trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.7586206896551726,
  "eval_steps": 100,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.55,
      "learning_rate": 5e-07,
      "logits/chosen": -2.442253589630127,
      "logits/rejected": -2.4321861267089844,
      "logps/chosen": -136.51828002929688,
      "logps/rejected": -126.99198913574219,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.55,
      "eval_logits/chosen": -2.2761800289154053,
      "eval_logits/rejected": -2.3014957904815674,
      "eval_logps/chosen": -191.0126495361328,
      "eval_logps/rejected": -132.9451446533203,
      "eval_loss": 0.6931473016738892,
      "eval_rewards/accuracies": 0.0,
      "eval_rewards/chosen": 0.0,
      "eval_rewards/margins": 0.0,
      "eval_rewards/rejected": 0.0,
      "eval_runtime": 7.3415,
      "eval_samples_per_second": 4.086,
      "eval_steps_per_second": 0.136,
      "step": 1
    },
    {
      "epoch": 1.66,
      "eval_logits/chosen": -2.2754929065704346,
      "eval_logits/rejected": -2.3016197681427,
      "eval_logps/chosen": -190.82786560058594,
      "eval_logps/rejected": -133.05657958984375,
      "eval_loss": 0.6928330659866333,
      "eval_rewards/accuracies": 0.5,
      "eval_rewards/chosen": 0.018478775396943092,
      "eval_rewards/margins": 0.029622457921504974,
      "eval_rewards/rejected": -0.011143684387207031,
      "eval_runtime": 6.7101,
      "eval_samples_per_second": 4.471,
      "eval_steps_per_second": 0.149,
      "step": 3
    },
    {
      "epoch": 2.76,
      "eval_logits/chosen": -2.274653911590576,
      "eval_logits/rejected": -2.2999308109283447,
      "eval_logps/chosen": -190.57220458984375,
      "eval_logps/rejected": -132.8740234375,
      "eval_loss": 0.68434077501297,
      "eval_rewards/accuracies": 0.5,
      "eval_rewards/chosen": 0.04404526203870773,
      "eval_rewards/margins": 0.03693275526165962,
      "eval_rewards/rejected": 0.0071125030517578125,
      "eval_runtime": 6.6968,
      "eval_samples_per_second": 4.48,
      "eval_steps_per_second": 0.149,
      "step": 5
    },
    {
      "epoch": 2.76,
      "step": 5,
      "total_flos": 0.0,
      "train_loss": 0.6927925229072571,
      "train_runtime": 943.7212,
      "train_samples_per_second": 4.885,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 10,
  "max_steps": 5,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}