{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 1,
  "global_step": 157,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.97,
      "grad_norm": 1.752838134765625,
      "learning_rate": 0.00013103448275862068,
      "loss": 1.9056,
      "step": 19
    },
    {
      "epoch": 0.97,
      "eval_loss": 1.2314808368682861,
      "eval_runtime": 41.0888,
      "eval_samples_per_second": 3.407,
      "eval_steps_per_second": 3.407,
      "step": 19
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.3157634139060974,
      "learning_rate": 0.000197828447339848,
      "loss": 0.7969,
      "step": 39
    },
    {
      "epoch": 1.99,
      "eval_loss": 0.6621134877204895,
      "eval_runtime": 41.1024,
      "eval_samples_per_second": 3.406,
      "eval_steps_per_second": 3.406,
      "step": 39
    },
    {
      "epoch": 2.96,
      "grad_norm": 0.46234801411628723,
      "learning_rate": 0.0001937024972855592,
      "loss": 0.564,
      "step": 58
    },
    {
      "epoch": 2.96,
      "eval_loss": 0.5718151330947876,
      "eval_runtime": 41.1043,
      "eval_samples_per_second": 3.406,
      "eval_steps_per_second": 3.406,
      "step": 58
    },
    {
      "epoch": 3.97,
      "grad_norm": 0.5510583519935608,
      "learning_rate": 0.00018935939196525515,
      "loss": 0.3962,
      "step": 78
    },
    {
      "epoch": 3.97,
      "eval_loss": 0.5229483246803284,
      "eval_runtime": 41.1671,
      "eval_samples_per_second": 3.401,
      "eval_steps_per_second": 3.401,
      "step": 78
    },
    {
      "epoch": 4.99,
      "grad_norm": 4.286730766296387,
      "learning_rate": 0.00018501628664495113,
      "loss": 0.3022,
      "step": 98
    },
    {
      "epoch": 4.99,
      "eval_loss": 0.5116992592811584,
      "eval_runtime": 41.1636,
      "eval_samples_per_second": 3.401,
      "eval_steps_per_second": 3.401,
      "step": 98
    },
    {
      "epoch": 5.96,
      "grad_norm": 0.6076149940490723,
      "learning_rate": 0.00018089033659066234,
      "loss": 0.2504,
      "step": 117
    },
    {
      "epoch": 5.96,
      "eval_loss": 0.5122354030609131,
      "eval_runtime": 41.1631,
      "eval_samples_per_second": 3.401,
      "eval_steps_per_second": 3.401,
      "step": 117
    },
    {
      "epoch": 6.98,
      "grad_norm": 0.5720144510269165,
      "learning_rate": 0.0001765472312703583,
      "loss": 0.1934,
      "step": 137
    },
    {
      "epoch": 6.98,
      "eval_loss": 0.5105646252632141,
      "eval_runtime": 41.1792,
      "eval_samples_per_second": 3.4,
      "eval_steps_per_second": 3.4,
      "step": 137
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.6266606450080872,
      "learning_rate": 0.0001722041259500543,
      "loss": 0.1632,
      "step": 157
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.523033082485199,
      "eval_runtime": 41.2092,
      "eval_samples_per_second": 3.397,
      "eval_steps_per_second": 3.397,
      "step": 157
    }
  ],
  "logging_steps": 1,
  "max_steps": 950,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 1,
  "total_flos": 5.266736889053184e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}