{
  "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": 2.7450246810913086,
      "learning_rate": 1.267605633802817e-05,
      "loss": 1.3679,
      "step": 54
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.3037077188491821,
      "eval_runtime": 54.835,
      "eval_samples_per_second": 4.103,
      "eval_steps_per_second": 1.039,
      "step": 54
    },
    {
      "epoch": 0.51,
      "grad_norm": 2.912886619567871,
      "learning_rate": 2.535211267605634e-05,
      "loss": 1.1493,
      "step": 108
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.0368825197219849,
      "eval_runtime": 54.8575,
      "eval_samples_per_second": 4.102,
      "eval_steps_per_second": 1.039,
      "step": 108
    },
    {
      "epoch": 0.76,
      "grad_norm": 3.442351818084717,
      "learning_rate": 3.802816901408451e-05,
      "loss": 1.0122,
      "step": 162
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.9716250896453857,
      "eval_runtime": 54.8562,
      "eval_samples_per_second": 4.102,
      "eval_steps_per_second": 1.039,
      "step": 162
    },
    {
      "epoch": 1.01,
      "grad_norm": 3.029773712158203,
      "learning_rate": 4.992175273865415e-05,
      "loss": 0.9696,
      "step": 216
    },
    {
      "epoch": 1.01,
      "eval_loss": 0.9415866136550903,
      "eval_runtime": 54.8671,
      "eval_samples_per_second": 4.101,
      "eval_steps_per_second": 1.039,
      "step": 216
    },
    {
      "epoch": 1.27,
      "grad_norm": 2.115858316421509,
      "learning_rate": 4.85133020344288e-05,
      "loss": 0.9081,
      "step": 270
    },
    {
      "epoch": 1.27,
      "eval_loss": 0.9336543679237366,
      "eval_runtime": 54.8599,
      "eval_samples_per_second": 4.101,
      "eval_steps_per_second": 1.039,
      "step": 270
    },
    {
      "epoch": 1.52,
      "grad_norm": 2.4567699432373047,
      "learning_rate": 4.710485133020345e-05,
      "loss": 0.9124,
      "step": 324
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.9218262434005737,
      "eval_runtime": 54.864,
      "eval_samples_per_second": 4.101,
      "eval_steps_per_second": 1.039,
      "step": 324
    },
    {
      "epoch": 1.77,
      "grad_norm": 2.404057025909424,
      "learning_rate": 4.569640062597809e-05,
      "loss": 0.9196,
      "step": 378
    },
    {
      "epoch": 1.77,
      "eval_loss": 0.9132243394851685,
      "eval_runtime": 54.8557,
      "eval_samples_per_second": 4.102,
      "eval_steps_per_second": 1.039,
      "step": 378
    },
    {
      "epoch": 2.03,
      "grad_norm": 2.8699450492858887,
      "learning_rate": 4.428794992175274e-05,
      "loss": 0.9057,
      "step": 432
    },
    {
      "epoch": 2.03,
      "eval_loss": 0.911640465259552,
      "eval_runtime": 54.8441,
      "eval_samples_per_second": 4.103,
      "eval_steps_per_second": 1.039,
      "step": 432
    },
    {
      "epoch": 2.28,
      "grad_norm": 3.179083824157715,
      "learning_rate": 4.287949921752739e-05,
      "loss": 0.8291,
      "step": 486
    },
    {
      "epoch": 2.28,
      "eval_loss": 0.9131847620010376,
      "eval_runtime": 54.8559,
      "eval_samples_per_second": 4.102,
      "eval_steps_per_second": 1.039,
      "step": 486
    },
    {
      "epoch": 2.54,
      "grad_norm": 3.405515193939209,
      "learning_rate": 4.1471048513302035e-05,
      "loss": 0.8384,
      "step": 540
    },
    {
      "epoch": 2.54,
      "eval_loss": 0.9128340482711792,
      "eval_runtime": 54.8514,
      "eval_samples_per_second": 4.102,
      "eval_steps_per_second": 1.039,
      "step": 540
    },
    {
      "epoch": 2.79,
      "grad_norm": 2.4128098487854004,
      "learning_rate": 4.0062597809076686e-05,
      "loss": 0.8584,
      "step": 594
    },
    {
      "epoch": 2.79,
      "eval_loss": 0.9089391231536865,
      "eval_runtime": 54.8548,
      "eval_samples_per_second": 4.102,
      "eval_steps_per_second": 1.039,
      "step": 594
    }
  ],
  "logging_steps": 54,
  "max_steps": 2130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 54,
  "total_flos": 4.321586242245427e+16,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}