{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.948148148148148,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3,
      "learning_rate": 6.25e-05,
      "loss": 4.423,
      "step": 5
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.995728791936504e-05,
      "loss": 3.358,
      "step": 10
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.947761466636014e-05,
      "loss": 2.8568,
      "step": 15
    },
    {
      "epoch": 0.95,
      "eval_loss": 2.593674421310425,
      "eval_runtime": 7.3362,
      "eval_samples_per_second": 61.613,
      "eval_steps_per_second": 15.403,
      "step": 16
    },
    {
      "epoch": 1.24,
      "learning_rate": 9.847001329696653e-05,
      "loss": 3.0543,
      "step": 20
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.694523495787149e-05,
      "loss": 2.3947,
      "step": 25
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.491954909459895e-05,
      "loss": 2.2487,
      "step": 30
    },
    {
      "epoch": 1.95,
      "eval_loss": 2.1050145626068115,
      "eval_runtime": 7.5204,
      "eval_samples_per_second": 60.104,
      "eval_steps_per_second": 15.026,
      "step": 32
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.241456985587868e-05,
      "loss": 2.4987,
      "step": 35
    },
    {
      "epoch": 2.47,
      "learning_rate": 8.945702546981969e-05,
      "loss": 2.0039,
      "step": 40
    },
    {
      "epoch": 2.77,
      "learning_rate": 8.60784730526531e-05,
      "loss": 1.9011,
      "step": 45
    },
    {
      "epoch": 2.95,
      "eval_loss": 1.80819571018219,
      "eval_runtime": 7.4002,
      "eval_samples_per_second": 61.079,
      "eval_steps_per_second": 15.27,
      "step": 48
    },
    {
      "epoch": 3.12,
      "learning_rate": 8.231496189304704e-05,
      "loss": 2.1532,
      "step": 50
    },
    {
      "epoch": 3.41,
      "learning_rate": 7.820664880476256e-05,
      "loss": 1.7384,
      "step": 55
    },
    {
      "epoch": 3.71,
      "learning_rate": 7.379736965185368e-05,
      "loss": 1.6837,
      "step": 60
    },
    {
      "epoch": 3.95,
      "eval_loss": 1.6178017854690552,
      "eval_runtime": 7.6447,
      "eval_samples_per_second": 59.126,
      "eval_steps_per_second": 14.781,
      "step": 64
    },
    {
      "epoch": 4.06,
      "learning_rate": 6.91341716182545e-05,
      "loss": 1.9079,
      "step": 65
    },
    {
      "epoch": 4.36,
      "learning_rate": 6.426681121245527e-05,
      "loss": 1.5592,
      "step": 70
    },
    {
      "epoch": 4.65,
      "learning_rate": 5.924722336357793e-05,
      "loss": 1.522,
      "step": 75
    },
    {
      "epoch": 4.95,
      "learning_rate": 5.4128967273616625e-05,
      "loss": 1.4887,
      "step": 80
    },
    {
      "epoch": 4.95,
      "eval_loss": 1.4896912574768066,
      "eval_runtime": 7.3322,
      "eval_samples_per_second": 61.646,
      "eval_steps_per_second": 15.411,
      "step": 80
    },
    {
      "epoch": 5.3,
      "learning_rate": 4.8966654938622295e-05,
      "loss": 1.6886,
      "step": 85
    },
    {
      "epoch": 5.59,
      "learning_rate": 4.381536843653262e-05,
      "loss": 1.4043,
      "step": 90
    },
    {
      "epoch": 5.89,
      "learning_rate": 3.87300721992097e-05,
      "loss": 1.3812,
      "step": 95
    },
    {
      "epoch": 5.95,
      "eval_loss": 1.4017213582992554,
      "eval_runtime": 7.2083,
      "eval_samples_per_second": 62.706,
      "eval_steps_per_second": 15.676,
      "step": 96
    },
    {
      "epoch": 6.24,
      "learning_rate": 3.3765026539765834e-05,
      "loss": 1.5807,
      "step": 100
    },
    {
      "epoch": 6.53,
      "learning_rate": 2.8973208692864624e-05,
      "loss": 1.3233,
      "step": 105
    },
    {
      "epoch": 6.83,
      "learning_rate": 2.4405747545519963e-05,
      "loss": 1.2944,
      "step": 110
    },
    {
      "epoch": 6.95,
      "eval_loss": 1.3437390327453613,
      "eval_runtime": 7.1152,
      "eval_samples_per_second": 63.526,
      "eval_steps_per_second": 15.882,
      "step": 112
    },
    {
      "epoch": 7.18,
      "learning_rate": 2.0111378089837956e-05,
      "loss": 1.5042,
      "step": 115
    },
    {
      "epoch": 7.47,
      "learning_rate": 1.6135921418712956e-05,
      "loss": 1.2607,
      "step": 120
    },
    {
      "epoch": 7.77,
      "learning_rate": 1.2521795812943704e-05,
      "loss": 1.2574,
      "step": 125
    },
    {
      "epoch": 7.95,
      "eval_loss": 1.3126752376556396,
      "eval_runtime": 7.3998,
      "eval_samples_per_second": 61.083,
      "eval_steps_per_second": 15.271,
      "step": 128
    },
    {
      "epoch": 8.12,
      "learning_rate": 9.307564136490254e-06,
      "loss": 1.4587,
      "step": 130
    },
    {
      "epoch": 8.41,
      "learning_rate": 6.527522369181655e-06,
      "loss": 1.2245,
      "step": 135
    },
    {
      "epoch": 8.71,
      "learning_rate": 4.2113336672471245e-06,
      "loss": 1.2325,
      "step": 140
    },
    {
      "epoch": 8.95,
      "eval_loss": 1.3008612394332886,
      "eval_runtime": 7.4909,
      "eval_samples_per_second": 60.34,
      "eval_steps_per_second": 15.085,
      "step": 144
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.3837118562592797e-06,
      "loss": 1.4374,
      "step": 145
    },
    {
      "epoch": 9.36,
      "learning_rate": 1.064157733632276e-06,
      "loss": 1.2213,
      "step": 150
    },
    {
      "epoch": 9.65,
      "learning_rate": 2.667509943378721e-07,
      "loss": 1.2213,
      "step": 155
    },
    {
      "epoch": 9.95,
      "learning_rate": 0.0,
      "loss": 1.2223,
      "step": 160
    },
    {
      "epoch": 9.95,
      "eval_loss": 1.2986704111099243,
      "eval_runtime": 7.4403,
      "eval_samples_per_second": 60.751,
      "eval_steps_per_second": 15.188,
      "step": 160
    },
    {
      "epoch": 9.95,
      "step": 160,
      "total_flos": 8.010001210461389e+16,
      "train_loss": 1.8282876014709473,
      "train_runtime": 2056.929,
      "train_samples_per_second": 42.068,
      "train_steps_per_second": 0.078
    }
  ],
  "max_steps": 160,
  "num_train_epochs": 10,
  "total_flos": 8.010001210461389e+16,
  "trial_name": null,
  "trial_params": null
}