trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8284789644012945,
  "eval_steps": 500,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20711974110032363,
      "grad_norm": 16.30859852459602,
      "learning_rate": 2e-05,
      "loss": 0.6532,
      "mean_token_accuracy": 0.8563957028090954,
      "num_tokens": 523828.0,
      "step": 1
    },
    {
      "epoch": 0.41423948220064727,
      "grad_norm": 15.747215265537687,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.4982,
      "mean_token_accuracy": 0.8662941353395581,
      "num_tokens": 1047723.0,
      "step": 2
    },
    {
      "epoch": 0.6213592233009708,
      "grad_norm": 3.538420138606602,
      "learning_rate": 1e-05,
      "loss": 0.4344,
      "mean_token_accuracy": 0.8762655928730965,
      "num_tokens": 1565521.0,
      "step": 3
    },
    {
      "epoch": 0.8284789644012945,
      "grad_norm": 3.378345970394599,
      "learning_rate": 5e-06,
      "loss": 0.4061,
      "mean_token_accuracy": 0.8822592282667756,
      "num_tokens": 2089809.0,
      "step": 4
    },
    {
      "epoch": 0.8284789644012945,
      "step": 4,
      "total_flos": 4162967633920.0,
      "train_loss": 0.49796468019485474,
      "train_runtime": 721.0845,
      "train_samples_per_second": 0.856,
      "train_steps_per_second": 0.006
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 4,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4162967633920.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}