{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.7777777777777777,
  "eval_steps": 25,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "grad_norm": 6.271986484527588,
      "learning_rate": 2.3847695390781562e-05,
      "loss": 0.9018,
      "step": 25
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.6543389558792114,
      "eval_runtime": 33.9351,
      "eval_samples_per_second": 1.798,
      "eval_steps_per_second": 0.236,
      "step": 25
    },
    {
      "epoch": 0.28,
      "grad_norm": 6.715949535369873,
      "learning_rate": 2.2595190380761526e-05,
      "loss": 0.6032,
      "step": 50
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.6594853401184082,
      "eval_runtime": 33.9905,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 50
    },
    {
      "epoch": 0.42,
      "grad_norm": 5.575837135314941,
      "learning_rate": 2.1342685370741483e-05,
      "loss": 0.5277,
      "step": 75
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.6358431577682495,
      "eval_runtime": 33.9553,
      "eval_samples_per_second": 1.796,
      "eval_steps_per_second": 0.236,
      "step": 75
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.133337020874023,
      "learning_rate": 2.0090180360721444e-05,
      "loss": 0.4387,
      "step": 100
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.6550642251968384,
      "eval_runtime": 33.9823,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 100
    },
    {
      "epoch": 0.69,
      "grad_norm": 7.314888954162598,
      "learning_rate": 1.8837675350701404e-05,
      "loss": 0.4692,
      "step": 125
    },
    {
      "epoch": 0.69,
      "eval_loss": 0.6102645993232727,
      "eval_runtime": 33.9789,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 125
    },
    {
      "epoch": 0.83,
      "grad_norm": 5.678727149963379,
      "learning_rate": 1.7585170340681365e-05,
      "loss": 0.4364,
      "step": 150
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.6225055456161499,
      "eval_runtime": 34.038,
      "eval_samples_per_second": 1.792,
      "eval_steps_per_second": 0.235,
      "step": 150
    },
    {
      "epoch": 0.97,
      "grad_norm": 5.900119304656982,
      "learning_rate": 1.633266533066132e-05,
      "loss": 0.4211,
      "step": 175
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.6140083074569702,
      "eval_runtime": 33.9476,
      "eval_samples_per_second": 1.797,
      "eval_steps_per_second": 0.236,
      "step": 175
    },
    {
      "epoch": 1.11,
      "grad_norm": 5.989519119262695,
      "learning_rate": 1.5080160320641284e-05,
      "loss": 0.2941,
      "step": 200
    },
    {
      "epoch": 1.11,
      "eval_loss": 0.6877502799034119,
      "eval_runtime": 33.9316,
      "eval_samples_per_second": 1.798,
      "eval_steps_per_second": 0.236,
      "step": 200
    },
    {
      "epoch": 1.25,
      "grad_norm": 2.5047194957733154,
      "learning_rate": 1.3827655310621242e-05,
      "loss": 0.2562,
      "step": 225
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.7026296257972717,
      "eval_runtime": 33.9018,
      "eval_samples_per_second": 1.799,
      "eval_steps_per_second": 0.236,
      "step": 225
    },
    {
      "epoch": 1.39,
      "grad_norm": 5.878956317901611,
      "learning_rate": 1.2575150300601201e-05,
      "loss": 0.183,
      "step": 250
    },
    {
      "epoch": 1.39,
      "eval_loss": 0.7297132015228271,
      "eval_runtime": 33.9801,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 250
    },
    {
      "epoch": 1.53,
      "grad_norm": 6.625136852264404,
      "learning_rate": 1.1322645290581163e-05,
      "loss": 0.245,
      "step": 275
    },
    {
      "epoch": 1.53,
      "eval_loss": 0.7251414060592651,
      "eval_runtime": 33.928,
      "eval_samples_per_second": 1.798,
      "eval_steps_per_second": 0.236,
      "step": 275
    },
    {
      "epoch": 1.67,
      "grad_norm": 5.02877950668335,
      "learning_rate": 1.0120240480961924e-05,
      "loss": 0.2323,
      "step": 300
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.6742722392082214,
      "eval_runtime": 33.9146,
      "eval_samples_per_second": 1.799,
      "eval_steps_per_second": 0.236,
      "step": 300
    },
    {
      "epoch": 1.81,
      "grad_norm": 3.24204158782959,
      "learning_rate": 8.867735470941884e-06,
      "loss": 0.2426,
      "step": 325
    },
    {
      "epoch": 1.81,
      "eval_loss": 0.6821221709251404,
      "eval_runtime": 33.9704,
      "eval_samples_per_second": 1.796,
      "eval_steps_per_second": 0.235,
      "step": 325
    },
    {
      "epoch": 1.94,
      "grad_norm": 5.852811336517334,
      "learning_rate": 7.615230460921845e-06,
      "loss": 0.2487,
      "step": 350
    },
    {
      "epoch": 1.94,
      "eval_loss": 0.6622934341430664,
      "eval_runtime": 33.9863,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 350
    },
    {
      "epoch": 2.08,
      "grad_norm": 2.1195895671844482,
      "learning_rate": 6.362725450901804e-06,
      "loss": 0.1755,
      "step": 375
    },
    {
      "epoch": 2.08,
      "eval_loss": 0.7334879040718079,
      "eval_runtime": 33.9725,
      "eval_samples_per_second": 1.796,
      "eval_steps_per_second": 0.235,
      "step": 375
    },
    {
      "epoch": 2.22,
      "grad_norm": 2.2676243782043457,
      "learning_rate": 5.110220440881764e-06,
      "loss": 0.1357,
      "step": 400
    },
    {
      "epoch": 2.22,
      "eval_loss": 0.8047094345092773,
      "eval_runtime": 33.8765,
      "eval_samples_per_second": 1.801,
      "eval_steps_per_second": 0.236,
      "step": 400
    },
    {
      "epoch": 2.36,
      "grad_norm": 2.1398653984069824,
      "learning_rate": 3.857715430861724e-06,
      "loss": 0.1517,
      "step": 425
    },
    {
      "epoch": 2.36,
      "eval_loss": 0.8226982355117798,
      "eval_runtime": 33.9677,
      "eval_samples_per_second": 1.796,
      "eval_steps_per_second": 0.236,
      "step": 425
    },
    {
      "epoch": 2.5,
      "grad_norm": 3.4427037239074707,
      "learning_rate": 2.6052104208416833e-06,
      "loss": 0.1596,
      "step": 450
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.7910816073417664,
      "eval_runtime": 33.9419,
      "eval_samples_per_second": 1.797,
      "eval_steps_per_second": 0.236,
      "step": 450
    },
    {
      "epoch": 2.64,
      "grad_norm": 2.103198289871216,
      "learning_rate": 1.3527054108216433e-06,
      "loss": 0.149,
      "step": 475
    },
    {
      "epoch": 2.64,
      "eval_loss": 0.7897896766662598,
      "eval_runtime": 33.9551,
      "eval_samples_per_second": 1.796,
      "eval_steps_per_second": 0.236,
      "step": 475
    },
    {
      "epoch": 2.78,
      "grad_norm": 5.470449447631836,
      "learning_rate": 1.002004008016032e-07,
      "loss": 0.1392,
      "step": 500
    },
    {
      "epoch": 2.78,
      "eval_loss": 0.7936234474182129,
      "eval_runtime": 33.9247,
      "eval_samples_per_second": 1.798,
      "eval_steps_per_second": 0.236,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "total_flos": 2.206098384538829e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}