{
  "best_metric": 0.47333332896232605,
  "best_model_checkpoint": "./zephyr/05-04-24-Weni-WeniGPT-Agents-Zephyr-1.0.1-KTO_testing kto dataset during training-3_max_steps-145_batch_16_2024-04-05_ppid_9/checkpoint-100",
  "epoch": 0.684931506849315,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "grad_norm": 4.650723457336426,
      "kl": 4.865365505218506,
      "learning_rate": 0.00018285714285714286,
      "logps/chosen": -310.4614562988281,
      "logps/rejected": -364.2945556640625,
      "loss": 0.45,
      "rewards/chosen": -2.890953779220581,
      "rewards/margins": 3.259232521057129,
      "rewards/rejected": -5.586672306060791,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.0,
      "kl": 1.256190538406372,
      "learning_rate": 0.0001542857142857143,
      "logps/chosen": -1522.9342041015625,
      "logps/rejected": -1577.1060791015625,
      "loss": 0.4463,
      "rewards/chosen": -123.79159545898438,
      "rewards/margins": 4.876855373382568,
      "rewards/rejected": -128.13319396972656,
      "step": 40
    },
    {
      "epoch": 0.34,
      "eval_kl": 0.0,
      "eval_logps/chosen": -1422.4874267578125,
      "eval_logps/rejected": -1508.553955078125,
      "eval_loss": 0.473336786031723,
      "eval_rewards/chosen": -114.58648681640625,
      "eval_rewards/margins": 10.318593978881836,
      "eval_rewards/rejected": -124.07293701171875,
      "eval_runtime": 140.0926,
      "eval_samples_per_second": 2.141,
      "eval_steps_per_second": 0.535,
      "step": 50
    },
    {
      "epoch": 0.41,
      "grad_norm": 3.258587184973294e-06,
      "kl": 0.0,
      "learning_rate": 0.00012571428571428572,
      "logps/chosen": -1605.4840087890625,
      "logps/rejected": -1731.5294189453125,
      "loss": 0.4812,
      "rewards/chosen": -133.641357421875,
      "rewards/margins": 11.06646728515625,
      "rewards/rejected": -144.48353576660156,
      "step": 60
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 9.714285714285715e-05,
      "logps/chosen": -2286.116455078125,
      "logps/rejected": -2312.602294921875,
      "loss": 0.4688,
      "rewards/chosen": -200.4071044921875,
      "rewards/margins": 2.4775140285491943,
      "rewards/rejected": -200.7866973876953,
      "step": 80
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 6.857142857142858e-05,
      "logps/chosen": -2744.5634765625,
      "logps/rejected": -2576.698974609375,
      "loss": 0.4719,
      "rewards/chosen": -246.3837890625,
      "rewards/margins": -13.795512199401855,
      "rewards/rejected": -228.7157440185547,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_kl": 0.0,
      "eval_logps/chosen": -2469.165771484375,
      "eval_logps/rejected": -2225.683349609375,
      "eval_loss": 0.47333332896232605,
      "eval_rewards/chosen": -219.25433349609375,
      "eval_rewards/margins": -26.235870361328125,
      "eval_rewards/rejected": -195.78585815429688,
      "eval_runtime": 139.9824,
      "eval_samples_per_second": 2.143,
      "eval_steps_per_second": 0.536,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 145,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}