trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.951219512195122,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1951219512195122,
      "grad_norm": 4.03995494053413,
      "learning_rate": 9.755282581475769e-06,
      "loss": 0.5899,
      "step": 1
    },
    {
      "epoch": 0.3902439024390244,
      "grad_norm": 2.1553787750471396,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.4927,
      "step": 2
    },
    {
      "epoch": 0.5853658536585366,
      "grad_norm": 1.9968512780483432,
      "learning_rate": 7.938926261462366e-06,
      "loss": 0.4451,
      "step": 3
    },
    {
      "epoch": 0.7804878048780488,
      "grad_norm": 1.033843079794326,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.4112,
      "step": 4
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 2.578275255160274,
      "learning_rate": 5e-06,
      "loss": 0.4277,
      "step": 5
    },
    {
      "epoch": 1.170731707317073,
      "grad_norm": 2.95430153489752,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.8008,
      "step": 6
    },
    {
      "epoch": 1.3658536585365852,
      "grad_norm": 1.0165512987095444,
      "learning_rate": 2.061073738537635e-06,
      "loss": 0.3747,
      "step": 7
    },
    {
      "epoch": 1.5609756097560976,
      "grad_norm": 0.8528550920383245,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.3647,
      "step": 8
    },
    {
      "epoch": 1.7560975609756098,
      "grad_norm": 0.827469247048934,
      "learning_rate": 2.447174185242324e-07,
      "loss": 0.3668,
      "step": 9
    },
    {
      "epoch": 1.951219512195122,
      "grad_norm": 0.8143010619430315,
      "learning_rate": 0.0,
      "loss": 0.3622,
      "step": 10
    },
    {
      "epoch": 1.951219512195122,
      "step": 10,
      "total_flos": 5439441862656.0,
      "train_loss": 0.4635750144720078,
      "train_runtime": 268.2002,
      "train_samples_per_second": 1.223,
      "train_steps_per_second": 0.037
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 70000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5439441862656.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}