{
  "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": 2.04457688331604,
      "learning_rate": 1.267605633802817e-05,
      "loss": 0.8663,
      "step": 54
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.8013755679130554,
      "eval_runtime": 56.6879,
      "eval_samples_per_second": 3.969,
      "eval_steps_per_second": 1.006,
      "step": 54
    },
    {
      "epoch": 0.51,
      "grad_norm": 2.8853890895843506,
      "learning_rate": 2.535211267605634e-05,
      "loss": 0.6963,
      "step": 108
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.5671647191047668,
      "eval_runtime": 56.7063,
      "eval_samples_per_second": 3.968,
      "eval_steps_per_second": 1.005,
      "step": 108
    },
    {
      "epoch": 0.76,
      "grad_norm": 2.1792943477630615,
      "learning_rate": 3.802816901408451e-05,
      "loss": 0.5304,
      "step": 162
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.4987545311450958,
      "eval_runtime": 56.6968,
      "eval_samples_per_second": 3.968,
      "eval_steps_per_second": 1.005,
      "step": 162
    },
    {
      "epoch": 1.01,
      "grad_norm": 1.5920032262802124,
      "learning_rate": 4.992175273865415e-05,
      "loss": 0.5226,
      "step": 216
    },
    {
      "epoch": 1.01,
      "eval_loss": 0.47676804661750793,
      "eval_runtime": 56.7097,
      "eval_samples_per_second": 3.968,
      "eval_steps_per_second": 1.005,
      "step": 216
    },
    {
      "epoch": 1.27,
      "grad_norm": 2.3091418743133545,
      "learning_rate": 4.85133020344288e-05,
      "loss": 0.4447,
      "step": 270
    },
    {
      "epoch": 1.27,
      "eval_loss": 0.4610518515110016,
      "eval_runtime": 56.7044,
      "eval_samples_per_second": 3.968,
      "eval_steps_per_second": 1.005,
      "step": 270
    },
    {
      "epoch": 1.52,
      "grad_norm": 2.401709794998169,
      "learning_rate": 4.710485133020345e-05,
      "loss": 0.4398,
      "step": 324
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.451718807220459,
      "eval_runtime": 56.7057,
      "eval_samples_per_second": 3.968,
      "eval_steps_per_second": 1.005,
      "step": 324
    },
    {
      "epoch": 1.77,
      "grad_norm": 2.706294059753418,
      "learning_rate": 4.569640062597809e-05,
      "loss": 0.4576,
      "step": 378
    },
    {
      "epoch": 1.77,
      "eval_loss": 0.4466288983821869,
      "eval_runtime": 56.7181,
      "eval_samples_per_second": 3.967,
      "eval_steps_per_second": 1.005,
      "step": 378
    },
    {
      "epoch": 2.03,
      "grad_norm": 1.9919854402542114,
      "learning_rate": 4.428794992175274e-05,
      "loss": 0.4311,
      "step": 432
    },
    {
      "epoch": 2.03,
      "eval_loss": 0.4444536566734314,
      "eval_runtime": 56.726,
      "eval_samples_per_second": 3.966,
      "eval_steps_per_second": 1.005,
      "step": 432
    },
    {
      "epoch": 2.28,
      "grad_norm": 2.293198823928833,
      "learning_rate": 4.287949921752739e-05,
      "loss": 0.3878,
      "step": 486
    },
    {
      "epoch": 2.28,
      "eval_loss": 0.4513635039329529,
      "eval_runtime": 56.7175,
      "eval_samples_per_second": 3.967,
      "eval_steps_per_second": 1.005,
      "step": 486
    },
    {
      "epoch": 2.54,
      "grad_norm": 2.474388360977173,
      "learning_rate": 4.1471048513302035e-05,
      "loss": 0.4075,
      "step": 540
    },
    {
      "epoch": 2.54,
      "eval_loss": 0.4425508379936218,
      "eval_runtime": 56.7136,
      "eval_samples_per_second": 3.967,
      "eval_steps_per_second": 1.005,
      "step": 540
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.9115936756134033,
      "learning_rate": 4.0062597809076686e-05,
      "loss": 0.3908,
      "step": 594
    },
    {
      "epoch": 2.79,
      "eval_loss": 0.4394964575767517,
      "eval_runtime": 56.7192,
      "eval_samples_per_second": 3.967,
      "eval_steps_per_second": 1.005,
      "step": 594
    }
  ],
  "logging_steps": 54,
  "max_steps": 2130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 54,
  "total_flos": 4.41642351905833e+16,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}