last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00881057268722467,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000881057268722467,
      "grad_norm": 34.048126220703125,
      "learning_rate": 2e-05,
      "loss": 12.5332,
      "step": 1
    },
    {
      "epoch": 0.000881057268722467,
      "eval_loss": 11.890963554382324,
      "eval_runtime": 121.6154,
      "eval_samples_per_second": 3.93,
      "eval_steps_per_second": 1.965,
      "step": 1
    },
    {
      "epoch": 0.001762114537444934,
      "grad_norm": 35.782840728759766,
      "learning_rate": 4e-05,
      "loss": 12.3024,
      "step": 2
    },
    {
      "epoch": 0.0026431718061674008,
      "grad_norm": 29.762380599975586,
      "learning_rate": 6e-05,
      "loss": 10.853,
      "step": 3
    },
    {
      "epoch": 0.0026431718061674008,
      "eval_loss": 11.337156295776367,
      "eval_runtime": 121.9438,
      "eval_samples_per_second": 3.92,
      "eval_steps_per_second": 1.96,
      "step": 3
    },
    {
      "epoch": 0.003524229074889868,
      "grad_norm": 37.080467224121094,
      "learning_rate": 8e-05,
      "loss": 10.3349,
      "step": 4
    },
    {
      "epoch": 0.004405286343612335,
      "grad_norm": 34.49464797973633,
      "learning_rate": 0.0001,
      "loss": 8.5361,
      "step": 5
    },
    {
      "epoch": 0.0052863436123348016,
      "grad_norm": 32.58349609375,
      "learning_rate": 0.00012,
      "loss": 8.3812,
      "step": 6
    },
    {
      "epoch": 0.0052863436123348016,
      "eval_loss": 7.188342094421387,
      "eval_runtime": 121.9598,
      "eval_samples_per_second": 3.919,
      "eval_steps_per_second": 1.96,
      "step": 6
    },
    {
      "epoch": 0.006167400881057269,
      "grad_norm": 32.926856994628906,
      "learning_rate": 0.00014,
      "loss": 7.5167,
      "step": 7
    },
    {
      "epoch": 0.007048458149779736,
      "grad_norm": 30.971134185791016,
      "learning_rate": 0.00016,
      "loss": 6.2016,
      "step": 8
    },
    {
      "epoch": 0.007929515418502203,
      "grad_norm": 27.42681121826172,
      "learning_rate": 0.00018,
      "loss": 6.2979,
      "step": 9
    },
    {
      "epoch": 0.007929515418502203,
      "eval_loss": 5.0941572189331055,
      "eval_runtime": 121.9511,
      "eval_samples_per_second": 3.92,
      "eval_steps_per_second": 1.96,
      "step": 9
    },
    {
      "epoch": 0.00881057268722467,
      "grad_norm": 21.287389755249023,
      "learning_rate": 0.0002,
      "loss": 5.2527,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4021193162096640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}