{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.6923076923076925,
  "eval_steps": 10,
  "global_step": 140,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 21.498493194580078,
      "learning_rate": 0.00019470198675496689,
      "loss": 14.4551,
      "mean_token_accuracy": 0.6612499989569187,
      "num_tokens": 49120.0,
      "step": 10
    },
    {
      "epoch": 0.19230769230769232,
      "eval_loss": 1.0248602628707886,
      "eval_mean_token_accuracy": 0.8969230560156015,
      "eval_num_tokens": 49120.0,
      "eval_runtime": 17.9265,
      "eval_samples_per_second": 2.789,
      "eval_steps_per_second": 0.725,
      "step": 10
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 6.777862071990967,
      "learning_rate": 0.00018145695364238411,
      "loss": 2.185,
      "mean_token_accuracy": 0.9445000052452087,
      "num_tokens": 98240.0,
      "step": 20
    },
    {
      "epoch": 0.38461538461538464,
      "eval_loss": 0.13720029592514038,
      "eval_mean_token_accuracy": 0.9761538688953106,
      "eval_num_tokens": 98240.0,
      "eval_runtime": 17.8731,
      "eval_samples_per_second": 2.797,
      "eval_steps_per_second": 0.727,
      "step": 20
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 4.832788467407227,
      "learning_rate": 0.00016821192052980132,
      "loss": 0.2858,
      "mean_token_accuracy": 0.9777500212192536,
      "num_tokens": 147360.0,
      "step": 30
    },
    {
      "epoch": 0.5769230769230769,
      "eval_loss": 0.03728929162025452,
      "eval_mean_token_accuracy": 0.9765384839131281,
      "eval_num_tokens": 147360.0,
      "eval_runtime": 18.2308,
      "eval_samples_per_second": 2.743,
      "eval_steps_per_second": 0.713,
      "step": 30
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 6.547402858734131,
      "learning_rate": 0.00015496688741721855,
      "loss": 0.1824,
      "mean_token_accuracy": 0.9781250208616257,
      "num_tokens": 196480.0,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 0.05031127855181694,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 196480.0,
      "eval_runtime": 17.8889,
      "eval_samples_per_second": 2.795,
      "eval_steps_per_second": 0.727,
      "step": 40
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 4.728590965270996,
      "learning_rate": 0.00014172185430463575,
      "loss": 0.1721,
      "mean_token_accuracy": 0.9783750206232071,
      "num_tokens": 245600.0,
      "step": 50
    },
    {
      "epoch": 0.9615384615384616,
      "eval_loss": 0.04732300713658333,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 245600.0,
      "eval_runtime": 18.0023,
      "eval_samples_per_second": 2.777,
      "eval_steps_per_second": 0.722,
      "step": 50
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 1.990704894065857,
      "learning_rate": 0.00012847682119205298,
      "loss": 0.1396,
      "mean_token_accuracy": 0.9791250199079513,
      "num_tokens": 294413.0,
      "step": 60
    },
    {
      "epoch": 1.1538461538461537,
      "eval_loss": 0.040356263518333435,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 294413.0,
      "eval_runtime": 17.8824,
      "eval_samples_per_second": 2.796,
      "eval_steps_per_second": 0.727,
      "step": 60
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 0.9954031109809875,
      "learning_rate": 0.0001152317880794702,
      "loss": 0.1391,
      "mean_token_accuracy": 0.9780000209808349,
      "num_tokens": 343533.0,
      "step": 70
    },
    {
      "epoch": 1.3461538461538463,
      "eval_loss": 0.03649154305458069,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 343533.0,
      "eval_runtime": 18.0872,
      "eval_samples_per_second": 2.764,
      "eval_steps_per_second": 0.719,
      "step": 70
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.5908142328262329,
      "learning_rate": 0.00010198675496688744,
      "loss": 0.1376,
      "mean_token_accuracy": 0.9785000190138817,
      "num_tokens": 392653.0,
      "step": 80
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 0.03940116986632347,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 392653.0,
      "eval_runtime": 17.8649,
      "eval_samples_per_second": 2.799,
      "eval_steps_per_second": 0.728,
      "step": 80
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 1.166898250579834,
      "learning_rate": 8.874172185430464e-05,
      "loss": 0.1379,
      "mean_token_accuracy": 0.9782500207424164,
      "num_tokens": 441773.0,
      "step": 90
    },
    {
      "epoch": 1.7307692307692308,
      "eval_loss": 0.03707318380475044,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 441773.0,
      "eval_runtime": 18.0365,
      "eval_samples_per_second": 2.772,
      "eval_steps_per_second": 0.721,
      "step": 90
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.38323092460632324,
      "learning_rate": 7.549668874172185e-05,
      "loss": 0.1415,
      "mean_token_accuracy": 0.9765000149607659,
      "num_tokens": 490893.0,
      "step": 100
    },
    {
      "epoch": 1.9230769230769231,
      "eval_loss": 0.036779891699552536,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 490893.0,
      "eval_runtime": 17.8928,
      "eval_samples_per_second": 2.794,
      "eval_steps_per_second": 0.727,
      "step": 100
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 1.039596676826477,
      "learning_rate": 6.225165562913907e-05,
      "loss": 0.1391,
      "mean_token_accuracy": 0.9800000190734863,
      "num_tokens": 539706.0,
      "step": 110
    },
    {
      "epoch": 2.1153846153846154,
      "eval_loss": 0.036059826612472534,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 539706.0,
      "eval_runtime": 17.9952,
      "eval_samples_per_second": 2.779,
      "eval_steps_per_second": 0.722,
      "step": 110
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.8448509573936462,
      "learning_rate": 4.900662251655629e-05,
      "loss": 0.1347,
      "mean_token_accuracy": 0.9772500187158585,
      "num_tokens": 588826.0,
      "step": 120
    },
    {
      "epoch": 2.3076923076923075,
      "eval_loss": 0.035431090742349625,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 588826.0,
      "eval_runtime": 17.9637,
      "eval_samples_per_second": 2.783,
      "eval_steps_per_second": 0.724,
      "step": 120
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.932663083076477,
      "learning_rate": 3.576158940397351e-05,
      "loss": 0.1377,
      "mean_token_accuracy": 0.9767500206828117,
      "num_tokens": 637946.0,
      "step": 130
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.040411509573459625,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 637946.0,
      "eval_runtime": 18.0796,
      "eval_samples_per_second": 2.766,
      "eval_steps_per_second": 0.719,
      "step": 130
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.6561686992645264,
      "learning_rate": 2.2516556291390732e-05,
      "loss": 0.1378,
      "mean_token_accuracy": 0.9785000205039978,
      "num_tokens": 687066.0,
      "step": 140
    },
    {
      "epoch": 2.6923076923076925,
      "eval_loss": 0.0409255214035511,
      "eval_mean_token_accuracy": 0.9769230989309458,
      "eval_num_tokens": 687066.0,
      "eval_runtime": 17.9047,
      "eval_samples_per_second": 2.793,
      "eval_steps_per_second": 0.726,
      "step": 140
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7865881654018112e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}