run-16/checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.1446540880503147,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6741935483870968,
      "eval_loss": 0.28739064931869507,
      "eval_runtime": 5.7413,
      "eval_samples_per_second": 539.95,
      "eval_steps_per_second": 11.322,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.4601,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8464516129032258,
      "eval_loss": 0.11687513440847397,
      "eval_runtime": 5.8927,
      "eval_samples_per_second": 526.074,
      "eval_steps_per_second": 11.031,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8925806451612903,
      "eval_loss": 0.06558599323034286,
      "eval_runtime": 5.7291,
      "eval_samples_per_second": 541.094,
      "eval_steps_per_second": 11.346,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1366,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 461297311410264.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8775428700427587,
    "num_train_epochs": 9,
    "temperature": 3
  }
}