trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.84,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 1.1282271146774292,
      "learning_rate": 0.00018333333333333334,
      "loss": 9.4167,
      "step": 2
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.5277754068374634,
      "learning_rate": 0.0001666666666666667,
      "loss": 9.031,
      "step": 4
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.8196742534637451,
      "learning_rate": 0.00015000000000000001,
      "loss": 8.5383,
      "step": 6
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.224635362625122,
      "learning_rate": 0.00013333333333333334,
      "loss": 8.0253,
      "step": 8
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.598311185836792,
      "learning_rate": 0.00011666666666666668,
      "loss": 7.8484,
      "step": 10
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.8537167310714722,
      "learning_rate": 0.0001,
      "loss": 7.676,
      "step": 12
    },
    {
      "epoch": 2.24,
      "grad_norm": 1.2156755924224854,
      "learning_rate": 8.333333333333334e-05,
      "loss": 7.4581,
      "step": 14
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.1320364475250244,
      "learning_rate": 6.666666666666667e-05,
      "loss": 7.3431,
      "step": 16
    },
    {
      "epoch": 2.88,
      "grad_norm": 1.965053915977478,
      "learning_rate": 5e-05,
      "loss": 7.1658,
      "step": 18
    },
    {
      "epoch": 3.2,
      "grad_norm": 1.133592963218689,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 7.1966,
      "step": 20
    },
    {
      "epoch": 3.52,
      "grad_norm": 1.4115341901779175,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 7.1508,
      "step": 22
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.9762312173843384,
      "learning_rate": 0.0,
      "loss": 7.0999,
      "step": 24
    },
    {
      "epoch": 3.84,
      "step": 24,
      "total_flos": 125128268611296.0,
      "train_loss": 7.829166213671367,
      "train_runtime": 119.3929,
      "train_samples_per_second": 3.35,
      "train_steps_per_second": 0.201
    }
  ],
  "logging_steps": 2,
  "max_steps": 24,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 125128268611296.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}