trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06424945923371812,
  "eval_steps": 500,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005354121602809843,
      "grad_norm": NaN,
      "learning_rate": 0.00025,
      "loss": 4.8867,
      "step": 500
    },
    {
      "epoch": 0.010708243205619685,
      "grad_norm": NaN,
      "learning_rate": 0.0002777777777777778,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 0.01606236480842953,
      "grad_norm": NaN,
      "learning_rate": 0.00025,
      "loss": 0.0,
      "step": 1500
    },
    {
      "epoch": 0.02141648641123937,
      "grad_norm": NaN,
      "learning_rate": 0.00022222222222222218,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 0.026770608014049214,
      "grad_norm": NaN,
      "learning_rate": 0.00019444444444444443,
      "loss": 0.0,
      "step": 2500
    },
    {
      "epoch": 0.03212472961685906,
      "grad_norm": NaN,
      "learning_rate": 0.00016666666666666666,
      "loss": 0.0,
      "step": 3000
    },
    {
      "epoch": 0.0374788512196689,
      "grad_norm": NaN,
      "learning_rate": 0.0001388888888888889,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 0.04283297282247874,
      "grad_norm": NaN,
      "learning_rate": 0.00011111111111111109,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 0.04818709442528859,
      "grad_norm": NaN,
      "learning_rate": 8.333333333333333e-05,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 0.05354121602809843,
      "grad_norm": NaN,
      "learning_rate": 5.5555555555555545e-05,
      "loss": 0.0,
      "step": 5000
    },
    {
      "epoch": 0.058895337630908276,
      "grad_norm": NaN,
      "learning_rate": 2.7777777777777772e-05,
      "loss": 0.0,
      "step": 5500
    },
    {
      "epoch": 0.06424945923371812,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 6000
    },
    {
      "epoch": 0.06424945923371812,
      "step": 6000,
      "total_flos": 2.5256461458914173e+19,
      "train_loss": 0.4072261555989583,
      "train_runtime": 8835.4203,
      "train_samples_per_second": 2.716,
      "train_steps_per_second": 0.679
    }
  ],
  "logging_steps": 500,
  "max_steps": 6000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5256461458914173e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}