{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.624639076034649,
  "global_step": 15000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "learning_rate": 4.98395378690629e-05,
      "loss": 0.4205,
      "step": 500
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.96790757381258e-05,
      "loss": 0.2948,
      "step": 1000
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.951861360718871e-05,
      "loss": 0.2779,
      "step": 1500
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.935815147625161e-05,
      "loss": 0.2354,
      "step": 2000
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.9197689345314504e-05,
      "loss": 0.2299,
      "step": 2500
    },
    {
      "epoch": 1.92,
      "learning_rate": 4.9037227214377404e-05,
      "loss": 0.2296,
      "step": 3000
    },
    {
      "epoch": 2.25,
      "learning_rate": 4.887676508344031e-05,
      "loss": 0.1991,
      "step": 3500
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.871630295250321e-05,
      "loss": 0.1887,
      "step": 4000
    },
    {
      "epoch": 2.89,
      "learning_rate": 4.855584082156611e-05,
      "loss": 0.1911,
      "step": 4500
    },
    {
      "epoch": 3.21,
      "learning_rate": 4.839537869062901e-05,
      "loss": 0.1662,
      "step": 5000
    },
    {
      "epoch": 3.53,
      "learning_rate": 4.823491655969192e-05,
      "loss": 0.1579,
      "step": 5500
    },
    {
      "epoch": 3.85,
      "learning_rate": 4.807445442875482e-05,
      "loss": 0.1623,
      "step": 6000
    },
    {
      "epoch": 4.17,
      "learning_rate": 4.791399229781772e-05,
      "loss": 0.1452,
      "step": 6500
    },
    {
      "epoch": 4.49,
      "learning_rate": 4.775353016688062e-05,
      "loss": 0.1324,
      "step": 7000
    },
    {
      "epoch": 4.81,
      "learning_rate": 4.759306803594352e-05,
      "loss": 0.1362,
      "step": 7500
    },
    {
      "epoch": 5.13,
      "learning_rate": 4.743260590500642e-05,
      "loss": 0.1275,
      "step": 8000
    },
    {
      "epoch": 5.45,
      "learning_rate": 4.727214377406932e-05,
      "loss": 0.114,
      "step": 8500
    },
    {
      "epoch": 5.77,
      "learning_rate": 4.711168164313222e-05,
      "loss": 0.1171,
      "step": 9000
    },
    {
      "epoch": 6.1,
      "learning_rate": 4.695121951219512e-05,
      "loss": 0.1104,
      "step": 9500
    },
    {
      "epoch": 6.42,
      "learning_rate": 4.679075738125803e-05,
      "loss": 0.0965,
      "step": 10000
    },
    {
      "epoch": 6.74,
      "learning_rate": 4.663029525032093e-05,
      "loss": 0.0998,
      "step": 10500
    },
    {
      "epoch": 7.06,
      "learning_rate": 4.646983311938383e-05,
      "loss": 0.0983,
      "step": 11000
    },
    {
      "epoch": 7.38,
      "learning_rate": 4.630937098844673e-05,
      "loss": 0.0817,
      "step": 11500
    },
    {
      "epoch": 7.7,
      "learning_rate": 4.614890885750963e-05,
      "loss": 0.0854,
      "step": 12000
    },
    {
      "epoch": 8.02,
      "learning_rate": 4.598844672657253e-05,
      "loss": 0.0867,
      "step": 12500
    },
    {
      "epoch": 8.34,
      "learning_rate": 4.582798459563543e-05,
      "loss": 0.0698,
      "step": 13000
    },
    {
      "epoch": 8.66,
      "learning_rate": 4.566752246469833e-05,
      "loss": 0.0722,
      "step": 13500
    },
    {
      "epoch": 8.98,
      "learning_rate": 4.550706033376124e-05,
      "loss": 0.0752,
      "step": 14000
    },
    {
      "epoch": 9.3,
      "learning_rate": 4.534659820282414e-05,
      "loss": 0.0593,
      "step": 14500
    },
    {
      "epoch": 9.62,
      "learning_rate": 4.518613607188704e-05,
      "loss": 0.0617,
      "step": 15000
    }
  ],
  "max_steps": 155800,
  "num_train_epochs": 100,
  "total_flos": 5.217329001215232e+17,
  "trial_name": null,
  "trial_params": null
}