last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016568463999834316,
  "eval_steps": 10,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.284231999917157e-05,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 8.284231999917157e-05,
      "eval_loss": NaN,
      "eval_runtime": 1482.1917,
      "eval_samples_per_second": 6.859,
      "eval_steps_per_second": 3.429,
      "step": 1
    },
    {
      "epoch": 0.00016568463999834315,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00024852695999751475,
      "grad_norm": NaN,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0003313692799966863,
      "grad_norm": NaN,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0004142115999958579,
      "grad_norm": NaN,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0004970539199950295,
      "grad_norm": NaN,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.000579896239994201,
      "grad_norm": NaN,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0006627385599933726,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0007455808799925441,
      "grad_norm": NaN,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0008284231999917158,
      "grad_norm": NaN,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0008284231999917158,
      "eval_loss": NaN,
      "eval_runtime": 1482.4697,
      "eval_samples_per_second": 6.857,
      "eval_steps_per_second": 3.429,
      "step": 10
    },
    {
      "epoch": 0.0009112655199908873,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.000994107839990059,
      "grad_norm": NaN,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0010769501599892305,
      "grad_norm": NaN,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.001159792479988402,
      "grad_norm": NaN,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0012426347999875736,
      "grad_norm": NaN,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0013254771199867452,
      "grad_norm": NaN,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.0014083194399859167,
      "grad_norm": NaN,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.0014911617599850883,
      "grad_norm": NaN,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.00157400407998426,
      "grad_norm": NaN,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0016568463999834316,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0016568463999834316,
      "eval_loss": NaN,
      "eval_runtime": 1482.283,
      "eval_samples_per_second": 6.858,
      "eval_steps_per_second": 3.429,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8466919858241536e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}