last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06309148264984227,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.006309148264984227,
      "grad_norm": 27.031143188476562,
      "learning_rate": 2e-05,
      "loss": 11.3864,
      "step": 1
    },
    {
      "epoch": 0.006309148264984227,
      "eval_loss": 11.524320602416992,
      "eval_runtime": 2.137,
      "eval_samples_per_second": 31.353,
      "eval_steps_per_second": 15.91,
      "step": 1
    },
    {
      "epoch": 0.012618296529968454,
      "grad_norm": 27.037906646728516,
      "learning_rate": 4e-05,
      "loss": 13.0125,
      "step": 2
    },
    {
      "epoch": 0.01892744479495268,
      "grad_norm": 26.173236846923828,
      "learning_rate": 6e-05,
      "loss": 11.2995,
      "step": 3
    },
    {
      "epoch": 0.01892744479495268,
      "eval_loss": 11.12730884552002,
      "eval_runtime": 2.1326,
      "eval_samples_per_second": 31.416,
      "eval_steps_per_second": 15.943,
      "step": 3
    },
    {
      "epoch": 0.025236593059936908,
      "grad_norm": 29.974369049072266,
      "learning_rate": 8e-05,
      "loss": 11.5018,
      "step": 4
    },
    {
      "epoch": 0.031545741324921134,
      "grad_norm": 27.482757568359375,
      "learning_rate": 0.0001,
      "loss": 11.4434,
      "step": 5
    },
    {
      "epoch": 0.03785488958990536,
      "grad_norm": 32.79706954956055,
      "learning_rate": 0.00012,
      "loss": 8.8625,
      "step": 6
    },
    {
      "epoch": 0.03785488958990536,
      "eval_loss": 6.5188446044921875,
      "eval_runtime": 2.1634,
      "eval_samples_per_second": 30.97,
      "eval_steps_per_second": 15.716,
      "step": 6
    },
    {
      "epoch": 0.04416403785488959,
      "grad_norm": 29.903905868530273,
      "learning_rate": 0.00014,
      "loss": 6.9666,
      "step": 7
    },
    {
      "epoch": 0.050473186119873815,
      "grad_norm": 23.711275100708008,
      "learning_rate": 0.00016,
      "loss": 4.5659,
      "step": 8
    },
    {
      "epoch": 0.056782334384858045,
      "grad_norm": 16.82898712158203,
      "learning_rate": 0.00018,
      "loss": 2.8434,
      "step": 9
    },
    {
      "epoch": 0.056782334384858045,
      "eval_loss": 1.9197030067443848,
      "eval_runtime": 2.1341,
      "eval_samples_per_second": 31.395,
      "eval_steps_per_second": 15.932,
      "step": 9
    },
    {
      "epoch": 0.06309148264984227,
      "grad_norm": 10.088571548461914,
      "learning_rate": 0.0002,
      "loss": 1.6594,
      "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": 562609497047040.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}