checkpoint-140/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 56.0,
  "eval_steps": 500,
  "global_step": 140,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.6,
      "learning_rate": 0.0005714285714285714,
      "loss": 1.2446,
      "step": 4
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0011428571428571427,
      "loss": 0.9828,
      "step": 8
    },
    {
      "epoch": 4.8,
      "learning_rate": 0.0017142857142857142,
      "loss": 0.7103,
      "step": 12
    },
    {
      "epoch": 6.4,
      "learning_rate": 0.001968253968253968,
      "loss": 0.4313,
      "step": 16
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0019047619047619048,
      "loss": 0.2152,
      "step": 20
    },
    {
      "epoch": 9.6,
      "learning_rate": 0.0018412698412698413,
      "loss": 0.1221,
      "step": 24
    },
    {
      "epoch": 11.2,
      "learning_rate": 0.0017777777777777776,
      "loss": 0.0657,
      "step": 28
    },
    {
      "epoch": 12.8,
      "learning_rate": 0.001746031746031746,
      "loss": 0.0725,
      "step": 32
    },
    {
      "epoch": 14.4,
      "learning_rate": 0.0016825396825396826,
      "loss": 0.0622,
      "step": 36
    },
    {
      "epoch": 16.0,
      "learning_rate": 0.0016190476190476191,
      "loss": 0.0362,
      "step": 40
    },
    {
      "epoch": 17.6,
      "learning_rate": 0.0015555555555555557,
      "loss": 0.0266,
      "step": 44
    },
    {
      "epoch": 19.2,
      "learning_rate": 0.001492063492063492,
      "loss": 0.0241,
      "step": 48
    },
    {
      "epoch": 20.8,
      "learning_rate": 0.0014285714285714286,
      "loss": 0.018,
      "step": 52
    },
    {
      "epoch": 22.4,
      "learning_rate": 0.0013650793650793651,
      "loss": 0.0187,
      "step": 56
    },
    {
      "epoch": 24.0,
      "learning_rate": 0.0013015873015873017,
      "loss": 0.0159,
      "step": 60
    },
    {
      "epoch": 25.6,
      "learning_rate": 0.0012698412698412698,
      "loss": 0.0463,
      "step": 64
    },
    {
      "epoch": 27.2,
      "learning_rate": 0.0012063492063492064,
      "loss": 0.0279,
      "step": 68
    },
    {
      "epoch": 28.8,
      "learning_rate": 0.0011428571428571427,
      "loss": 0.0133,
      "step": 72
    },
    {
      "epoch": 30.4,
      "learning_rate": 0.0010793650793650793,
      "loss": 0.0148,
      "step": 76
    },
    {
      "epoch": 32.0,
      "learning_rate": 0.0010158730158730158,
      "loss": 0.0115,
      "step": 80
    },
    {
      "epoch": 33.6,
      "learning_rate": 0.0009523809523809524,
      "loss": 0.0139,
      "step": 84
    },
    {
      "epoch": 35.2,
      "learning_rate": 0.0008888888888888888,
      "loss": 0.0105,
      "step": 88
    },
    {
      "epoch": 36.8,
      "learning_rate": 0.0008253968253968254,
      "loss": 0.0096,
      "step": 92
    },
    {
      "epoch": 38.4,
      "learning_rate": 0.0007619047619047619,
      "loss": 0.0129,
      "step": 96
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0006984126984126984,
      "loss": 0.0069,
      "step": 100
    },
    {
      "epoch": 41.6,
      "learning_rate": 0.0006349206349206349,
      "loss": 0.0063,
      "step": 104
    },
    {
      "epoch": 43.2,
      "learning_rate": 0.0005714285714285714,
      "loss": 0.0053,
      "step": 108
    },
    {
      "epoch": 44.8,
      "learning_rate": 0.0005079365079365079,
      "loss": 0.0047,
      "step": 112
    },
    {
      "epoch": 46.4,
      "learning_rate": 0.0004444444444444444,
      "loss": 0.0055,
      "step": 116
    },
    {
      "epoch": 48.0,
      "learning_rate": 0.00038095238095238096,
      "loss": 0.0033,
      "step": 120
    },
    {
      "epoch": 49.6,
      "learning_rate": 0.00031746031746031746,
      "loss": 0.0038,
      "step": 124
    },
    {
      "epoch": 51.2,
      "learning_rate": 0.00025396825396825396,
      "loss": 0.0036,
      "step": 128
    },
    {
      "epoch": 52.8,
      "learning_rate": 0.00019047619047619048,
      "loss": 0.0029,
      "step": 132
    },
    {
      "epoch": 54.4,
      "learning_rate": 0.00012698412698412698,
      "loss": 0.0035,
      "step": 136
    },
    {
      "epoch": 56.0,
      "learning_rate": 6.349206349206349e-05,
      "loss": 0.0026,
      "step": 140
    }
  ],
  "logging_steps": 4,
  "max_steps": 140,
  "num_train_epochs": 70,
  "save_steps": 500,
  "total_flos": 4.89773450877993e+16,
  "trial_name": null,
  "trial_params": null
}