trainer_state.json

{
  "best_metric": 0.7751432587577413,
  "best_model_checkpoint": "2-en-de-**1-wikispan-*unsup-ensemble-last-64-768-6*-64-768-3e-5-8600/checkpoint-1800**-64-128-3e-5-2600/checkpoint-2501",
  "epoch": 1.0,
  "global_step": 2501,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 1.5e-05,
      "loss": 33.9859,
      "step": 100
    },
    {
      "epoch": 0.08,
      "learning_rate": 3e-05,
      "loss": 21.6865,
      "step": 200
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.875e-05,
      "loss": 20.6193,
      "step": 300
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.75e-05,
      "loss": 17.9514,
      "step": 400
    },
    {
      "epoch": 0.2,
      "learning_rate": 2.625e-05,
      "loss": 17.5616,
      "step": 500
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.5e-05,
      "loss": 16.6805,
      "step": 600
    },
    {
      "epoch": 0.28,
      "learning_rate": 2.3749999999999998e-05,
      "loss": 15.423,
      "step": 700
    },
    {
      "epoch": 0.32,
      "learning_rate": 2.25e-05,
      "loss": 14.8845,
      "step": 800
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.125e-05,
      "loss": 14.5805,
      "step": 900
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 14.6016,
      "step": 1000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 13.951,
      "step": 1100
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 12.9799,
      "step": 1200
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.625e-05,
      "loss": 12.8452,
      "step": 1300
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.5e-05,
      "loss": 13.2246,
      "step": 1400
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.375e-05,
      "loss": 12.2712,
      "step": 1500
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.25e-05,
      "loss": 11.9556,
      "step": 1600
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.125e-05,
      "loss": 11.5849,
      "step": 1700
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.999999999999999e-06,
      "loss": 11.5427,
      "step": 1800
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.750000000000001e-06,
      "loss": 11.3689,
      "step": 1900
    },
    {
      "epoch": 0.8,
      "learning_rate": 7.5e-06,
      "loss": 10.4619,
      "step": 2000
    },
    {
      "epoch": 0.84,
      "learning_rate": 6.25e-06,
      "loss": 11.2686,
      "step": 2100
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 10.772,
      "step": 2200
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.75e-06,
      "loss": 11.0352,
      "step": 2300
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.4999999999999998e-06,
      "loss": 10.7418,
      "step": 2400
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.2499999999999999e-06,
      "loss": 10.8549,
      "step": 2500
    }
  ],
  "max_steps": 2600,
  "num_train_epochs": 2,
  "total_flos": 165708798492672.0,
  "trial_name": null,
  "trial_params": null
}