trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 22.0,
  "global_step": 215380,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 5.4101,
      "step": 3896
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.9e-05,
      "loss": 3.7774,
      "step": 7792
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.85e-05,
      "loss": 3.4166,
      "step": 11688
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.8e-05,
      "loss": 3.2281,
      "step": 15584
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.75e-05,
      "loss": 3.105,
      "step": 19480
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.7e-05,
      "loss": 3.0041,
      "step": 23376
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.6500000000000005e-05,
      "loss": 2.9418,
      "step": 27272
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 2.8758,
      "step": 31168
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.55e-05,
      "loss": 2.8427,
      "step": 35064
    },
    {
      "epoch": 10.0,
      "learning_rate": 4.5e-05,
      "loss": 2.791,
      "step": 38960
    },
    {
      "epoch": 11.0,
      "learning_rate": 4.4500000000000004e-05,
      "loss": 2.7628,
      "step": 42856
    },
    {
      "epoch": 12.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 2.7277,
      "step": 46752
    },
    {
      "epoch": 13.0,
      "learning_rate": 4.35e-05,
      "loss": 2.7199,
      "step": 50648
    },
    {
      "epoch": 14.0,
      "learning_rate": 4.3e-05,
      "loss": 2.6665,
      "step": 54544
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.25e-05,
      "loss": 2.6518,
      "step": 58440
    },
    {
      "epoch": 16.0,
      "learning_rate": 4.2e-05,
      "loss": 2.6451,
      "step": 62336
    },
    {
      "epoch": 17.0,
      "learning_rate": 4.15e-05,
      "loss": 2.6108,
      "step": 66232
    },
    {
      "epoch": 18.0,
      "learning_rate": 4.1e-05,
      "loss": 2.5972,
      "step": 70128
    },
    {
      "epoch": 19.0,
      "learning_rate": 4.05e-05,
      "loss": 2.5859,
      "step": 74024
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 2.9427,
      "step": 78320
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.55e-05,
      "loss": 2.9183,
      "step": 88110
    },
    {
      "epoch": 10.0,
      "learning_rate": 4.5e-05,
      "loss": 2.8809,
      "step": 97900
    },
    {
      "epoch": 11.0,
      "learning_rate": 4.4500000000000004e-05,
      "loss": 2.8521,
      "step": 107690
    },
    {
      "epoch": 12.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 2.8272,
      "step": 117480
    },
    {
      "epoch": 13.0,
      "learning_rate": 4.35e-05,
      "loss": 2.8104,
      "step": 127270
    },
    {
      "epoch": 14.0,
      "learning_rate": 4.3e-05,
      "loss": 2.7856,
      "step": 137060
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.25e-05,
      "loss": 2.7625,
      "step": 146850
    },
    {
      "epoch": 16.0,
      "learning_rate": 4.2e-05,
      "loss": 2.7459,
      "step": 156640
    },
    {
      "epoch": 17.0,
      "learning_rate": 4.15e-05,
      "loss": 2.734,
      "step": 166430
    },
    {
      "epoch": 18.0,
      "learning_rate": 4.1e-05,
      "loss": 2.7209,
      "step": 176220
    },
    {
      "epoch": 19.0,
      "learning_rate": 4.05e-05,
      "loss": 2.7129,
      "step": 186010
    },
    {
      "epoch": 20.0,
      "learning_rate": 4e-05,
      "loss": 2.6932,
      "step": 195800
    },
    {
      "epoch": 21.0,
      "learning_rate": 3.9500000000000005e-05,
      "loss": 2.6885,
      "step": 205590
    },
    {
      "epoch": 22.0,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 2.679,
      "step": 215380
    }
  ],
  "max_steps": 979000,
  "num_train_epochs": 100,
  "total_flos": 5.872346596487864e+23,
  "trial_name": null,
  "trial_params": null
}