{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006135345726731701,
  "eval_steps": 20,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012270691453463403,
      "grad_norm": 0.07455573976039886,
      "learning_rate": 0.00019981588314717073,
      "loss": 2.6233,
      "step": 20
    },
    {
      "epoch": 0.0012270691453463403,
      "eval_loss": 2.2018821239471436,
      "eval_runtime": 23.583,
      "eval_samples_per_second": 4.24,
      "eval_steps_per_second": 0.551,
      "step": 20
    },
    {
      "epoch": 0.0024541382906926807,
      "grad_norm": 0.0842185765504837,
      "learning_rate": 0.00019957039401006504,
      "loss": 2.1109,
      "step": 40
    },
    {
      "epoch": 0.0024541382906926807,
      "eval_loss": 1.9889661073684692,
      "eval_runtime": 23.7122,
      "eval_samples_per_second": 4.217,
      "eval_steps_per_second": 0.548,
      "step": 40
    },
    {
      "epoch": 0.003681207436039021,
      "grad_norm": 0.1341264545917511,
      "learning_rate": 0.00019932490487295938,
      "loss": 1.9978,
      "step": 60
    },
    {
      "epoch": 0.003681207436039021,
      "eval_loss": 1.8857674598693848,
      "eval_runtime": 24.0663,
      "eval_samples_per_second": 4.155,
      "eval_steps_per_second": 0.54,
      "step": 60
    },
    {
      "epoch": 0.004908276581385361,
      "grad_norm": 0.13033552467823029,
      "learning_rate": 0.00019907941573585368,
      "loss": 1.8977,
      "step": 80
    },
    {
      "epoch": 0.004908276581385361,
      "eval_loss": 1.8246879577636719,
      "eval_runtime": 23.578,
      "eval_samples_per_second": 4.241,
      "eval_steps_per_second": 0.551,
      "step": 80
    },
    {
      "epoch": 0.006135345726731701,
      "grad_norm": 0.11537110805511475,
      "learning_rate": 0.00019883392659874802,
      "loss": 1.8383,
      "step": 100
    },
    {
      "epoch": 0.006135345726731701,
      "eval_loss": 1.7761367559432983,
      "eval_runtime": 23.7248,
      "eval_samples_per_second": 4.215,
      "eval_steps_per_second": 0.548,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 16299,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "total_flos": 6919962687590400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}