{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 105,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3463203463203463,
      "grad_norm": 9.76123928801732,
      "learning_rate": 7.272727272727273e-06,
      "loss": 1.7913,
      "mean_token_accuracy": 0.6742337845265866,
      "num_tokens": 3541090.0,
      "step": 5
    },
    {
      "epoch": 0.6926406926406926,
      "grad_norm": 1.7292229461629567,
      "learning_rate": 1.6363636363636366e-05,
      "loss": 1.1842,
      "mean_token_accuracy": 0.7258017435669899,
      "num_tokens": 6936595.0,
      "step": 10
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.958735263554765,
      "learning_rate": 1.994977815088504e-05,
      "loss": 0.6688,
      "mean_token_accuracy": 0.8406233997412131,
      "num_tokens": 9820463.0,
      "step": 15
    },
    {
      "epoch": 1.3463203463203464,
      "grad_norm": 0.6172390521588779,
      "learning_rate": 1.964469175054377e-05,
      "loss": 0.5533,
      "mean_token_accuracy": 0.8736346215009689,
      "num_tokens": 13203490.0,
      "step": 20
    },
    {
      "epoch": 1.6926406926406927,
      "grad_norm": 0.1996301013239303,
      "learning_rate": 1.907090913734341e-05,
      "loss": 0.5113,
      "mean_token_accuracy": 0.8854866914451123,
      "num_tokens": 16677343.0,
      "step": 25
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4762125617018596,
      "learning_rate": 1.8244415603417603e-05,
      "loss": 0.557,
      "mean_token_accuracy": 0.8772783690774945,
      "num_tokens": 19657455.0,
      "step": 30
    },
    {
      "epoch": 2.346320346320346,
      "grad_norm": 0.17061055880750653,
      "learning_rate": 1.7188236838779297e-05,
      "loss": 0.4643,
      "mean_token_accuracy": 0.8905399434268475,
      "num_tokens": 23134171.0,
      "step": 35
    },
    {
      "epoch": 2.6926406926406927,
      "grad_norm": 0.17654122424112745,
      "learning_rate": 1.5931797447293553e-05,
      "loss": 0.4953,
      "mean_token_accuracy": 0.8895446695387363,
      "num_tokens": 26513134.0,
      "step": 40
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4198550622011698,
      "learning_rate": 1.451010119216102e-05,
      "loss": 0.5039,
      "mean_token_accuracy": 0.8886955749820655,
      "num_tokens": 29483157.0,
      "step": 45
    },
    {
      "epoch": 3.346320346320346,
      "grad_norm": 0.18028115485047716,
      "learning_rate": 1.2962755808856341e-05,
      "loss": 0.4518,
      "mean_token_accuracy": 0.9004578970372676,
      "num_tokens": 32726053.0,
      "step": 50
    },
    {
      "epoch": 3.6926406926406927,
      "grad_norm": 0.18706797363551947,
      "learning_rate": 1.133286955373779e-05,
      "loss": 0.46,
      "mean_token_accuracy": 0.8930317558348179,
      "num_tokens": 36258177.0,
      "step": 55
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.15856177406249233,
      "learning_rate": 9.665850229923258e-06,
      "loss": 0.4212,
      "mean_token_accuracy": 0.8997706342750872,
      "num_tokens": 39301701.0,
      "step": 60
    },
    {
      "epoch": 4.346320346320346,
      "grad_norm": 0.17718136029490175,
      "learning_rate": 8.008140148961642e-06,
      "loss": 0.4743,
      "mean_token_accuracy": 0.8929261632263661,
      "num_tokens": 42943792.0,
      "step": 65
    },
    {
      "epoch": 4.692640692640692,
      "grad_norm": 0.17848902736372635,
      "learning_rate": 6.405922271624874e-06,
      "loss": 0.4136,
      "mean_token_accuracy": 0.9024837173521518,
      "num_tokens": 46472250.0,
      "step": 70
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2562768112814367,
      "learning_rate": 4.903833574080825e-06,
      "loss": 0.3746,
      "mean_token_accuracy": 0.9082856144703609,
      "num_tokens": 49176637.0,
      "step": 75
    },
    {
      "epoch": 5.346320346320346,
      "grad_norm": 0.17829278326006745,
      "learning_rate": 3.543721484411976e-06,
      "loss": 0.4164,
      "mean_token_accuracy": 0.9040168270468711,
      "num_tokens": 52544587.0,
      "step": 80
    },
    {
      "epoch": 5.692640692640692,
      "grad_norm": 0.159769199751552,
      "learning_rate": 2.3634780345266805e-06,
      "loss": 0.4005,
      "mean_token_accuracy": 0.9068072281777859,
      "num_tokens": 56073263.0,
      "step": 85
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.17497872253097468,
      "learning_rate": 1.3959842073986085e-06,
      "loss": 0.3873,
      "mean_token_accuracy": 0.907542069193343,
      "num_tokens": 59025335.0,
      "step": 90
    },
    {
      "epoch": 6.346320346320346,
      "grad_norm": 0.16214766397656102,
      "learning_rate": 6.681938895839746e-07,
      "loss": 0.421,
      "mean_token_accuracy": 0.9028209887444973,
      "num_tokens": 62643177.0,
      "step": 95
    },
    {
      "epoch": 6.692640692640692,
      "grad_norm": 0.17433731599885693,
      "learning_rate": 2.0038294963413251e-07,
      "loss": 0.3946,
      "mean_token_accuracy": 0.9086535051465034,
      "num_tokens": 66032974.0,
      "step": 100
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.26625567760062313,
      "learning_rate": 5.584362697453882e-09,
      "loss": 0.3765,
      "mean_token_accuracy": 0.910740121149681,
      "num_tokens": 68862598.0,
      "step": 105
    },
    {
      "epoch": 7.0,
      "step": 105,
      "total_flos": 100971315986432.0,
      "train_loss": 0.5581429311207362,
      "train_runtime": 2344.1731,
      "train_samples_per_second": 5.509,
      "train_steps_per_second": 0.045
    }
  ],
  "logging_steps": 5,
  "max_steps": 105,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 100971315986432.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}