trainer_state.json

{
  "best_metric": 1.763041377067566,
  "best_model_checkpoint": "output/loud-luxury/checkpoint-55",
  "epoch": 5.0,
  "global_step": 55,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.45,
      "learning_rate": 7.836279790554737e-05,
      "loss": 2.5226,
      "step": 5
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.778782010045488e-06,
      "loss": 2.2569,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.33217453956604,
      "eval_runtime": 0.7668,
      "eval_samples_per_second": 22.17,
      "eval_steps_per_second": 3.912,
      "step": 11
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.0102530108070535e-05,
      "loss": 2.051,
      "step": 15
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.000126309992352219,
      "loss": 2.1034,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.14093017578125,
      "eval_runtime": 0.7701,
      "eval_samples_per_second": 22.075,
      "eval_steps_per_second": 3.896,
      "step": 22
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.00011352344634864655,
      "loss": 1.7721,
      "step": 25
    },
    {
      "epoch": 2.73,
      "learning_rate": 2.3676553651353525e-05,
      "loss": 1.7551,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.9571082592010498,
      "eval_runtime": 0.7658,
      "eval_samples_per_second": 22.199,
      "eval_steps_per_second": 3.917,
      "step": 33
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.0890007647780947e-05,
      "loss": 1.5651,
      "step": 35
    },
    {
      "epoch": 3.64,
      "learning_rate": 9.709746989192932e-05,
      "loss": 1.4126,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.8464750051498413,
      "eval_runtime": 0.7656,
      "eval_samples_per_second": 22.204,
      "eval_steps_per_second": 3.918,
      "step": 44
    },
    {
      "epoch": 4.09,
      "learning_rate": 0.00013442121798995453,
      "loss": 1.4854,
      "step": 45
    },
    {
      "epoch": 4.55,
      "learning_rate": 5.883720209445262e-05,
      "loss": 1.0912,
      "step": 50
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 1.4044,
      "step": 55
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.763041377067566,
      "eval_runtime": 0.7602,
      "eval_samples_per_second": 22.361,
      "eval_steps_per_second": 3.946,
      "step": 55
    }
  ],
  "max_steps": 55,
  "num_train_epochs": 5,
  "total_flos": 56177786880000.0,
  "trial_name": null,
  "trial_params": null
}