trainer_state.json

{
  "best_global_step": 8000,
  "best_metric": 0.13777850568294525,
  "best_model_checkpoint": "check-point-qwen2.5-bags/checkpoint-8000",
  "epoch": 1.9176005273566248,
  "eval_steps": 2000,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.479415113561455,
      "eval_entropy": 0.15452208999515504,
      "eval_loss": 0.15754735469818115,
      "eval_mean_token_accuracy": 0.9525203127269597,
      "eval_num_tokens": 32768000.0,
      "eval_runtime": 11.8915,
      "eval_samples_per_second": 86.784,
      "eval_steps_per_second": 21.696,
      "step": 2000
    },
    {
      "epoch": 0.95883022712291,
      "eval_entropy": 0.1454699612403101,
      "eval_loss": 0.14876286685466766,
      "eval_mean_token_accuracy": 0.954681836819464,
      "eval_num_tokens": 65536000.0,
      "eval_runtime": 11.8969,
      "eval_samples_per_second": 86.745,
      "eval_steps_per_second": 21.686,
      "step": 4000
    },
    {
      "epoch": 1.43818541379517,
      "eval_entropy": 0.1354526253633721,
      "eval_loss": 0.14411340653896332,
      "eval_mean_token_accuracy": 0.9560297101505043,
      "eval_num_tokens": 98297856.0,
      "eval_runtime": 11.9057,
      "eval_samples_per_second": 86.682,
      "eval_steps_per_second": 21.67,
      "step": 6000
    },
    {
      "epoch": 1.9176005273566248,
      "eval_entropy": 0.13540720385174418,
      "eval_loss": 0.13777850568294525,
      "eval_mean_token_accuracy": 0.9574325225611989,
      "eval_num_tokens": 131065856.0,
      "eval_runtime": 11.8969,
      "eval_samples_per_second": 86.745,
      "eval_steps_per_second": 21.686,
      "step": 8000
    }
  ],
  "logging_steps": 8344,
  "max_steps": 41720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8144933446038323e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}