{
  "best_metric": 0.6390000581741333,
  "best_model_checkpoint": "./zephyr/09-04-24-Weni-WeniGPT-Agents-Zephyr-1.0.8-KTO_Hyperparameter search, altering desired and undesired weights for KTO task.-2_max_steps-145_batch_16_2024-04-09_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": 0.0,
      "kl": 7.392268180847168,
      "learning_rate": 0.00018142857142857142,
      "logps/chosen": -556.1337890625,
      "logps/rejected": -563.4813842773438,
      "loss": 0.5396,
      "rewards/chosen": -27.30255699157715,
      "rewards/margins": 7.028589725494385,
      "rewards/rejected": -27.570302963256836,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 0.00015285714285714287,
      "logps/chosen": -1588.400634765625,
      "logps/rejected": -1780.9019775390625,
      "loss": 0.6413,
      "rewards/chosen": -131.71377563476562,
      "rewards/margins": 18.5042781829834,
      "rewards/rejected": -147.60687255859375,
      "step": 40
    },
    {
      "epoch": 0.34,
      "eval_kl": 0.0,
      "eval_logps/chosen": -1972.53662109375,
      "eval_logps/rejected": -1894.372314453125,
      "eval_loss": 0.6390000581741333,
      "eval_rewards/chosen": -168.53868103027344,
      "eval_rewards/margins": -6.742273807525635,
      "eval_rewards/rejected": -162.69952392578125,
      "eval_runtime": 141.4233,
      "eval_samples_per_second": 2.121,
      "eval_steps_per_second": 0.53,
      "step": 50
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 0.00012428571428571428,
      "logps/chosen": -2004.6998291015625,
      "logps/rejected": -1917.18212890625,
      "loss": 0.6159,
      "rewards/chosen": -172.33108520507812,
      "rewards/margins": -6.893044948577881,
      "rewards/rejected": -164.2748260498047,
      "step": 60
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 9.571428571428573e-05,
      "logps/chosen": -2267.27490234375,
      "logps/rejected": -2191.0869140625,
      "loss": 0.6539,
      "rewards/chosen": -196.35650634765625,
      "rewards/margins": -5.303529739379883,
      "rewards/rejected": -189.41534423828125,
      "step": 80
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 6.714285714285714e-05,
      "logps/chosen": -2088.027587890625,
      "logps/rejected": -2137.646728515625,
      "loss": 0.637,
      "rewards/chosen": -180.6759796142578,
      "rewards/margins": 5.185708045959473,
      "rewards/rejected": -185.26431274414062,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_kl": 0.0,
      "eval_logps/chosen": -2049.0234375,
      "eval_logps/rejected": -1959.040283203125,
      "eval_loss": 0.6390000581741333,
      "eval_rewards/chosen": -176.1873779296875,
      "eval_rewards/margins": -7.898090839385986,
      "eval_rewards/rejected": -169.1663055419922,
      "eval_runtime": 141.3411,
      "eval_samples_per_second": 2.123,
      "eval_steps_per_second": 0.531,
      "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
}