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": 1.1328805578866286,
      "learning_rate": 5e-06,
      "loss": 0.6261,
      "step": 10
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.5655437748332994,
      "learning_rate": 5e-06,
      "loss": 0.5229,
      "step": 20
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4255185947425879,
      "learning_rate": 5e-06,
      "loss": 0.4914,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.48239657282829285,
      "eval_runtime": 11.6513,
      "eval_samples_per_second": 69.349,
      "eval_steps_per_second": 1.116,
      "step": 30
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.4517651743600907,
      "learning_rate": 5e-06,
      "loss": 0.4487,
      "step": 40
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.3649131361498104,
      "learning_rate": 5e-06,
      "loss": 0.438,
      "step": 50
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3830225718161711,
      "learning_rate": 5e-06,
      "loss": 0.4301,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.45320141315460205,
      "eval_runtime": 11.8663,
      "eval_samples_per_second": 68.092,
      "eval_steps_per_second": 1.096,
      "step": 60
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.47711530562220955,
      "learning_rate": 5e-06,
      "loss": 0.3922,
      "step": 70
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.47041140445746077,
      "learning_rate": 5e-06,
      "loss": 0.3841,
      "step": 80
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.43944682743936286,
      "learning_rate": 5e-06,
      "loss": 0.3817,
      "step": 90
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4424304664134979,
      "eval_runtime": 11.6409,
      "eval_samples_per_second": 69.41,
      "eval_steps_per_second": 1.117,
      "step": 90
    },
    {
      "epoch": 3.0,
      "step": 90,
      "total_flos": 150543972433920.0,
      "train_loss": 0.4572353945838081,
      "train_runtime": 2587.8961,
      "train_samples_per_second": 17.787,
      "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
}