trainer_state.json

{
  "best_metric": 3.056016445159912,
  "best_model_checkpoint": "output/bill-wurtz/checkpoint-54",
  "epoch": 2.0,
  "global_step": 54,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 0.00012591446386292745,
      "loss": 3.5789,
      "step": 5
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.577107195028616e-05,
      "loss": 3.3728,
      "step": 10
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.668773501204858e-05,
      "loss": 3.2429,
      "step": 15
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.152382364220488e-05,
      "loss": 3.2491,
      "step": 20
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.849121878224087e-06,
      "loss": 3.1415,
      "step": 25
    },
    {
      "epoch": 1.0,
      "eval_loss": 3.1433706283569336,
      "eval_runtime": 0.6351,
      "eval_samples_per_second": 42.515,
      "eval_steps_per_second": 6.299,
      "step": 27
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.137086214086682e-06,
      "loss": 2.9586,
      "step": 30
    },
    {
      "epoch": 1.3,
      "learning_rate": 2.7634920609188867e-05,
      "loss": 3.0624,
      "step": 35
    },
    {
      "epoch": 1.48,
      "learning_rate": 6.461126473674133e-05,
      "loss": 2.9663,
      "step": 40
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00010290000000000001,
      "loss": 2.8978,
      "step": 45
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001299031991261861,
      "loss": 2.9441,
      "step": 50
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.056016445159912,
      "eval_runtime": 0.6316,
      "eval_samples_per_second": 42.746,
      "eval_steps_per_second": 6.333,
      "step": 54
    }
  ],
  "max_steps": 135,
  "num_train_epochs": 5,
  "total_flos": 56439078912000.0,
  "trial_name": null,
  "trial_params": null
}