last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011379800853485065,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011379800853485065,
      "grad_norm": 11.28669261932373,
      "learning_rate": 2e-05,
      "loss": 4.9874,
      "step": 1
    },
    {
      "epoch": 0.0011379800853485065,
      "eval_loss": 5.444311618804932,
      "eval_runtime": 263.497,
      "eval_samples_per_second": 1.404,
      "eval_steps_per_second": 0.702,
      "step": 1
    },
    {
      "epoch": 0.002275960170697013,
      "grad_norm": 10.132047653198242,
      "learning_rate": 4e-05,
      "loss": 4.0493,
      "step": 2
    },
    {
      "epoch": 0.0034139402560455193,
      "grad_norm": 13.788056373596191,
      "learning_rate": 6e-05,
      "loss": 5.1544,
      "step": 3
    },
    {
      "epoch": 0.0034139402560455193,
      "eval_loss": 5.3216962814331055,
      "eval_runtime": 263.6528,
      "eval_samples_per_second": 1.403,
      "eval_steps_per_second": 0.702,
      "step": 3
    },
    {
      "epoch": 0.004551920341394026,
      "grad_norm": 9.466752052307129,
      "learning_rate": 8e-05,
      "loss": 4.2767,
      "step": 4
    },
    {
      "epoch": 0.005689900426742532,
      "grad_norm": 21.691171646118164,
      "learning_rate": 0.0001,
      "loss": 6.9685,
      "step": 5
    },
    {
      "epoch": 0.0068278805120910386,
      "grad_norm": 16.943603515625,
      "learning_rate": 0.00012,
      "loss": 4.566,
      "step": 6
    },
    {
      "epoch": 0.0068278805120910386,
      "eval_loss": 4.045292854309082,
      "eval_runtime": 264.6413,
      "eval_samples_per_second": 1.398,
      "eval_steps_per_second": 0.699,
      "step": 6
    },
    {
      "epoch": 0.007965860597439544,
      "grad_norm": 9.648813247680664,
      "learning_rate": 0.00014,
      "loss": 3.3786,
      "step": 7
    },
    {
      "epoch": 0.009103840682788052,
      "grad_norm": 8.042987823486328,
      "learning_rate": 0.00016,
      "loss": 2.8221,
      "step": 8
    },
    {
      "epoch": 0.010241820768136558,
      "grad_norm": 7.113251686096191,
      "learning_rate": 0.00018,
      "loss": 2.8156,
      "step": 9
    },
    {
      "epoch": 0.010241820768136558,
      "eval_loss": 2.4009945392608643,
      "eval_runtime": 144.3345,
      "eval_samples_per_second": 2.563,
      "eval_steps_per_second": 1.282,
      "step": 9
    },
    {
      "epoch": 0.011379800853485065,
      "grad_norm": 4.849100112915039,
      "learning_rate": 0.0002,
      "loss": 1.8402,
      "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": 621231451668480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}