trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 25.3735801083628,
      "learning_rate": 1e-05,
      "loss": 1.0647,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.061476469039917,
      "eval_runtime": 2.8031,
      "eval_samples_per_second": 6.778,
      "eval_steps_per_second": 0.357,
      "step": 1
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.1875081062316895,
      "eval_runtime": 2.6772,
      "eval_samples_per_second": 7.097,
      "eval_steps_per_second": 0.374,
      "step": 2
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.985565185546875,
      "eval_runtime": 2.6682,
      "eval_samples_per_second": 7.121,
      "eval_steps_per_second": 0.375,
      "step": 3
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6195911169052124,
      "eval_runtime": 2.6829,
      "eval_samples_per_second": 7.082,
      "eval_steps_per_second": 0.373,
      "step": 4
    },
    {
      "epoch": 5.0,
      "grad_norm": 13.767739837044488,
      "learning_rate": 5.8682408883346535e-06,
      "loss": 0.969,
      "step": 5
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.4429333806037903,
      "eval_runtime": 2.6868,
      "eval_samples_per_second": 7.072,
      "eval_steps_per_second": 0.372,
      "step": 5
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.3589998185634613,
      "eval_runtime": 2.6805,
      "eval_samples_per_second": 7.088,
      "eval_steps_per_second": 0.373,
      "step": 6
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.24026596546173096,
      "eval_runtime": 2.671,
      "eval_samples_per_second": 7.113,
      "eval_steps_per_second": 0.374,
      "step": 7
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.18603171408176422,
      "eval_runtime": 2.7105,
      "eval_samples_per_second": 7.01,
      "eval_steps_per_second": 0.369,
      "step": 8
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.15581761300563812,
      "eval_runtime": 2.6748,
      "eval_samples_per_second": 7.103,
      "eval_steps_per_second": 0.374,
      "step": 9
    },
    {
      "epoch": 10.0,
      "grad_norm": 2.4836614664887184,
      "learning_rate": 0.0,
      "loss": 0.2782,
      "step": 10
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.14684411883354187,
      "eval_runtime": 2.6767,
      "eval_samples_per_second": 7.098,
      "eval_steps_per_second": 0.374,
      "step": 10
    },
    {
      "epoch": 10.0,
      "step": 10,
      "total_flos": 8375186227200.0,
      "train_loss": 0.6331865787506104,
      "train_runtime": 307.8989,
      "train_samples_per_second": 0.617,
      "train_steps_per_second": 0.032
    }
  ],
  "logging_steps": 5,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8375186227200.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}