last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00024032524015834764,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.6702804462038626e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.1234,
      "step": 1
    },
    {
      "epoch": 2.6702804462038626e-05,
      "eval_loss": 5.431859493255615,
      "eval_runtime": 1660.4484,
      "eval_samples_per_second": 4.749,
      "eval_steps_per_second": 4.749,
      "step": 1
    },
    {
      "epoch": 5.340560892407725e-05,
      "grad_norm": 7.5764665603637695,
      "learning_rate": 2e-05,
      "loss": 4.6044,
      "step": 2
    },
    {
      "epoch": 8.010841338611588e-05,
      "grad_norm": 8.277660369873047,
      "learning_rate": 4e-05,
      "loss": 5.0017,
      "step": 3
    },
    {
      "epoch": 8.010841338611588e-05,
      "eval_loss": 5.314225196838379,
      "eval_runtime": 1711.0308,
      "eval_samples_per_second": 4.608,
      "eval_steps_per_second": 4.608,
      "step": 3
    },
    {
      "epoch": 0.0001068112178481545,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 5.4632,
      "step": 4
    },
    {
      "epoch": 0.00013351402231019313,
      "grad_norm": 8.344459533691406,
      "learning_rate": 6e-05,
      "loss": 5.1562,
      "step": 5
    },
    {
      "epoch": 0.00016021682677223176,
      "grad_norm": 7.8158040046691895,
      "learning_rate": 8e-05,
      "loss": 4.5659,
      "step": 6
    },
    {
      "epoch": 0.00016021682677223176,
      "eval_loss": 4.253037929534912,
      "eval_runtime": 1868.5451,
      "eval_samples_per_second": 4.22,
      "eval_steps_per_second": 4.22,
      "step": 6
    },
    {
      "epoch": 0.00018691963123427038,
      "grad_norm": 9.651495933532715,
      "learning_rate": 0.0001,
      "loss": 3.8852,
      "step": 7
    },
    {
      "epoch": 0.000213622435696309,
      "grad_norm": 8.860259056091309,
      "learning_rate": 0.00012,
      "loss": 3.537,
      "step": 8
    },
    {
      "epoch": 0.00024032524015834764,
      "grad_norm": 7.234774112701416,
      "learning_rate": 0.00014,
      "loss": 2.4743,
      "step": 9
    },
    {
      "epoch": 0.00024032524015834764,
      "eval_loss": 2.026648759841919,
      "eval_runtime": 1650.4991,
      "eval_samples_per_second": 4.777,
      "eval_steps_per_second": 4.777,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6676985552044032.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}