{
  "best_metric": 0.47333332896232605,
  "best_model_checkpoint": "./mixstral/05-04-24-Weni-WeniGPT-Agents-Mixstral-Instruct-3.0.1-KTO_KTO with Agents 1.2.0 dataset and Mixstral model with tokenization zephyr chat template-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": 0.10322337597608566,
      "kl": 0.26181185245513916,
      "learning_rate": 0.00018142857142857142,
      "logps/chosen": -314.7935485839844,
      "logps/rejected": -306.157958984375,
      "loss": 0.458,
      "rewards/chosen": -2.649841785430908,
      "rewards/margins": 0.5847967267036438,
      "rewards/rejected": -3.3113763332366943,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.0019746168982237577,
      "kl": 0.0,
      "learning_rate": 0.00015285714285714287,
      "logps/chosen": -632.806396484375,
      "logps/rejected": -873.2008056640625,
      "loss": 0.4414,
      "rewards/chosen": -38.71037292480469,
      "rewards/margins": 23.247425079345703,
      "rewards/rejected": -59.540672302246094,
      "step": 40
    },
    {
      "epoch": 0.34,
      "eval_kl": 0.0,
      "eval_logps/chosen": -1706.5499267578125,
      "eval_logps/rejected": -1665.1412353515625,
      "eval_loss": 0.47333332896232605,
      "eval_rewards/chosen": -144.7169647216797,
      "eval_rewards/margins": -3.485215187072754,
      "eval_rewards/rejected": -141.496337890625,
      "eval_runtime": 336.8962,
      "eval_samples_per_second": 0.89,
      "eval_steps_per_second": 0.223,
      "step": 50
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.3855821831841872e-11,
      "kl": 0.0,
      "learning_rate": 0.00012428571428571428,
      "logps/chosen": -1769.3804931640625,
      "logps/rejected": -1863.0657958984375,
      "loss": 0.5219,
      "rewards/chosen": -151.40240478515625,
      "rewards/margins": 10.106590270996094,
      "rewards/rejected": -157.65440368652344,
      "step": 60
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 9.571428571428573e-05,
      "logps/chosen": -2173.3388671875,
      "logps/rejected": -2083.44091796875,
      "loss": 0.4875,
      "rewards/chosen": -190.8201446533203,
      "rewards/margins": -6.008725166320801,
      "rewards/rejected": -181.0947723388672,
      "step": 80
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 6.714285714285714e-05,
      "logps/chosen": -2457.9619140625,
      "logps/rejected": -1933.36376953125,
      "loss": 0.45,
      "rewards/chosen": -216.5624542236328,
      "rewards/margins": -47.97616958618164,
      "rewards/rejected": -167.77308654785156,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_kl": 0.0,
      "eval_logps/chosen": -1931.6995849609375,
      "eval_logps/rejected": -1860.81591796875,
      "eval_loss": 0.47333332896232605,
      "eval_rewards/chosen": -167.23193359375,
      "eval_rewards/margins": -6.333158493041992,
      "eval_rewards/rejected": -161.0637969970703,
      "eval_runtime": 336.6145,
      "eval_samples_per_second": 0.891,
      "eval_steps_per_second": 0.223,
      "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
}