{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9992429977289932,
  "eval_steps": 100,
  "global_step": 165,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.9411764705882356e-07,
      "logits/chosen": -2.4206371307373047,
      "logits/rejected": -2.4444966316223145,
      "logps/chosen": -285.239013671875,
      "logps/rejected": -269.48089599609375,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9411764705882355e-06,
      "logits/chosen": -2.4849026203155518,
      "logits/rejected": -2.361117362976074,
      "logps/chosen": -271.2438659667969,
      "logps/rejected": -257.04443359375,
      "loss": 0.689,
      "rewards/accuracies": 0.5208333134651184,
      "rewards/chosen": -0.010696162469685078,
      "rewards/margins": 0.008694707415997982,
      "rewards/rejected": -0.01939087174832821,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.994932636402032e-06,
      "logits/chosen": -2.162980318069458,
      "logits/rejected": -2.0703816413879395,
      "logps/chosen": -297.11322021484375,
      "logps/rejected": -317.59942626953125,
      "loss": 0.6434,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.3237840533256531,
      "rewards/margins": 0.11028959602117538,
      "rewards/rejected": -0.43407368659973145,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.905416503522124e-06,
      "logits/chosen": -1.9069904088974,
      "logits/rejected": -1.5060116052627563,
      "logps/chosen": -316.3459777832031,
      "logps/rejected": -310.5841064453125,
      "loss": 0.594,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.2567785978317261,
      "rewards/margins": 0.2907828688621521,
      "rewards/rejected": -0.5475614666938782,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.707922373336524e-06,
      "logits/chosen": -1.2782038450241089,
      "logits/rejected": -1.1451029777526855,
      "logps/chosen": -334.9066467285156,
      "logps/rejected": -359.5328674316406,
      "loss": 0.6188,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.6329762935638428,
      "rewards/margins": 0.26158708333969116,
      "rewards/rejected": -0.8945633769035339,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.411315662967732e-06,
      "logits/chosen": -1.3727542161941528,
      "logits/rejected": -1.0713155269622803,
      "logps/chosen": -302.04876708984375,
      "logps/rejected": -323.1622009277344,
      "loss": 0.5896,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.33613428473472595,
      "rewards/margins": 0.32212409377098083,
      "rewards/rejected": -0.658258318901062,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.028910905897229e-06,
      "logits/chosen": -1.2992875576019287,
      "logits/rejected": -0.787635862827301,
      "logps/chosen": -339.11273193359375,
      "logps/rejected": -345.2218017578125,
      "loss": 0.5665,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.41970473527908325,
      "rewards/margins": 0.42981767654418945,
      "rewards/rejected": -0.8495222926139832,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.577874068920446e-06,
      "logits/chosen": -1.1777417659759521,
      "logits/rejected": -0.7731892466545105,
      "logps/chosen": -331.8875427246094,
      "logps/rejected": -348.7061767578125,
      "loss": 0.5574,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.45568642020225525,
      "rewards/margins": 0.4189928472042084,
      "rewards/rejected": -0.8746792674064636,
      "step": 70
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.0784519801008546e-06,
      "logits/chosen": -1.0813881158828735,
      "logits/rejected": -0.695389986038208,
      "logps/chosen": -328.33599853515625,
      "logps/rejected": -378.7125549316406,
      "loss": 0.5757,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.41106706857681274,
      "rewards/margins": 0.49909764528274536,
      "rewards/rejected": -0.9101647138595581,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.553063458334059e-06,
      "logits/chosen": -1.1919687986373901,
      "logits/rejected": -0.7594654560089111,
      "logps/chosen": -311.298095703125,
      "logps/rejected": -356.6019592285156,
      "loss": 0.5572,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.4122340679168701,
      "rewards/margins": 0.47736555337905884,
      "rewards/rejected": -0.889599621295929,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.025292943281429e-06,
      "logits/chosen": -1.1734464168548584,
      "logits/rejected": -0.8028995394706726,
      "logps/chosen": -314.7601013183594,
      "logps/rejected": -354.06524658203125,
      "loss": 0.5669,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.37006908655166626,
      "rewards/margins": 0.4851018786430359,
      "rewards/rejected": -0.8551709055900574,
      "step": 100
    },
    {
      "epoch": 0.61,
      "eval_logits/chosen": -1.1060620546340942,
      "eval_logits/rejected": -0.6833169460296631,
      "eval_logps/chosen": -329.36163330078125,
      "eval_logps/rejected": -354.41693115234375,
      "eval_loss": 0.5602970123291016,
      "eval_rewards/accuracies": 0.6980000138282776,
      "eval_rewards/chosen": -0.4443123936653137,
      "eval_rewards/margins": 0.428010493516922,
      "eval_rewards/rejected": -0.8723229169845581,
      "eval_runtime": 384.7387,
      "eval_samples_per_second": 5.198,
      "eval_steps_per_second": 0.65,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5188318011445907e-06,
      "logits/chosen": -1.08719801902771,
      "logits/rejected": -0.7771456241607666,
      "logps/chosen": -318.55706787109375,
      "logps/rejected": -360.88641357421875,
      "loss": 0.5582,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.4783196449279785,
      "rewards/margins": 0.4534730315208435,
      "rewards/rejected": -0.9317927360534668,
      "step": 110
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0564148305586296e-06,
      "logits/chosen": -1.264317274093628,
      "logits/rejected": -0.6377102136611938,
      "logps/chosen": -338.55364990234375,
      "logps/rejected": -344.6549072265625,
      "loss": 0.552,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.4587250351905823,
      "rewards/margins": 0.43041476607322693,
      "rewards/rejected": -0.8891397714614868,
      "step": 120
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.587997083462197e-07,
      "logits/chosen": -1.205304503440857,
      "logits/rejected": -0.7666274905204773,
      "logps/chosen": -337.1952209472656,
      "logps/rejected": -378.85015869140625,
      "loss": 0.5546,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.481029212474823,
      "rewards/margins": 0.47466927766799927,
      "rewards/rejected": -0.9556983709335327,
      "step": 130
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.438351873250492e-07,
      "logits/chosen": -1.018686056137085,
      "logits/rejected": -0.5739496946334839,
      "logps/chosen": -331.6724853515625,
      "logps/rejected": -385.0194396972656,
      "loss": 0.5522,
      "rewards/accuracies": 0.7593749761581421,
      "rewards/chosen": -0.5046229958534241,
      "rewards/margins": 0.5355193018913269,
      "rewards/rejected": -1.040142297744751,
      "step": 140
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.2565987432367032e-07,
      "logits/chosen": -1.0615447759628296,
      "logits/rejected": -0.6152147650718689,
      "logps/chosen": -330.9660949707031,
      "logps/rejected": -370.95867919921875,
      "loss": 0.5513,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.4836733937263489,
      "rewards/margins": 0.5048274397850037,
      "rewards/rejected": -0.9885008931159973,
      "step": 150
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4067554877743861e-08,
      "logits/chosen": -1.0293254852294922,
      "logits/rejected": -0.6487714052200317,
      "logps/chosen": -317.6270751953125,
      "logps/rejected": -361.36053466796875,
      "loss": 0.5338,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.4425296187400818,
      "rewards/margins": 0.528509259223938,
      "rewards/rejected": -0.9710389375686646,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 165,
      "total_flos": 0.0,
      "train_loss": 0.5778744011214285,
      "train_runtime": 7573.0657,
      "train_samples_per_second": 2.791,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 165,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}