{
  "source": "raw\\runs\\20261030-010019\\checkpoints\\checkpoint-600\\training_args.bin",
  "training_arguments": {
    "output_dir": "C:\\Users\\USER\\Downloads\\mnist-decision-lab\\runs\\20261030-010019\\checkpoints",
    "per_device_train_batch_size": 32,
    "num_train_epochs": 3.0,
    "max_steps": 600,
    "learning_rate": 0.0002,
    "lr_scheduler_type": "cosine",
    "lr_scheduler_kwargs": null,
    "warmup_steps": 30,
    "optim": "adamw_torch_fused",
    "optim_args": null,
    "weight_decay": 0.01,
    "adam_beta1": 0.9,
    "adam_beta2": 0.999,
    "adam_epsilon": 1e-08,
    "optim_target_modules": null,
    "gradient_accumulation_steps": 1,
    "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": false,
    "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": 5,
    "logging_first_step": false,
    "log_on_each_node": true,
    "logging_nan_inf_filter": true,
    "include_num_input_tokens_seen": "no",
    "log_level": "passive",
    "log_level_replica": "warning",
    "disable_tqdm": false,
    "report_to": [],
    "run_name": null,
    "project": "huggingface",
    "trackio_space_id": null,
    "trackio_bucket_id": null,
    "trackio_static_space_id": null,
    "eval_strategy": "no",
    "eval_steps": null,
    "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": false,
    "save_strategy": "steps",
    "save_steps": 100,
    "save_on_each_node": false,
    "save_total_limit": 2,
    "enable_jit_checkpoint": false,
    "push_to_hub": false,
    "hub_token": "<HUB_TOKEN>",
    "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": null,
    "greater_is_better": null,
    "ignore_data_skip": false,
    "restore_callback_states_from_checkpoint": false,
    "full_determinism": false,
    "seed": 3407,
    "data_seed": null,
    "use_cpu": false,
    "accelerator_config": {
      "split_batches": false,
      "dispatch_batches": null,
      "even_batches": true,
      "use_seedable_sampler": true,
      "non_blocking": false,
      "gradient_accumulation_kwargs": null
    },
    "parallelism_config": null,
    "dataloader_drop_last": false,
    "dataloader_num_workers": 0,
    "dataloader_pin_memory": true,
    "dataloader_persistent_workers": false,
    "dataloader_prefetch_factor": null,
    "dataloader_multiprocessing_context": null,
    "dataloader_in_order": true,
    "remove_unused_columns": false,
    "label_names": [
      "target"
    ],
    "train_sampling_strategy": "random",
    "length_column_name": "length",
    "ddp_find_unused_parameters": null,
    "ddp_bucket_cap_mb": null,
    "ddp_broadcast_buffers": null,
    "ddp_static_graph": null,
    "ddp_backend": null,
    "ddp_timeout": 1800,
    "fsdp": null,
    "fsdp_config": null,
    "deepspeed": null,
    "debug": [],
    "skip_memory_metrics": true,
    "do_train": false,
    "do_eval": false,
    "do_predict": false,
    "resume_from_checkpoint": null,
    "local_rank": -1
  }
}