run-4/checkpoint-1500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.716981132075472,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7190322580645161,
      "eval_loss": 3.268949508666992,
      "eval_runtime": 2.384,
      "eval_samples_per_second": 1300.335,
      "eval_steps_per_second": 27.265,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.371069182389937e-05,
      "loss": 3.7735,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8429032258064516,
      "eval_loss": 1.8548803329467773,
      "eval_runtime": 2.405,
      "eval_samples_per_second": 1288.981,
      "eval_steps_per_second": 27.027,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8925806451612903,
      "eval_loss": 1.1410753726959229,
      "eval_runtime": 2.417,
      "eval_samples_per_second": 1282.579,
      "eval_steps_per_second": 26.893,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 7.421383647798742e-06,
      "loss": 1.6771,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9135483870967742,
      "eval_loss": 0.8444640040397644,
      "eval_runtime": 2.38,
      "eval_samples_per_second": 1302.521,
      "eval_steps_per_second": 27.311,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.8911,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1590,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 389479376069112.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.682022026080148,
    "num_train_epochs": 5,
    "temperature": 20
  }
}