{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.99581589958159,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -2.323209762573242,
      "logits/rejected": -2.2975621223449707,
      "logps/chosen": -281.906982421875,
      "logps/rejected": -356.19207763671875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.4348256587982178,
      "logits/rejected": -2.431809663772583,
      "logps/chosen": -233.62814331054688,
      "logps/rejected": -294.4069519042969,
      "loss": 0.69,
      "rewards/accuracies": 0.4652777910232544,
      "rewards/chosen": -0.004595045931637287,
      "rewards/margins": 0.009994587861001492,
      "rewards/rejected": -0.014589633792638779,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.3493409156799316,
      "logits/rejected": -2.289567470550537,
      "logps/chosen": -273.939208984375,
      "logps/rejected": -290.79986572265625,
      "loss": 0.6886,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.3175145089626312,
      "rewards/margins": 0.04773712903261185,
      "rewards/rejected": -0.36525166034698486,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.3267781734466553,
      "logits/rejected": -2.322356700897217,
      "logps/chosen": -268.1922302246094,
      "logps/rejected": -330.12933349609375,
      "loss": 0.6712,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.252364844083786,
      "rewards/margins": 0.18057797849178314,
      "rewards/rejected": -0.43294280767440796,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.3110787868499756,
      "logits/rejected": -2.285353899002075,
      "logps/chosen": -280.7604675292969,
      "logps/rejected": -335.8335876464844,
      "loss": 0.6648,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.24384799599647522,
      "rewards/margins": 0.17965658009052277,
      "rewards/rejected": -0.4235045909881592,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.2896323204040527,
      "logits/rejected": -2.26826548576355,
      "logps/chosen": -313.2692565917969,
      "logps/rejected": -327.1571044921875,
      "loss": 0.6986,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.4340161681175232,
      "rewards/margins": 0.04799577593803406,
      "rewards/rejected": -0.48201197385787964,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.33687162399292,
      "logits/rejected": -2.3146133422851562,
      "logps/chosen": -273.87420654296875,
      "logps/rejected": -346.2785949707031,
      "loss": 0.6462,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.31471648812294006,
      "rewards/margins": 0.17776905000209808,
      "rewards/rejected": -0.49248558282852173,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.362096071243286,
      "logits/rejected": -2.312239170074463,
      "logps/chosen": -280.3855285644531,
      "logps/rejected": -351.7784729003906,
      "loss": 0.6541,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.24225488305091858,
      "rewards/margins": 0.20938357710838318,
      "rewards/rejected": -0.45163851976394653,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.352534770965576,
      "logits/rejected": -2.33256196975708,
      "logps/chosen": -302.2196044921875,
      "logps/rejected": -284.39703369140625,
      "loss": 0.6473,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.3088821768760681,
      "rewards/margins": 0.1018916592001915,
      "rewards/rejected": -0.4107738435268402,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.354665517807007,
      "logits/rejected": -2.3205881118774414,
      "logps/chosen": -290.73431396484375,
      "logps/rejected": -303.2606506347656,
      "loss": 0.6515,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.3215685188770294,
      "rewards/margins": 0.10454748570919037,
      "rewards/rejected": -0.4261159896850586,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.354948043823242,
      "logits/rejected": -2.326176166534424,
      "logps/chosen": -291.93096923828125,
      "logps/rejected": -360.5430908203125,
      "loss": 0.6545,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.3707358241081238,
      "rewards/margins": 0.19102896749973297,
      "rewards/rejected": -0.5617648363113403,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -2.356266498565674,
      "logits/rejected": -2.33392596244812,
      "logps/chosen": -285.56732177734375,
      "logps/rejected": -340.04534912109375,
      "loss": 0.6384,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.3159911036491394,
      "rewards/margins": 0.20389732718467712,
      "rewards/rejected": -0.5198885202407837,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.663738803703244,
      "train_runtime": 1999.9245,
      "train_samples_per_second": 7.642,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 119,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}