{
  "best_metric": 0.11026904731988907,
  "best_model_checkpoint": "outputs/checkpoint-270",
  "epoch": 4.920273348519363,
  "eval_steps": 500,
  "global_step": 270,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36446469248291574,
      "grad_norm": 3.687648057937622,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.3728,
      "step": 20
    },
    {
      "epoch": 0.7289293849658315,
      "grad_norm": 1.6180248260498047,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.2874,
      "step": 40
    },
    {
      "epoch": 0.9840546697038725,
      "eval_loss": 0.1868637651205063,
      "eval_runtime": 47.7396,
      "eval_samples_per_second": 3.519,
      "eval_steps_per_second": 0.44,
      "step": 54
    },
    {
      "epoch": 1.0933940774487472,
      "grad_norm": 0.6248186230659485,
      "learning_rate": 5.8e-06,
      "loss": 0.1808,
      "step": 60
    },
    {
      "epoch": 1.4578587699316627,
      "grad_norm": 0.4216831922531128,
      "learning_rate": 7.800000000000002e-06,
      "loss": 0.134,
      "step": 80
    },
    {
      "epoch": 1.8223234624145785,
      "grad_norm": 1.7952409982681274,
      "learning_rate": 9.800000000000001e-06,
      "loss": 0.1238,
      "step": 100
    },
    {
      "epoch": 1.9863325740318907,
      "eval_loss": 0.1356915980577469,
      "eval_runtime": 47.72,
      "eval_samples_per_second": 3.521,
      "eval_steps_per_second": 0.44,
      "step": 109
    },
    {
      "epoch": 2.1867881548974943,
      "grad_norm": 0.481364369392395,
      "learning_rate": 9.7259191408041e-06,
      "loss": 0.1034,
      "step": 120
    },
    {
      "epoch": 2.55125284738041,
      "grad_norm": 0.6242639422416687,
      "learning_rate": 8.816991413705515e-06,
      "loss": 0.0873,
      "step": 140
    },
    {
      "epoch": 2.9157175398633255,
      "grad_norm": 0.40104907751083374,
      "learning_rate": 7.392557845506433e-06,
      "loss": 0.0863,
      "step": 160
    },
    {
      "epoch": 2.988610478359909,
      "eval_loss": 0.11830399930477142,
      "eval_runtime": 47.7245,
      "eval_samples_per_second": 3.52,
      "eval_steps_per_second": 0.44,
      "step": 164
    },
    {
      "epoch": 3.2801822323462413,
      "grad_norm": 0.659656822681427,
      "learning_rate": 5.644996082651018e-06,
      "loss": 0.0846,
      "step": 180
    },
    {
      "epoch": 3.644646924829157,
      "grad_norm": 0.5718321800231934,
      "learning_rate": 3.8103240247869077e-06,
      "loss": 0.0774,
      "step": 200
    },
    {
      "epoch": 3.990888382687927,
      "eval_loss": 0.11092903465032578,
      "eval_runtime": 47.7405,
      "eval_samples_per_second": 3.519,
      "eval_steps_per_second": 0.44,
      "step": 219
    },
    {
      "epoch": 4.009111617312073,
      "grad_norm": 0.6463438272476196,
      "learning_rate": 2.136324299597474e-06,
      "loss": 0.075,
      "step": 220
    },
    {
      "epoch": 4.373576309794989,
      "grad_norm": 0.44310250878334045,
      "learning_rate": 8.490798459222477e-07,
      "loss": 0.0736,
      "step": 240
    },
    {
      "epoch": 4.738041002277904,
      "grad_norm": 0.5197251439094543,
      "learning_rate": 1.22440160097817e-07,
      "loss": 0.0693,
      "step": 260
    },
    {
      "epoch": 4.920273348519363,
      "eval_loss": 0.11026904731988907,
      "eval_runtime": 47.7998,
      "eval_samples_per_second": 3.515,
      "eval_steps_per_second": 0.439,
      "step": 270
    }
  ],
  "logging_steps": 20,
  "max_steps": 270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.0107505880948736e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}