trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.131060202450719,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2663825253063399,
      "grad_norm": 4.165382385253906,
      "learning_rate": 1.822766826496182e-05,
      "loss": 2.1046,
      "mean_token_accuracy": 0.5178786942958832,
      "num_tokens": 6104570.0,
      "step": 500
    },
    {
      "epoch": 0.5327650506126798,
      "grad_norm": 4.388947010040283,
      "learning_rate": 1.6451784762919554e-05,
      "loss": 1.9567,
      "mean_token_accuracy": 0.5437576373815537,
      "num_tokens": 12233138.0,
      "step": 1000
    },
    {
      "epoch": 0.7991475759190197,
      "grad_norm": 4.746854782104492,
      "learning_rate": 1.4675901260877288e-05,
      "loss": 1.7979,
      "mean_token_accuracy": 0.5719504290819168,
      "num_tokens": 18348342.0,
      "step": 1500
    },
    {
      "epoch": 1.0655301012253595,
      "grad_norm": 5.537892818450928,
      "learning_rate": 1.2900017758835022e-05,
      "loss": 1.4819,
      "mean_token_accuracy": 0.6352935122847557,
      "num_tokens": 24492082.0,
      "step": 2000
    },
    {
      "epoch": 1.3319126265316994,
      "grad_norm": 5.861352443695068,
      "learning_rate": 1.1124134256792756e-05,
      "loss": 0.8953,
      "mean_token_accuracy": 0.7642805895209313,
      "num_tokens": 30611797.0,
      "step": 2500
    },
    {
      "epoch": 1.5982951518380393,
      "grad_norm": 6.586839199066162,
      "learning_rate": 9.348250754750488e-06,
      "loss": 0.7746,
      "mean_token_accuracy": 0.7995102950930595,
      "num_tokens": 36744673.0,
      "step": 3000
    },
    {
      "epoch": 1.8646776771443794,
      "grad_norm": 5.596500396728516,
      "learning_rate": 7.572367252708223e-06,
      "loss": 0.6492,
      "mean_token_accuracy": 0.8341795219779015,
      "num_tokens": 42844868.0,
      "step": 3500
    },
    {
      "epoch": 2.131060202450719,
      "grad_norm": 5.42270040512085,
      "learning_rate": 5.796483750665957e-06,
      "loss": 0.3917,
      "mean_token_accuracy": 0.9007325413227081,
      "num_tokens": 48963290.0,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 5631,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4237684301247283e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}