{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9951497860199715,
  "eval_steps": 100,
  "global_step": 109,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.7272727272727274e-07,
      "logits/chosen": -2.779836893081665,
      "logits/rejected": -2.772892951965332,
      "logps/chosen": -67.39044952392578,
      "logps/rejected": -65.7892074584961,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.09,
      "learning_rate": 2.7272727272727272e-06,
      "logits/chosen": -2.764448881149292,
      "logits/rejected": -2.7583670616149902,
      "logps/chosen": -63.044189453125,
      "logps/rejected": -65.00924682617188,
      "loss": 0.6834,
      "rewards/accuracies": 0.6076388955116272,
      "rewards/chosen": 0.011253755539655685,
      "rewards/margins": 0.020065149292349815,
      "rewards/rejected": -0.008811394684016705,
      "step": 10
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.7244897959183674e-06,
      "logits/chosen": -2.7638134956359863,
      "logits/rejected": -2.7512309551239014,
      "logps/chosen": -60.88775634765625,
      "logps/rejected": -64.6989517211914,
      "loss": 0.5807,
      "rewards/accuracies": 0.8656250238418579,
      "rewards/chosen": 0.074214868247509,
      "rewards/margins": 0.26364782452583313,
      "rewards/rejected": -0.18943293392658234,
      "step": 20
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.4183673469387754e-06,
      "logits/chosen": -2.7466235160827637,
      "logits/rejected": -2.7364418506622314,
      "logps/chosen": -62.1016845703125,
      "logps/rejected": -73.9842529296875,
      "loss": 0.4425,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": 0.031027918681502342,
      "rewards/margins": 0.7247552275657654,
      "rewards/rejected": -0.6937273740768433,
      "step": 30
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.112244897959184e-06,
      "logits/chosen": -2.720366954803467,
      "logits/rejected": -2.707810401916504,
      "logps/chosen": -64.14788818359375,
      "logps/rejected": -80.15257263183594,
      "loss": 0.3545,
      "rewards/accuracies": 0.909375011920929,
      "rewards/chosen": -0.217222660779953,
      "rewards/margins": 1.13102388381958,
      "rewards/rejected": -1.348246693611145,
      "step": 40
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.806122448979592e-06,
      "logits/chosen": -2.703882932662964,
      "logits/rejected": -2.688373327255249,
      "logps/chosen": -68.16423034667969,
      "logps/rejected": -81.43087005615234,
      "loss": 0.3228,
      "rewards/accuracies": 0.903124988079071,
      "rewards/chosen": -0.3540056347846985,
      "rewards/margins": 1.3236706256866455,
      "rewards/rejected": -1.6776764392852783,
      "step": 50
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.5e-06,
      "logits/chosen": -2.6891207695007324,
      "logits/rejected": -2.6766979694366455,
      "logps/chosen": -65.70265197753906,
      "logps/rejected": -84.64549255371094,
      "loss": 0.3174,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -0.44440922141075134,
      "rewards/margins": 1.486973762512207,
      "rewards/rejected": -1.9313831329345703,
      "step": 60
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.193877551020408e-06,
      "logits/chosen": -2.696516275405884,
      "logits/rejected": -2.6814627647399902,
      "logps/chosen": -68.18341827392578,
      "logps/rejected": -88.01350402832031,
      "loss": 0.2817,
      "rewards/accuracies": 0.903124988079071,
      "rewards/chosen": -0.6013726592063904,
      "rewards/margins": 1.7021477222442627,
      "rewards/rejected": -2.303520441055298,
      "step": 70
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.877551020408164e-07,
      "logits/chosen": -2.680356025695801,
      "logits/rejected": -2.662485122680664,
      "logps/chosen": -69.7511978149414,
      "logps/rejected": -89.8312759399414,
      "loss": 0.2427,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -0.5794776678085327,
      "rewards/margins": 1.9294792413711548,
      "rewards/rejected": -2.5089569091796875,
      "step": 80
    },
    {
      "epoch": 0.82,
      "learning_rate": 5.816326530612245e-07,
      "logits/chosen": -2.6861064434051514,
      "logits/rejected": -2.6741652488708496,
      "logps/chosen": -71.44612884521484,
      "logps/rejected": -92.25032806396484,
      "loss": 0.257,
      "rewards/accuracies": 0.8968750238418579,
      "rewards/chosen": -0.6901184320449829,
      "rewards/margins": 1.9418220520019531,
      "rewards/rejected": -2.6319406032562256,
      "step": 90
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.7551020408163265e-07,
      "logits/chosen": -2.6638271808624268,
      "logits/rejected": -2.6519863605499268,
      "logps/chosen": -72.42386627197266,
      "logps/rejected": -90.99058532714844,
      "loss": 0.2434,
      "rewards/accuracies": 0.9281250238418579,
      "rewards/chosen": -0.7177737355232239,
      "rewards/margins": 2.0420150756835938,
      "rewards/rejected": -2.759788751602173,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -2.6727941036224365,
      "eval_logits/rejected": -2.6590042114257812,
      "eval_logps/chosen": -70.43208312988281,
      "eval_logps/rejected": -92.74579620361328,
      "eval_loss": 0.24523001909255981,
      "eval_rewards/accuracies": 0.913241982460022,
      "eval_rewards/chosen": -0.7311916947364807,
      "eval_rewards/margins": 2.0473380088806152,
      "eval_rewards/rejected": -2.778529405593872,
      "eval_runtime": 288.5183,
      "eval_samples_per_second": 3.036,
      "eval_steps_per_second": 3.036,
      "step": 109
    },
    {
      "epoch": 1.0,
      "step": 109,
      "total_flos": 0.0,
      "train_loss": 0.3616804119643815,
      "train_runtime": 2568.179,
      "train_samples_per_second": 1.365,
      "train_steps_per_second": 0.042
    }
  ],
  "logging_steps": 10,
  "max_steps": 109,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}