,
+      "grad_norm": 0.20840346813201904,
+      "learning_rate": 7.708984990092528e-06,
+      "loss": 0.22673961520195007,
+      "step": 281,
+      "token_acc": 0.9184409845576723
+    },
+    {
+      "epoch": 1.0891041162227604,
+      "grad_norm": 0.21199366450309753,
+      "learning_rate": 7.690799243713825e-06,
+      "loss": 0.2788952887058258,
+      "step": 282,
+      "token_acc": 0.9002122640890617
+    },
+    {
+      "epoch": 1.0929782082324455,
+      "grad_norm": 0.23963455855846405,
+      "learning_rate": 7.672563250172278e-06,
+      "loss": 0.2703215479850769,
+      "step": 283,
+      "token_acc": 0.902904561306835
+    },
+    {
+      "epoch": 1.0968523002421307,
+      "grad_norm": 0.20739565789699554,
+      "learning_rate": 7.654277350001255e-06,
+      "loss": 0.2556743621826172,
+      "step": 284,
+      "token_acc": 0.9087778504769448
+    },
+    {
+      "epoch": 1.100726392251816,
+      "grad_norm": 0.3205340504646301,
+      "learning_rate": 7.635941884666072e-06,
+      "loss": 0.2660865783691406,
+      "step": 285,
+      "token_acc": 0.9052546447746934
+    },
+    {
+      "epoch": 1.1046004842615011,
+      "grad_norm": 0.20611628890037537,
+      "learning_rate": 7.617557196557601e-06,
+      "loss": 0.2590142488479614,
+      "step": 286,
+      "token_acc": 0.9070821077566713
+    },
+    {
+      "epoch": 1.1084745762711865,
+      "grad_norm": 0.1932753622531891,
+      "learning_rate": 7.599123628985894e-06,
+      "loss": 0.2396095246076584,
+      "step": 287,
+      "token_acc": 0.9135842317299648
+    },
+    {
+      "epoch": 1.1123486682808716,
+      "grad_norm": 0.21151748299598694,
+      "learning_rate": 7.580641526173758e-06,
+      "loss": 0.2544936537742615,
+      "step": 288,
+      "token_acc": 0.9088854539111634
+    },
+    {
+      "epoch": 1.116222760290557,
+      "grad_norm": 0.1992950737476349,
+      "learning_rate": 7.5621112332503325e-06,
+      "loss": 0.2544850707054138,
+      "step": 289,
+      "token_acc": 0.9090426161294457
+    },
+    {
+      "epoch": 1.1200968523002421,
+      "grad_norm": 0.20908565819263458,
+      "learning_rate": 7.543533096244644e-06,
+      "loss": 0.2762412428855896,
+      "step": 290,
+      "token_acc": 0.9013541447063986
+    },
+    {
+      "epoch": 1.1239709443099273,
+      "grad_norm": 0.2157965451478958,
+      "learning_rate": 7.524907462079149e-06,
+      "loss": 0.25533056259155273,
+      "step": 291,
+      "token_acc": 0.9080176353704462
+    },
+    {
+      "epoch": 1.1278450363196126,
+      "grad_norm": 0.19141145050525665,
+      "learning_rate": 7.506234678563248e-06,
+      "loss": 0.2362717241048813,
+      "step": 292,
+      "token_acc": 0.9155038610363999
+    },
+    {
+      "epoch": 1.1317191283292978,
+      "grad_norm": 0.21533732116222382,
+      "learning_rate": 7.487515094386792e-06,
+      "loss": 0.23099368810653687,
+      "step": 293,
+      "token_acc": 0.9173202498403009
+    },
+    {
+      "epoch": 1.1355932203389831,
+      "grad_norm": 0.20129309594631195,
+      "learning_rate": 7.468749059113578e-06,
+      "loss": 0.26144838333129883,
+      "step": 294,
+      "token_acc": 0.9057641431815713
+    },
+    {
+      "epoch": 1.1394673123486683,
+      "grad_norm": 0.3953739404678345,
+      "learning_rate": 7.449936923174813e-06,
+      "loss": 0.2557257413864136,
+      "step": 295,
+      "token_acc": 0.9087617787160037
+    },
+    {
+      "epoch": 1.1433414043583534,
+      "grad_norm": 0.21214410662651062,
+      "learning_rate": 7.431079037862575e-06,
+      "loss": 0.27983057498931885,
+      "step": 296,
+      "token_acc": 0.8996573827559394
+    },
+    {
+      "epoch": 1.1472154963680388,
+      "grad_norm": 0.20280665159225464,
+      "learning_rate": 7.412175755323254e-06,
+      "loss": 0.2772400677204132,
+      "step": 297,
+      "token_acc": 0.9010093723967251
+    },
+    {
+      "epoch": 1.151089588377724,
+      "grad_norm": 0.21776501834392548,
+      "learning_rate": 7.39322742855097e-06,
+      "loss": 0.24517808854579926,
+      "step": 298,
+      "token_acc": 0.9120538077359621
+    },
+    {
+      "epoch": 1.1549636803874093,
+      "grad_norm": 0.21630938351154327,
+      "learning_rate": 7.374234411380987e-06,
+      "loss": 0.2736694812774658,
+      "step": 299,
+      "token_acc": 0.9020631116999458
+    },
+    {
+      "epoch": 1.1588377723970944,
+      "grad_norm": 0.19338402152061462,
+      "learning_rate": 7.355197058483103e-06,
+      "loss": 0.24092288315296173,
+      "step": 300,
+      "token_acc": 0.9133508019967492
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 777,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1233780174946304e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82fc3e8dde96239190e8d51b6089ff008851cd3d6cf4700f6dc0f566e10a996c
+size 8977

video_preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 25165824,
+        "shortest_edge": 4096
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "video_processor_type": "Qwen3VLVideoProcessor"
+}