checkpoint-2814/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2814,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 1.715923547744751,
      "learning_rate": 0.00027878464818763325,
      "loss": 2.0994,
      "step": 200
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 1.554486870765686,
      "learning_rate": 0.00025746268656716415,
      "loss": 1.7202,
      "step": 400
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.6873255968093872,
      "learning_rate": 0.00023614072494669507,
      "loss": 1.6231,
      "step": 600
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 1.6301352977752686,
      "learning_rate": 0.000214818763326226,
      "loss": 1.6,
      "step": 800
    },
    {
      "epoch": 1.0661333333333334,
      "grad_norm": 1.7343031167984009,
      "learning_rate": 0.0001934968017057569,
      "loss": 1.4666,
      "step": 1000
    },
    {
      "epoch": 1.2794666666666665,
      "grad_norm": 1.3031765222549438,
      "learning_rate": 0.00017217484008528783,
      "loss": 1.4039,
      "step": 1200
    },
    {
      "epoch": 1.4928,
      "grad_norm": 1.2540940046310425,
      "learning_rate": 0.00015085287846481875,
      "loss": 1.4188,
      "step": 1400
    },
    {
      "epoch": 1.7061333333333333,
      "grad_norm": 1.3976752758026123,
      "learning_rate": 0.00012953091684434967,
      "loss": 1.3681,
      "step": 1600
    },
    {
      "epoch": 1.9194666666666667,
      "grad_norm": 1.8897106647491455,
      "learning_rate": 0.00010820895522388059,
      "loss": 1.3493,
      "step": 1800
    },
    {
      "epoch": 2.1322666666666668,
      "grad_norm": 1.1045914888381958,
      "learning_rate": 8.68869936034115e-05,
      "loss": 1.3455,
      "step": 2000
    },
    {
      "epoch": 2.3456,
      "grad_norm": 2.033783435821533,
      "learning_rate": 6.556503198294243e-05,
      "loss": 1.2918,
      "step": 2200
    },
    {
      "epoch": 2.558933333333333,
      "grad_norm": 1.3451552391052246,
      "learning_rate": 4.4243070362473345e-05,
      "loss": 1.2864,
      "step": 2400
    },
    {
      "epoch": 2.772266666666667,
      "grad_norm": 1.4732403755187988,
      "learning_rate": 2.2921108742004262e-05,
      "loss": 1.2869,
      "step": 2600
    },
    {
      "epoch": 2.9856,
      "grad_norm": 1.428038477897644,
      "learning_rate": 1.599147121535181e-06,
      "loss": 1.2471,
      "step": 2800
    }
  ],
  "logging_steps": 200,
  "max_steps": 2814,
  "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": 3044378482311168.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}