{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.788732394366197,
  "eval_steps": 54,
  "global_step": 594,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 1.9276669025421143,
      "learning_rate": 1.267605633802817e-05,
      "loss": 1.2779,
      "step": 54
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.222019910812378,
      "eval_runtime": 54.1446,
      "eval_samples_per_second": 4.156,
      "eval_steps_per_second": 1.053,
      "step": 54
    },
    {
      "epoch": 0.51,
      "grad_norm": 2.256397008895874,
      "learning_rate": 2.535211267605634e-05,
      "loss": 1.0801,
      "step": 108
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.9641969203948975,
      "eval_runtime": 54.1532,
      "eval_samples_per_second": 4.155,
      "eval_steps_per_second": 1.053,
      "step": 108
    },
    {
      "epoch": 0.76,
      "grad_norm": 2.1346447467803955,
      "learning_rate": 3.802816901408451e-05,
      "loss": 0.9403,
      "step": 162
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.9075038433074951,
      "eval_runtime": 54.1623,
      "eval_samples_per_second": 4.154,
      "eval_steps_per_second": 1.052,
      "step": 162
    },
    {
      "epoch": 1.01,
      "grad_norm": 3.086144208908081,
      "learning_rate": 4.992175273865415e-05,
      "loss": 0.9201,
      "step": 216
    },
    {
      "epoch": 1.01,
      "eval_loss": 0.888337254524231,
      "eval_runtime": 54.1657,
      "eval_samples_per_second": 4.154,
      "eval_steps_per_second": 1.052,
      "step": 216
    },
    {
      "epoch": 1.27,
      "grad_norm": 2.150139331817627,
      "learning_rate": 4.85133020344288e-05,
      "loss": 0.8568,
      "step": 270
    },
    {
      "epoch": 1.27,
      "eval_loss": 0.8800536394119263,
      "eval_runtime": 54.1792,
      "eval_samples_per_second": 4.153,
      "eval_steps_per_second": 1.052,
      "step": 270
    },
    {
      "epoch": 1.52,
      "grad_norm": 2.6009371280670166,
      "learning_rate": 4.710485133020345e-05,
      "loss": 0.8316,
      "step": 324
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.8666479587554932,
      "eval_runtime": 54.1596,
      "eval_samples_per_second": 4.154,
      "eval_steps_per_second": 1.052,
      "step": 324
    },
    {
      "epoch": 1.77,
      "grad_norm": 3.0447170734405518,
      "learning_rate": 4.569640062597809e-05,
      "loss": 0.863,
      "step": 378
    },
    {
      "epoch": 1.77,
      "eval_loss": 0.8673966526985168,
      "eval_runtime": 54.1837,
      "eval_samples_per_second": 4.153,
      "eval_steps_per_second": 1.052,
      "step": 378
    },
    {
      "epoch": 2.03,
      "grad_norm": 3.341876745223999,
      "learning_rate": 4.428794992175274e-05,
      "loss": 0.8366,
      "step": 432
    },
    {
      "epoch": 2.03,
      "eval_loss": 0.8631945848464966,
      "eval_runtime": 54.1757,
      "eval_samples_per_second": 4.153,
      "eval_steps_per_second": 1.052,
      "step": 432
    },
    {
      "epoch": 2.28,
      "grad_norm": 2.3844501972198486,
      "learning_rate": 4.287949921752739e-05,
      "loss": 0.7709,
      "step": 486
    },
    {
      "epoch": 2.28,
      "eval_loss": 0.8624871373176575,
      "eval_runtime": 54.1608,
      "eval_samples_per_second": 4.154,
      "eval_steps_per_second": 1.052,
      "step": 486
    },
    {
      "epoch": 2.54,
      "grad_norm": 3.0776872634887695,
      "learning_rate": 4.1471048513302035e-05,
      "loss": 0.7905,
      "step": 540
    },
    {
      "epoch": 2.54,
      "eval_loss": 0.864746630191803,
      "eval_runtime": 54.1833,
      "eval_samples_per_second": 4.153,
      "eval_steps_per_second": 1.052,
      "step": 540
    },
    {
      "epoch": 2.79,
      "grad_norm": 2.175617218017578,
      "learning_rate": 4.0062597809076686e-05,
      "loss": 0.7826,
      "step": 594
    },
    {
      "epoch": 2.79,
      "eval_loss": 0.8620367050170898,
      "eval_runtime": 54.1597,
      "eval_samples_per_second": 4.154,
      "eval_steps_per_second": 1.052,
      "step": 594
    }
  ],
  "logging_steps": 54,
  "max_steps": 2130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 54,
  "total_flos": 4.336297242697728e+16,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}