last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003962139555359894,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00044023772837332157,
      "grad_norm": 5.396112442016602,
      "learning_rate": 2e-05,
      "loss": 12.852,
      "step": 1
    },
    {
      "epoch": 0.00044023772837332157,
      "eval_loss": 12.87714672088623,
      "eval_runtime": 31.4762,
      "eval_samples_per_second": 15.218,
      "eval_steps_per_second": 15.218,
      "step": 1
    },
    {
      "epoch": 0.0008804754567466431,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 12.5049,
      "step": 2
    },
    {
      "epoch": 0.0013207131851199648,
      "grad_norm": 6.062294960021973,
      "learning_rate": 4e-05,
      "loss": 11.39,
      "step": 3
    },
    {
      "epoch": 0.0013207131851199648,
      "eval_loss": 12.858711242675781,
      "eval_runtime": 31.6988,
      "eval_samples_per_second": 15.111,
      "eval_steps_per_second": 15.111,
      "step": 3
    },
    {
      "epoch": 0.0017609509134932863,
      "grad_norm": 5.33453893661499,
      "learning_rate": 6e-05,
      "loss": 11.7484,
      "step": 4
    },
    {
      "epoch": 0.002201188641866608,
      "grad_norm": 5.404849052429199,
      "learning_rate": 8e-05,
      "loss": 13.2869,
      "step": 5
    },
    {
      "epoch": 0.0026414263702399295,
      "grad_norm": 6.6339240074157715,
      "learning_rate": 0.0001,
      "loss": 12.8958,
      "step": 6
    },
    {
      "epoch": 0.0026414263702399295,
      "eval_loss": 12.353913307189941,
      "eval_runtime": 32.0275,
      "eval_samples_per_second": 14.956,
      "eval_steps_per_second": 14.956,
      "step": 6
    },
    {
      "epoch": 0.0030816640986132513,
      "grad_norm": 6.760335922241211,
      "learning_rate": 0.00012,
      "loss": 11.4293,
      "step": 7
    },
    {
      "epoch": 0.0035219018269865726,
      "grad_norm": 10.758153915405273,
      "learning_rate": 0.00014,
      "loss": 12.74,
      "step": 8
    },
    {
      "epoch": 0.003962139555359894,
      "grad_norm": 11.617557525634766,
      "learning_rate": 0.00016,
      "loss": 10.8383,
      "step": 9
    },
    {
      "epoch": 0.003962139555359894,
      "eval_loss": 10.073469161987305,
      "eval_runtime": 32.1815,
      "eval_samples_per_second": 14.884,
      "eval_steps_per_second": 14.884,
      "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": 1478847803424768.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}