trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.998118178396688,
  "eval_steps": 500,
  "global_step": 6640,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38,
      "learning_rate": 4.9800000000000004e-05,
      "loss": 0.0423,
      "step": 500
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.919280091260669e-05,
      "loss": 0.0599,
      "step": 1000
    },
    {
      "epoch": 1.13,
      "learning_rate": 4.681083744859544e-05,
      "loss": 0.0611,
      "step": 1500
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.30091452205688e-05,
      "loss": 0.0538,
      "step": 2000
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.803518706991396e-05,
      "loss": 0.0524,
      "step": 2500
    },
    {
      "epoch": 2.26,
      "learning_rate": 3.221273192453827e-05,
      "loss": 0.0414,
      "step": 3000
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5920779768580868e-05,
      "loss": 0.0344,
      "step": 3500
    },
    {
      "epoch": 3.01,
      "learning_rate": 1.9568891467307694e-05,
      "loss": 0.03,
      "step": 4000
    },
    {
      "epoch": 3.39,
      "learning_rate": 1.3593288155812168e-05,
      "loss": 0.0209,
      "step": 4500
    },
    {
      "epoch": 3.76,
      "learning_rate": 8.335204956563933e-06,
      "loss": 0.0182,
      "step": 5000
    },
    {
      "epoch": 4.14,
      "learning_rate": 4.161880145220276e-06,
      "loss": 0.0133,
      "step": 5500
    },
    {
      "epoch": 4.52,
      "learning_rate": 1.344967173429157e-06,
      "loss": 0.0113,
      "step": 6000
    },
    {
      "epoch": 4.89,
      "learning_rate": 6.877170722064397e-08,
      "loss": 0.0086,
      "step": 6500
    }
  ],
  "logging_steps": 500,
  "max_steps": 6640,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5.628910677609677e+16,
  "trial_name": null,
  "trial_params": null
}