calibrate_margin_ibn_dd/checkpoint-10000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008527584604298755,
  "eval_steps": 2500,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 4.0477,
      "step": 500
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 3.362,
      "step": 1000
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 3.1288,
      "step": 1500
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.9448,
      "step": 2000
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.7412,
      "step": 2500
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.6923,
      "step": 3000
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.626,
      "step": 3500
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.573,
      "step": 4000
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.5107,
      "step": 4500
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.4911,
      "step": 5000
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.01,
      "loss": 2.4486,
      "step": 5500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.4078,
      "step": 6000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.4018,
      "step": 6500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.3685,
      "step": 7000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.3276,
      "step": 7500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.2816,
      "step": 8000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.2986,
      "step": 8500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.2612,
      "step": 9000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.2599,
      "step": 9500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.01,
      "loss": 2.236,
      "step": 10000
    }
  ],
  "logging_steps": 500,
  "max_steps": 20000,
  "num_train_epochs": 1,
  "save_steps": 10000,
  "total_flos": 4.829257277256499e+16,
  "trial_name": null,
  "trial_params": null
}