last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016909744240118367,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001690974424011837,
      "grad_norm": 7.734948635101318,
      "learning_rate": 2e-05,
      "loss": 6.6256,
      "step": 1
    },
    {
      "epoch": 0.0001690974424011837,
      "eval_loss": 6.817934036254883,
      "eval_runtime": 86.5425,
      "eval_samples_per_second": 28.772,
      "eval_steps_per_second": 14.386,
      "step": 1
    },
    {
      "epoch": 0.0003381948848023674,
      "grad_norm": 8.201972007751465,
      "learning_rate": 4e-05,
      "loss": 6.6485,
      "step": 2
    },
    {
      "epoch": 0.000507292327203551,
      "grad_norm": 7.980592250823975,
      "learning_rate": 6e-05,
      "loss": 6.9518,
      "step": 3
    },
    {
      "epoch": 0.000507292327203551,
      "eval_loss": 6.735722541809082,
      "eval_runtime": 85.3061,
      "eval_samples_per_second": 29.189,
      "eval_steps_per_second": 14.594,
      "step": 3
    },
    {
      "epoch": 0.0006763897696047348,
      "grad_norm": 8.044904708862305,
      "learning_rate": 8e-05,
      "loss": 6.3287,
      "step": 4
    },
    {
      "epoch": 0.0008454872120059184,
      "grad_norm": 8.073067665100098,
      "learning_rate": 0.0001,
      "loss": 6.4404,
      "step": 5
    },
    {
      "epoch": 0.001014584654407102,
      "grad_norm": 7.195446491241455,
      "learning_rate": 0.00012,
      "loss": 5.6174,
      "step": 6
    },
    {
      "epoch": 0.001014584654407102,
      "eval_loss": 5.374767780303955,
      "eval_runtime": 85.5385,
      "eval_samples_per_second": 29.11,
      "eval_steps_per_second": 14.555,
      "step": 6
    },
    {
      "epoch": 0.0011836820968082857,
      "grad_norm": 6.348795413970947,
      "learning_rate": 0.00014,
      "loss": 5.6487,
      "step": 7
    },
    {
      "epoch": 0.0013527795392094695,
      "grad_norm": 5.67942476272583,
      "learning_rate": 0.00016,
      "loss": 4.7154,
      "step": 8
    },
    {
      "epoch": 0.0015218769816106531,
      "grad_norm": 5.2752861976623535,
      "learning_rate": 0.00018,
      "loss": 4.32,
      "step": 9
    },
    {
      "epoch": 0.0015218769816106531,
      "eval_loss": 3.6447913646698,
      "eval_runtime": 85.335,
      "eval_samples_per_second": 29.179,
      "eval_steps_per_second": 14.59,
      "step": 9
    },
    {
      "epoch": 0.0016909744240118367,
      "grad_norm": 5.289436340332031,
      "learning_rate": 0.0002,
      "loss": 3.5508,
      "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": 1199811613163520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}