trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 17.999285459092533,
  "global_step": 5038,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 1.2306,
      "step": 128
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 1.0373,
      "step": 256
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.983,
      "step": 384
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.97,
      "step": 512
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0002,
      "loss": 0.9417,
      "step": 640
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0002,
      "loss": 0.9256,
      "step": 768
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0002,
      "loss": 0.9053,
      "step": 896
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.0002,
      "loss": 0.8893,
      "step": 1024
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.0002,
      "loss": 0.8817,
      "step": 1152
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.0002,
      "loss": 0.8662,
      "step": 1280
    },
    {
      "epoch": 5.03,
      "learning_rate": 0.0002,
      "loss": 0.8612,
      "step": 1408
    },
    {
      "epoch": 5.49,
      "learning_rate": 0.0002,
      "loss": 0.8391,
      "step": 1536
    },
    {
      "epoch": 5.94,
      "learning_rate": 0.0002,
      "loss": 0.8434,
      "step": 1664
    },
    {
      "epoch": 6.4,
      "learning_rate": 0.0002,
      "loss": 0.8167,
      "step": 1792
    },
    {
      "epoch": 6.86,
      "learning_rate": 0.0002,
      "loss": 0.8239,
      "step": 1920
    },
    {
      "epoch": 7.32,
      "learning_rate": 0.0002,
      "loss": 0.8046,
      "step": 2048
    },
    {
      "epoch": 7.77,
      "learning_rate": 0.0002,
      "loss": 0.8053,
      "step": 2176
    },
    {
      "epoch": 8.23,
      "learning_rate": 0.0002,
      "loss": 0.7903,
      "step": 2304
    },
    {
      "epoch": 8.69,
      "learning_rate": 0.0002,
      "loss": 0.7868,
      "step": 2432
    },
    {
      "epoch": 9.15,
      "learning_rate": 0.0002,
      "loss": 0.7763,
      "step": 2560
    },
    {
      "epoch": 9.6,
      "learning_rate": 0.0002,
      "loss": 0.7658,
      "step": 2688
    },
    {
      "epoch": 10.06,
      "learning_rate": 0.0002,
      "loss": 0.7669,
      "step": 2816
    },
    {
      "epoch": 10.52,
      "learning_rate": 0.0002,
      "loss": 0.7443,
      "step": 2944
    },
    {
      "epoch": 10.98,
      "learning_rate": 0.0002,
      "loss": 0.7597,
      "step": 3072
    },
    {
      "epoch": 11.43,
      "learning_rate": 0.0002,
      "loss": 0.7313,
      "step": 3200
    },
    {
      "epoch": 11.89,
      "learning_rate": 0.0002,
      "loss": 0.7388,
      "step": 3328
    },
    {
      "epoch": 12.35,
      "learning_rate": 0.0002,
      "loss": 0.7202,
      "step": 3456
    },
    {
      "epoch": 12.8,
      "learning_rate": 0.0002,
      "loss": 0.7231,
      "step": 3584
    },
    {
      "epoch": 13.26,
      "learning_rate": 0.0002,
      "loss": 0.711,
      "step": 3712
    },
    {
      "epoch": 13.72,
      "learning_rate": 0.0002,
      "loss": 0.7066,
      "step": 3840
    },
    {
      "epoch": 14.18,
      "learning_rate": 0.0002,
      "loss": 0.7015,
      "step": 3968
    },
    {
      "epoch": 14.63,
      "learning_rate": 0.0002,
      "loss": 0.6885,
      "step": 4096
    },
    {
      "epoch": 15.09,
      "learning_rate": 0.0002,
      "loss": 0.6938,
      "step": 4224
    },
    {
      "epoch": 15.55,
      "learning_rate": 0.0002,
      "loss": 0.6753,
      "step": 4352
    },
    {
      "epoch": 16.01,
      "learning_rate": 0.0002,
      "loss": 0.6863,
      "step": 4480
    },
    {
      "epoch": 16.46,
      "learning_rate": 0.0002,
      "loss": 0.6546,
      "step": 4608
    },
    {
      "epoch": 16.92,
      "learning_rate": 0.0002,
      "loss": 0.6739,
      "step": 4736
    },
    {
      "epoch": 17.38,
      "learning_rate": 0.0002,
      "loss": 0.6436,
      "step": 4864
    },
    {
      "epoch": 17.83,
      "learning_rate": 0.0002,
      "loss": 0.6589,
      "step": 4992
    }
  ],
  "max_steps": 5580,
  "num_train_epochs": 20,
  "total_flos": 2.989931152539648e+16,
  "trial_name": null,
  "trial_params": null
}