last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04343105320304017,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004343105320304018,
      "grad_norm": 1.5055750608444214,
      "learning_rate": 2e-05,
      "loss": 2.9357,
      "step": 1
    },
    {
      "epoch": 0.004343105320304018,
      "eval_loss": 0.9203988909721375,
      "eval_runtime": 16.8799,
      "eval_samples_per_second": 5.746,
      "eval_steps_per_second": 2.903,
      "step": 1
    },
    {
      "epoch": 0.008686210640608035,
      "grad_norm": 0.6354552507400513,
      "learning_rate": 4e-05,
      "loss": 3.224,
      "step": 2
    },
    {
      "epoch": 0.013029315960912053,
      "grad_norm": 0.8828385472297668,
      "learning_rate": 6e-05,
      "loss": 3.2607,
      "step": 3
    },
    {
      "epoch": 0.013029315960912053,
      "eval_loss": 0.9197360873222351,
      "eval_runtime": 16.9119,
      "eval_samples_per_second": 5.736,
      "eval_steps_per_second": 2.897,
      "step": 3
    },
    {
      "epoch": 0.01737242128121607,
      "grad_norm": 0.7923864722251892,
      "learning_rate": 8e-05,
      "loss": 3.549,
      "step": 4
    },
    {
      "epoch": 0.021715526601520086,
      "grad_norm": 0.9735816121101379,
      "learning_rate": 0.0001,
      "loss": 3.4902,
      "step": 5
    },
    {
      "epoch": 0.026058631921824105,
      "grad_norm": 1.5655157566070557,
      "learning_rate": 0.00012,
      "loss": 3.403,
      "step": 6
    },
    {
      "epoch": 0.026058631921824105,
      "eval_loss": 0.910624623298645,
      "eval_runtime": 17.0593,
      "eval_samples_per_second": 5.686,
      "eval_steps_per_second": 2.872,
      "step": 6
    },
    {
      "epoch": 0.03040173724212812,
      "grad_norm": 0.9918355345726013,
      "learning_rate": 0.00014,
      "loss": 3.3246,
      "step": 7
    },
    {
      "epoch": 0.03474484256243214,
      "grad_norm": 0.8918602466583252,
      "learning_rate": 0.00016,
      "loss": 3.1978,
      "step": 8
    },
    {
      "epoch": 0.03908794788273615,
      "grad_norm": 1.5681477785110474,
      "learning_rate": 0.00018,
      "loss": 3.2228,
      "step": 9
    },
    {
      "epoch": 0.03908794788273615,
      "eval_loss": 0.8799699544906616,
      "eval_runtime": 17.0714,
      "eval_samples_per_second": 5.682,
      "eval_steps_per_second": 2.87,
      "step": 9
    },
    {
      "epoch": 0.04343105320304017,
      "grad_norm": 1.3310284614562988,
      "learning_rate": 0.0002,
      "loss": 3.4842,
      "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": 6253253241077760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}