last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013951866062085804,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013951866062085804,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0013951866062085804,
      "eval_loss": NaN,
      "eval_runtime": 12.4465,
      "eval_samples_per_second": 24.264,
      "eval_steps_per_second": 12.132,
      "step": 1
    },
    {
      "epoch": 0.0027903732124171608,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.004185559818625741,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 1.9907,
      "step": 3
    },
    {
      "epoch": 0.004185559818625741,
      "eval_loss": NaN,
      "eval_runtime": 11.052,
      "eval_samples_per_second": 27.325,
      "eval_steps_per_second": 13.663,
      "step": 3
    },
    {
      "epoch": 0.0055807464248343215,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.006975933031042902,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 2.732,
      "step": 5
    },
    {
      "epoch": 0.008371119637251482,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.008371119637251482,
      "eval_loss": NaN,
      "eval_runtime": 11.1533,
      "eval_samples_per_second": 27.077,
      "eval_steps_per_second": 13.539,
      "step": 6
    },
    {
      "epoch": 0.009766306243460064,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 1.3876,
      "step": 7
    },
    {
      "epoch": 0.011161492849668643,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 3.5812,
      "step": 8
    },
    {
      "epoch": 0.012556679455877224,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 1.0287,
      "step": 9
    },
    {
      "epoch": 0.012556679455877224,
      "eval_loss": NaN,
      "eval_runtime": 11.1153,
      "eval_samples_per_second": 27.17,
      "eval_steps_per_second": 13.585,
      "step": 9
    },
    {
      "epoch": 0.013951866062085804,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.9319,
      "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": 2259350810787840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}