checkpoint-24/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.428571428571429,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 4.9999999999999996e-05,
      "loss": 1.0283,
      "step": 1
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.999999999999999e-05,
      "loss": 1.0447,
      "step": 2
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00015,
      "loss": 1.0075,
      "step": 3
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00019999999999999998,
      "loss": 0.9604,
      "step": 4
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00025,
      "loss": 0.8964,
      "step": 5
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0003,
      "loss": 0.8099,
      "step": 6
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.000294,
      "loss": 0.7356,
      "step": 7
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.00028799999999999995,
      "loss": 0.6955,
      "step": 8
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.00028199999999999997,
      "loss": 0.6318,
      "step": 9
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.000276,
      "loss": 0.594,
      "step": 10
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.00027,
      "loss": 0.5653,
      "step": 11
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.00026399999999999997,
      "loss": 0.5067,
      "step": 12
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.000258,
      "loss": 0.4433,
      "step": 13
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.00025199999999999995,
      "loss": 0.3947,
      "step": 14
    },
    {
      "epoch": 4.43,
      "learning_rate": 0.00024599999999999996,
      "loss": 0.3413,
      "step": 15
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.2837,
      "step": 16
    },
    {
      "epoch": 5.29,
      "learning_rate": 0.000234,
      "loss": 0.2606,
      "step": 17
    },
    {
      "epoch": 5.43,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.2437,
      "step": 18
    },
    {
      "epoch": 6.14,
      "learning_rate": 0.00022199999999999998,
      "loss": 0.2033,
      "step": 19
    },
    {
      "epoch": 6.29,
      "learning_rate": 0.00021599999999999996,
      "loss": 0.187,
      "step": 20
    },
    {
      "epoch": 6.43,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.162,
      "step": 21
    },
    {
      "epoch": 7.14,
      "learning_rate": 0.000204,
      "loss": 0.1494,
      "step": 22
    },
    {
      "epoch": 7.29,
      "learning_rate": 0.000198,
      "loss": 0.1454,
      "step": 23
    },
    {
      "epoch": 7.43,
      "learning_rate": 0.00019199999999999998,
      "loss": 0.1422,
      "step": 24
    }
  ],
  "logging_steps": 1,
  "max_steps": 56,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 3984531891683328.0,
  "trial_name": null,
  "trial_params": null
}