trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.9154215336812168,
      "learning_rate": 5e-06,
      "loss": 0.6823,
      "step": 10
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.7537911360343048,
      "learning_rate": 5e-06,
      "loss": 0.5691,
      "step": 20
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4949368852301628,
      "learning_rate": 5e-06,
      "loss": 0.53,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.518334150314331,
      "eval_runtime": 11.549,
      "eval_samples_per_second": 69.963,
      "eval_steps_per_second": 1.126,
      "step": 30
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.56431020956907,
      "learning_rate": 5e-06,
      "loss": 0.4812,
      "step": 40
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.5632165897700695,
      "learning_rate": 5e-06,
      "loss": 0.4679,
      "step": 50
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.733341343744141,
      "learning_rate": 5e-06,
      "loss": 0.4581,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4814133942127228,
      "eval_runtime": 11.8783,
      "eval_samples_per_second": 68.023,
      "eval_steps_per_second": 1.094,
      "step": 60
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.6395934420420472,
      "learning_rate": 5e-06,
      "loss": 0.4144,
      "step": 70
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.7408061967156113,
      "learning_rate": 5e-06,
      "loss": 0.4057,
      "step": 80
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6987607406583534,
      "learning_rate": 5e-06,
      "loss": 0.4024,
      "step": 90
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.46840161085128784,
      "eval_runtime": 12.5392,
      "eval_samples_per_second": 64.438,
      "eval_steps_per_second": 1.037,
      "step": 90
    },
    {
      "epoch": 3.0,
      "step": 90,
      "total_flos": 150543972433920.0,
      "train_loss": 0.4901230918036567,
      "train_runtime": 2567.367,
      "train_samples_per_second": 17.93,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 10,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 150543972433920.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}