{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.925925925925926,
  "eval_steps": 1,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 282.9060974121094,
      "learning_rate": 2.5e-05,
      "loss": 8.6695,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6212121212121212,
      "eval_loss": 7.576000690460205,
      "eval_runtime": 8.5741,
      "eval_samples_per_second": 15.395,
      "eval_steps_per_second": 0.816,
      "step": 1
    },
    {
      "epoch": 1.1851851851851851,
      "grad_norm": 62.49760437011719,
      "learning_rate": 5e-05,
      "loss": 1.941,
      "step": 2
    },
    {
      "epoch": 1.1851851851851851,
      "eval_accuracy": 0.5454545454545454,
      "eval_loss": 1.9975844621658325,
      "eval_runtime": 8.5606,
      "eval_samples_per_second": 15.419,
      "eval_steps_per_second": 0.818,
      "step": 2
    },
    {
      "epoch": 2.0,
      "grad_norm": 122.50921630859375,
      "learning_rate": 4.375e-05,
      "loss": 1.3554,
      "step": 3
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3787878787878788,
      "eval_loss": 15.970643997192383,
      "eval_runtime": 8.5528,
      "eval_samples_per_second": 15.434,
      "eval_steps_per_second": 0.818,
      "step": 3
    },
    {
      "epoch": 2.3703703703703702,
      "grad_norm": 142.13168334960938,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 4.8003,
      "step": 4
    },
    {
      "epoch": 2.3703703703703702,
      "eval_accuracy": 0.3787878787878788,
      "eval_loss": 8.323533058166504,
      "eval_runtime": 8.5446,
      "eval_samples_per_second": 15.448,
      "eval_steps_per_second": 0.819,
      "step": 4
    },
    {
      "epoch": 3.0,
      "grad_norm": 234.6129608154297,
      "learning_rate": 3.125e-05,
      "loss": 4.1677,
      "step": 5
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6818181818181818,
      "eval_loss": 1.3907140493392944,
      "eval_runtime": 8.5437,
      "eval_samples_per_second": 15.45,
      "eval_steps_per_second": 0.819,
      "step": 5
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 105.30667877197266,
      "learning_rate": 2.5e-05,
      "loss": 0.8659,
      "step": 6
    },
    {
      "epoch": 3.5555555555555554,
      "eval_accuracy": 0.6515151515151515,
      "eval_loss": 1.0335112810134888,
      "eval_runtime": 8.5605,
      "eval_samples_per_second": 15.42,
      "eval_steps_per_second": 0.818,
      "step": 6
    },
    {
      "epoch": 4.0,
      "grad_norm": 61.28445053100586,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.441,
      "step": 7
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4090909090909091,
      "eval_loss": 2.9479410648345947,
      "eval_runtime": 8.5489,
      "eval_samples_per_second": 15.441,
      "eval_steps_per_second": 0.819,
      "step": 7
    },
    {
      "epoch": 4.7407407407407405,
      "grad_norm": 232.58999633789062,
      "learning_rate": 1.25e-05,
      "loss": 1.5472,
      "step": 8
    },
    {
      "epoch": 4.7407407407407405,
      "eval_accuracy": 0.4090909090909091,
      "eval_loss": 2.773439884185791,
      "eval_runtime": 8.5456,
      "eval_samples_per_second": 15.447,
      "eval_steps_per_second": 0.819,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 73.13507843017578,
      "learning_rate": 6.25e-06,
      "loss": 0.4933,
      "step": 9
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5075757575757576,
      "eval_loss": 1.6608752012252808,
      "eval_runtime": 8.544,
      "eval_samples_per_second": 15.45,
      "eval_steps_per_second": 0.819,
      "step": 9
    },
    {
      "epoch": 5.925925925925926,
      "grad_norm": 179.47828674316406,
      "learning_rate": 0.0,
      "loss": 1.0334,
      "step": 10
    },
    {
      "epoch": 5.925925925925926,
      "eval_accuracy": 0.5681818181818182,
      "eval_loss": 1.06743323802948,
      "eval_runtime": 8.5485,
      "eval_samples_per_second": 15.441,
      "eval_steps_per_second": 0.819,
      "step": 10
    },
    {
      "epoch": 5.925925925925926,
      "step": 10,
      "total_flos": 1.4575787280384e+16,
      "train_loss": 2.531478080153465,
      "train_runtime": 897.3522,
      "train_samples_per_second": 5.862,
      "train_steps_per_second": 0.011
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.4575787280384e+16,
  "train_batch_size": 20,
  "trial_name": null,
  "trial_params": null
}