{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3192612137203166,
  "eval_steps": 25,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "grad_norm": 1.4790619611740112,
      "learning_rate": 2.3797595190380762e-05,
      "loss": 0.9713,
      "step": 25
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.7396946549415588,
      "eval_runtime": 75.3443,
      "eval_samples_per_second": 2.535,
      "eval_steps_per_second": 0.319,
      "step": 25
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.359966516494751,
      "learning_rate": 2.2545090180360722e-05,
      "loss": 0.6963,
      "step": 50
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.6846508979797363,
      "eval_runtime": 75.2251,
      "eval_samples_per_second": 2.539,
      "eval_steps_per_second": 0.319,
      "step": 50
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.3525006771087646,
      "learning_rate": 2.1292585170340683e-05,
      "loss": 0.6809,
      "step": 75
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.6628683805465698,
      "eval_runtime": 75.0453,
      "eval_samples_per_second": 2.545,
      "eval_steps_per_second": 0.32,
      "step": 75
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.3805192708969116,
      "learning_rate": 2.0040080160320643e-05,
      "loss": 0.6596,
      "step": 100
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.6511640548706055,
      "eval_runtime": 75.0686,
      "eval_samples_per_second": 2.544,
      "eval_steps_per_second": 0.32,
      "step": 100
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.2107349634170532,
      "learning_rate": 1.87875751503006e-05,
      "loss": 0.6552,
      "step": 125
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.6428065896034241,
      "eval_runtime": 75.0266,
      "eval_samples_per_second": 2.546,
      "eval_steps_per_second": 0.32,
      "step": 125
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.327737808227539,
      "learning_rate": 1.7535070140280564e-05,
      "loss": 0.6363,
      "step": 150
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.6354109048843384,
      "eval_runtime": 75.1344,
      "eval_samples_per_second": 2.542,
      "eval_steps_per_second": 0.319,
      "step": 150
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.2277601957321167,
      "learning_rate": 1.628256513026052e-05,
      "loss": 0.6441,
      "step": 175
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.6307269930839539,
      "eval_runtime": 75.0535,
      "eval_samples_per_second": 2.545,
      "eval_steps_per_second": 0.32,
      "step": 175
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.3676594495773315,
      "learning_rate": 1.5030060120240483e-05,
      "loss": 0.6317,
      "step": 200
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.6252032518386841,
      "eval_runtime": 75.0153,
      "eval_samples_per_second": 2.546,
      "eval_steps_per_second": 0.32,
      "step": 200
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.2495057582855225,
      "learning_rate": 1.3777555110220442e-05,
      "loss": 0.6222,
      "step": 225
    },
    {
      "epoch": 0.59,
      "eval_loss": 0.6214942336082458,
      "eval_runtime": 75.0103,
      "eval_samples_per_second": 2.546,
      "eval_steps_per_second": 0.32,
      "step": 225
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.2535713911056519,
      "learning_rate": 1.25250501002004e-05,
      "loss": 0.6127,
      "step": 250
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.6172995567321777,
      "eval_runtime": 75.023,
      "eval_samples_per_second": 2.546,
      "eval_steps_per_second": 0.32,
      "step": 250
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.3162541389465332,
      "learning_rate": 1.1272545090180361e-05,
      "loss": 0.6017,
      "step": 275
    },
    {
      "epoch": 0.73,
      "eval_loss": 0.614778459072113,
      "eval_runtime": 75.3906,
      "eval_samples_per_second": 2.533,
      "eval_steps_per_second": 0.318,
      "step": 275
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.3146350383758545,
      "learning_rate": 1.0020040080160322e-05,
      "loss": 0.6201,
      "step": 300
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.6113688945770264,
      "eval_runtime": 75.4374,
      "eval_samples_per_second": 2.532,
      "eval_steps_per_second": 0.318,
      "step": 300
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.311963438987732,
      "learning_rate": 8.767535070140282e-06,
      "loss": 0.5961,
      "step": 325
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.6093372702598572,
      "eval_runtime": 75.3949,
      "eval_samples_per_second": 2.533,
      "eval_steps_per_second": 0.318,
      "step": 325
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.3103692531585693,
      "learning_rate": 7.515030060120242e-06,
      "loss": 0.6044,
      "step": 350
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.606224775314331,
      "eval_runtime": 75.3575,
      "eval_samples_per_second": 2.535,
      "eval_steps_per_second": 0.318,
      "step": 350
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.4562299251556396,
      "learning_rate": 6.2625250501002e-06,
      "loss": 0.6064,
      "step": 375
    },
    {
      "epoch": 0.99,
      "eval_loss": 0.6035271286964417,
      "eval_runtime": 75.3932,
      "eval_samples_per_second": 2.533,
      "eval_steps_per_second": 0.318,
      "step": 375
    },
    {
      "epoch": 1.06,
      "grad_norm": 1.3848822116851807,
      "learning_rate": 5.010020040080161e-06,
      "loss": 0.5763,
      "step": 400
    },
    {
      "epoch": 1.06,
      "eval_loss": 0.6053218841552734,
      "eval_runtime": 75.3983,
      "eval_samples_per_second": 2.533,
      "eval_steps_per_second": 0.318,
      "step": 400
    },
    {
      "epoch": 1.12,
      "grad_norm": 1.466248631477356,
      "learning_rate": 3.757515030060121e-06,
      "loss": 0.5589,
      "step": 425
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.6024670600891113,
      "eval_runtime": 75.3786,
      "eval_samples_per_second": 2.534,
      "eval_steps_per_second": 0.318,
      "step": 425
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.4298962354660034,
      "learning_rate": 2.5050100200400804e-06,
      "loss": 0.5595,
      "step": 450
    },
    {
      "epoch": 1.19,
      "eval_loss": 0.6021297574043274,
      "eval_runtime": 75.3606,
      "eval_samples_per_second": 2.534,
      "eval_steps_per_second": 0.318,
      "step": 450
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.395114779472351,
      "learning_rate": 1.2525050100200402e-06,
      "loss": 0.5525,
      "step": 475
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.6009101867675781,
      "eval_runtime": 75.3723,
      "eval_samples_per_second": 2.534,
      "eval_steps_per_second": 0.318,
      "step": 475
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.4652175903320312,
      "learning_rate": 0.0,
      "loss": 0.5541,
      "step": 500
    },
    {
      "epoch": 1.32,
      "eval_loss": 0.6004253625869751,
      "eval_runtime": 75.3761,
      "eval_samples_per_second": 2.534,
      "eval_steps_per_second": 0.318,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 250,
  "total_flos": 7.87141357369344e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}