trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.888888888888889,
  "eval_steps": 1,
  "global_step": 7,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 1.6381683349609375,
      "eval_runtime": 22.2982,
      "eval_samples_per_second": 1.884,
      "eval_steps_per_second": 0.942,
      "step": 1
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.6295489072799683,
      "eval_runtime": 22.5248,
      "eval_samples_per_second": 1.865,
      "eval_steps_per_second": 0.932,
      "step": 2
    },
    {
      "epoch": 1.2222222222222223,
      "eval_loss": 1.5750688314437866,
      "eval_runtime": 22.6413,
      "eval_samples_per_second": 1.855,
      "eval_steps_per_second": 0.928,
      "step": 3
    },
    {
      "epoch": 1.6666666666666665,
      "eval_loss": 1.3997501134872437,
      "eval_runtime": 22.6484,
      "eval_samples_per_second": 1.854,
      "eval_steps_per_second": 0.927,
      "step": 4
    },
    {
      "epoch": 2.111111111111111,
      "eval_loss": 1.0638465881347656,
      "eval_runtime": 22.6523,
      "eval_samples_per_second": 1.854,
      "eval_steps_per_second": 0.927,
      "step": 5
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012,
      "loss": 1.292,
      "step": 6
    },
    {
      "epoch": 2.4444444444444446,
      "eval_loss": 0.7260665893554688,
      "eval_runtime": 22.642,
      "eval_samples_per_second": 1.855,
      "eval_steps_per_second": 0.927,
      "step": 6
    },
    {
      "epoch": 2.888888888888889,
      "eval_loss": 0.4600009024143219,
      "eval_runtime": 22.6317,
      "eval_samples_per_second": 1.856,
      "eval_steps_per_second": 0.928,
      "step": 7
    }
  ],
  "logging_steps": 6,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.140358009513574e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}