last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00047019548377237835,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.7019548377237836e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 8.4875,
      "step": 1
    },
    {
      "epoch": 4.7019548377237836e-05,
      "eval_loss": 9.813383102416992,
      "eval_runtime": 903.7894,
      "eval_samples_per_second": 4.955,
      "eval_steps_per_second": 4.955,
      "step": 1
    },
    {
      "epoch": 9.403909675447567e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 11.199,
      "step": 2
    },
    {
      "epoch": 0.00014105864513171352,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 11.0914,
      "step": 3
    },
    {
      "epoch": 0.00014105864513171352,
      "eval_loss": 9.813383102416992,
      "eval_runtime": 906.174,
      "eval_samples_per_second": 4.942,
      "eval_steps_per_second": 4.942,
      "step": 3
    },
    {
      "epoch": 0.00018807819350895135,
      "grad_norm": 19.602088928222656,
      "learning_rate": 2e-05,
      "loss": 9.625,
      "step": 4
    },
    {
      "epoch": 0.00023509774188618918,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 8.9464,
      "step": 5
    },
    {
      "epoch": 0.00028211729026342703,
      "grad_norm": 34.755611419677734,
      "learning_rate": 4e-05,
      "loss": 9.7495,
      "step": 6
    },
    {
      "epoch": 0.00028211729026342703,
      "eval_loss": 9.576491355895996,
      "eval_runtime": 898.0043,
      "eval_samples_per_second": 4.987,
      "eval_steps_per_second": 4.987,
      "step": 6
    },
    {
      "epoch": 0.00032913683864066484,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 9.5245,
      "step": 7
    },
    {
      "epoch": 0.0003761563870179027,
      "grad_norm": 36.6467170715332,
      "learning_rate": 6e-05,
      "loss": 9.123,
      "step": 8
    },
    {
      "epoch": 0.00042317593539514055,
      "grad_norm": 19.946910858154297,
      "learning_rate": 8e-05,
      "loss": 7.95,
      "step": 9
    },
    {
      "epoch": 0.00042317593539514055,
      "eval_loss": 7.195590972900391,
      "eval_runtime": 899.7507,
      "eval_samples_per_second": 4.977,
      "eval_steps_per_second": 4.977,
      "step": 9
    },
    {
      "epoch": 0.00047019548377237835,
      "grad_norm": 38.46210861206055,
      "learning_rate": 0.0001,
      "loss": 6.1035,
      "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": 7418872835604480.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}