last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.961077091265879e-05,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.961077091265878e-06,
      "grad_norm": 10.275520324707031,
      "learning_rate": 2e-05,
      "loss": 28.3132,
      "step": 1
    },
    {
      "epoch": 9.961077091265878e-06,
      "eval_loss": 7.1797685623168945,
      "eval_runtime": 2183.9928,
      "eval_samples_per_second": 19.354,
      "eval_steps_per_second": 9.677,
      "step": 1
    },
    {
      "epoch": 1.9922154182531756e-05,
      "grad_norm": 11.051509857177734,
      "learning_rate": 4e-05,
      "loss": 30.5221,
      "step": 2
    },
    {
      "epoch": 2.9883231273797637e-05,
      "grad_norm": 9.985577583312988,
      "learning_rate": 6e-05,
      "loss": 28.8905,
      "step": 3
    },
    {
      "epoch": 2.9883231273797637e-05,
      "eval_loss": 7.147799015045166,
      "eval_runtime": 2176.4219,
      "eval_samples_per_second": 19.422,
      "eval_steps_per_second": 9.711,
      "step": 3
    },
    {
      "epoch": 3.984430836506351e-05,
      "grad_norm": 10.860417366027832,
      "learning_rate": 8e-05,
      "loss": 25.7236,
      "step": 4
    },
    {
      "epoch": 4.980538545632939e-05,
      "grad_norm": 13.008605003356934,
      "learning_rate": 0.0001,
      "loss": 24.5482,
      "step": 5
    },
    {
      "epoch": 5.9766462547595274e-05,
      "grad_norm": 16.258377075195312,
      "learning_rate": 0.00012,
      "loss": 25.6603,
      "step": 6
    },
    {
      "epoch": 5.9766462547595274e-05,
      "eval_loss": 6.418942451477051,
      "eval_runtime": 2178.0656,
      "eval_samples_per_second": 19.407,
      "eval_steps_per_second": 9.704,
      "step": 6
    },
    {
      "epoch": 6.972753963886115e-05,
      "grad_norm": 20.417570114135742,
      "learning_rate": 0.00014,
      "loss": 26.6501,
      "step": 7
    },
    {
      "epoch": 7.968861673012702e-05,
      "grad_norm": 23.798818588256836,
      "learning_rate": 0.00016,
      "loss": 22.8077,
      "step": 8
    },
    {
      "epoch": 8.964969382139291e-05,
      "grad_norm": 30.659055709838867,
      "learning_rate": 0.00018,
      "loss": 18.2496,
      "step": 9
    },
    {
      "epoch": 8.964969382139291e-05,
      "eval_loss": 2.7002601623535156,
      "eval_runtime": 2181.454,
      "eval_samples_per_second": 19.377,
      "eval_steps_per_second": 9.688,
      "step": 9
    },
    {
      "epoch": 9.961077091265879e-05,
      "grad_norm": 26.626014709472656,
      "learning_rate": 0.0002,
      "loss": 9.7356,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}