trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.6268375515937805,
      "epoch": 0.42105263157894735,
      "grad_norm": 10.60391392169834,
      "learning_rate": 0.0,
      "loss": 1.2625,
      "mean_token_accuracy": 0.748258426785469,
      "num_tokens": 655168.0,
      "step": 1
    },
    {
      "entropy": 0.5894907414913177,
      "epoch": 0.8421052631578947,
      "grad_norm": 9.558516207360615,
      "learning_rate": 1e-05,
      "loss": 1.1435,
      "mean_token_accuracy": 0.7735518142580986,
      "num_tokens": 1387063.0,
      "step": 2
    },
    {
      "entropy": 0.8542178273200989,
      "epoch": 1.0,
      "grad_norm": 3.1100659452025194,
      "learning_rate": 9.619397662556434e-06,
      "loss": 1.1435,
      "mean_token_accuracy": 0.7454110582669576,
      "num_tokens": 1613757.0,
      "step": 3
    },
    {
      "entropy": 0.9231625646352768,
      "epoch": 1.4210526315789473,
      "grad_norm": 1.693128940790078,
      "learning_rate": 8.535533905932739e-06,
      "loss": 0.8642,
      "mean_token_accuracy": 0.7867787629365921,
      "num_tokens": 2325693.0,
      "step": 4
    },
    {
      "entropy": 0.9480910897254944,
      "epoch": 1.8421052631578947,
      "grad_norm": 1.708409786596246,
      "learning_rate": 6.913417161825449e-06,
      "loss": 0.8868,
      "mean_token_accuracy": 0.7862061262130737,
      "num_tokens": 2982268.0,
      "step": 5
    },
    {
      "entropy": 0.9154236714045206,
      "epoch": 2.0,
      "grad_norm": 1.5359512263556723,
      "learning_rate": 5e-06,
      "loss": 0.6313,
      "mean_token_accuracy": 0.8401732444763184,
      "num_tokens": 3234680.0,
      "step": 6
    },
    {
      "entropy": 0.8867324888706207,
      "epoch": 2.4210526315789473,
      "grad_norm": 1.5346075141788051,
      "learning_rate": 3.0865828381745515e-06,
      "loss": 0.6152,
      "mean_token_accuracy": 0.8460986465215683,
      "num_tokens": 3925339.0,
      "step": 7
    },
    {
      "entropy": 0.7957010418176651,
      "epoch": 2.8421052631578947,
      "grad_norm": 1.7349674932563635,
      "learning_rate": 1.4644660940672628e-06,
      "loss": 0.5403,
      "mean_token_accuracy": 0.8625246286392212,
      "num_tokens": 4628235.0,
      "step": 8
    },
    {
      "entropy": 0.8120348453521729,
      "epoch": 3.0,
      "grad_norm": 2.2259970607185178,
      "learning_rate": 3.8060233744356634e-07,
      "loss": 0.5885,
      "mean_token_accuracy": 0.8512998024622599,
      "num_tokens": 4859504.0,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 9,
  "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": 16213663940608.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}