checkpoint-200/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4691358024691357,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 69.9074478149414,
      "learning_rate": 9.952e-06,
      "loss": 4.4273,
      "step": 30
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 96.12484741210938,
      "learning_rate": 9.892000000000001e-06,
      "loss": 2.6895,
      "step": 60
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 43.14171600341797,
      "learning_rate": 9.832e-06,
      "loss": 2.4001,
      "step": 90
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 41.42817306518555,
      "learning_rate": 9.772e-06,
      "loss": 2.0886,
      "step": 120
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 41.4298095703125,
      "learning_rate": 9.712e-06,
      "loss": 1.9545,
      "step": 150
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 32.89754104614258,
      "learning_rate": 9.652e-06,
      "loss": 1.8443,
      "step": 180
    }
  ],
  "logging_steps": 30,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 62,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.64486332416e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}