trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 34,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14705882352941177,
      "grad_norm": 0.04209648817777634,
      "learning_rate": 0.0002,
      "loss": 0.7597,
      "step": 5
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 0.04242026433348656,
      "learning_rate": 0.0002,
      "loss": 0.7107,
      "step": 10
    },
    {
      "epoch": 0.4411764705882353,
      "grad_norm": 0.042179469019174576,
      "learning_rate": 0.0002,
      "loss": 0.6837,
      "step": 15
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.03848237544298172,
      "learning_rate": 0.0002,
      "loss": 0.676,
      "step": 20
    },
    {
      "epoch": 0.7352941176470589,
      "grad_norm": 0.03739458695054054,
      "learning_rate": 0.0002,
      "loss": 0.6331,
      "step": 25
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 0.03209580481052399,
      "learning_rate": 0.0002,
      "loss": 0.6468,
      "step": 30
    },
    {
      "epoch": 1.0,
      "step": 34,
      "total_flos": 5.080206342144e+16,
      "train_loss": 0.6800254232743207,
      "train_runtime": 539.7583,
      "train_samples_per_second": 1.004,
      "train_steps_per_second": 0.063
    }
  ],
  "logging_steps": 5,
  "max_steps": 34,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.080206342144e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}