checkpoint-33/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 33,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 3.409710544347763,
      "epoch": 0.9302325581395349,
      "grad_norm": 3.647934913635254,
      "learning_rate": 0.0002,
      "loss": 3.7439,
      "mean_token_accuracy": 0.38004436269402503,
      "num_tokens": 1515.0,
      "step": 10
    },
    {
      "entropy": 1.9291726778715084,
      "epoch": 1.8372093023255816,
      "grad_norm": 2.9668562412261963,
      "learning_rate": 0.0002,
      "loss": 1.6097,
      "mean_token_accuracy": 0.6647876837314703,
      "num_tokens": 3024.0,
      "step": 20
    },
    {
      "entropy": 1.4819405537385206,
      "epoch": 2.744186046511628,
      "grad_norm": 3.0126216411590576,
      "learning_rate": 0.0002,
      "loss": 1.0587,
      "mean_token_accuracy": 0.7289426051653348,
      "num_tokens": 4458.0,
      "step": 30
    }
  ],
  "logging_steps": 10,
  "max_steps": 33,
  "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": 79329025290240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}