{
  "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": -2.4571897983551025,
      "logits/rejected": -2.364729166030884,
      "logps/chosen": -393.0217590332031,
      "logps/rejected": -392.042236328125,
      "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": -2.4598512649536133,
      "logits/rejected": -2.5237348079681396,
      "logps/chosen": -319.2628479003906,
      "logps/rejected": -363.0235290527344,
      "loss": 0.6922,
      "rewards/accuracies": 0.4513888955116272,
      "rewards/chosen": -0.005169111769646406,
      "rewards/margins": 0.0021531255915760994,
      "rewards/rejected": -0.007322237826883793,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.994932636402032e-06,
      "logits/chosen": -2.4682602882385254,
      "logits/rejected": -2.506861925125122,
      "logps/chosen": -310.16632080078125,
      "logps/rejected": -378.22735595703125,
      "loss": 0.6703,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 0.03243725374341011,
      "rewards/margins": 0.057667654007673264,
      "rewards/rejected": -0.02523040398955345,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.905416503522124e-06,
      "logits/chosen": -2.458878993988037,
      "logits/rejected": -2.5221245288848877,
      "logps/chosen": -329.98956298828125,
      "logps/rejected": -392.0107421875,
      "loss": 0.6549,
      "rewards/accuracies": 0.6468750238418579,
      "rewards/chosen": -0.022809982299804688,
      "rewards/margins": 0.09913644939661026,
      "rewards/rejected": -0.12194643169641495,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.707922373336524e-06,
      "logits/chosen": -2.3397133350372314,
      "logits/rejected": -2.440927505493164,
      "logps/chosen": -318.717041015625,
      "logps/rejected": -400.6865539550781,
      "loss": 0.6365,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.11206980794668198,
      "rewards/margins": 0.15860508382320404,
      "rewards/rejected": -0.2706748843193054,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.411315662967732e-06,
      "logits/chosen": -2.3495843410491943,
      "logits/rejected": -2.3737149238586426,
      "logps/chosen": -344.6326599121094,
      "logps/rejected": -397.68157958984375,
      "loss": 0.627,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.17710785567760468,
      "rewards/margins": 0.15342721343040466,
      "rewards/rejected": -0.33053505420684814,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.028910905897229e-06,
      "logits/chosen": -2.30446720123291,
      "logits/rejected": -2.3595499992370605,
      "logps/chosen": -353.98602294921875,
      "logps/rejected": -411.59515380859375,
      "loss": 0.6064,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.18608322739601135,
      "rewards/margins": 0.23026804625988007,
      "rewards/rejected": -0.4163512587547302,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.577874068920446e-06,
      "logits/chosen": -2.239515781402588,
      "logits/rejected": -2.3043527603149414,
      "logps/chosen": -348.39581298828125,
      "logps/rejected": -414.6133728027344,
      "loss": 0.5915,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.22679059207439423,
      "rewards/margins": 0.23025627434253693,
      "rewards/rejected": -0.45704689621925354,
      "step": 70
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.0784519801008546e-06,
      "logits/chosen": -2.16733980178833,
      "logits/rejected": -2.2091336250305176,
      "logps/chosen": -339.76129150390625,
      "logps/rejected": -419.670654296875,
      "loss": 0.595,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.2677019238471985,
      "rewards/margins": 0.26828354597091675,
      "rewards/rejected": -0.53598552942276,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.553063458334059e-06,
      "logits/chosen": -2.177290439605713,
      "logits/rejected": -2.2133171558380127,
      "logps/chosen": -337.5870666503906,
      "logps/rejected": -410.07464599609375,
      "loss": 0.5779,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.21148499846458435,
      "rewards/margins": 0.3073815107345581,
      "rewards/rejected": -0.5188664793968201,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.025292943281429e-06,
      "logits/chosen": -2.137645721435547,
      "logits/rejected": -2.2139902114868164,
      "logps/chosen": -362.3526611328125,
      "logps/rejected": -451.25543212890625,
      "loss": 0.5749,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.38997265696525574,
      "rewards/margins": 0.3522176146507263,
      "rewards/rejected": -0.7421903610229492,
      "step": 100
    },
    {
      "epoch": 0.61,
      "eval_logits/chosen": -2.1221249103546143,
      "eval_logits/rejected": -2.0325379371643066,
      "eval_logps/chosen": -324.94329833984375,
      "eval_logps/rejected": -311.2961730957031,
      "eval_loss": 0.688376247882843,
      "eval_rewards/accuracies": 0.5440000295639038,
      "eval_rewards/chosen": -0.18393579125404358,
      "eval_rewards/margins": 0.05143474042415619,
      "eval_rewards/rejected": -0.23537051677703857,
      "eval_runtime": 384.2989,
      "eval_samples_per_second": 5.204,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5188318011445907e-06,
      "logits/chosen": -2.0991530418395996,
      "logits/rejected": -2.1866981983184814,
      "logps/chosen": -343.84478759765625,
      "logps/rejected": -443.0243225097656,
      "loss": 0.5695,
      "rewards/accuracies": 0.753125011920929,
      "rewards/chosen": -0.24366803467273712,
      "rewards/margins": 0.37467828392982483,
      "rewards/rejected": -0.6183463335037231,
      "step": 110
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0564148305586296e-06,
      "logits/chosen": -2.12058424949646,
      "logits/rejected": -2.183859348297119,
      "logps/chosen": -358.75238037109375,
      "logps/rejected": -454.56982421875,
      "loss": 0.5503,
      "rewards/accuracies": 0.796875,
      "rewards/chosen": -0.3431812822818756,
      "rewards/margins": 0.4134922921657562,
      "rewards/rejected": -0.7566735148429871,
      "step": 120
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.587997083462197e-07,
      "logits/chosen": -2.1187376976013184,
      "logits/rejected": -2.1680569648742676,
      "logps/chosen": -377.47918701171875,
      "logps/rejected": -470.26470947265625,
      "loss": 0.5556,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.4671626091003418,
      "rewards/margins": 0.353724867105484,
      "rewards/rejected": -0.8208874464035034,
      "step": 130
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.438351873250492e-07,
      "logits/chosen": -2.0515809059143066,
      "logits/rejected": -2.1326162815093994,
      "logps/chosen": -359.46771240234375,
      "logps/rejected": -465.7957458496094,
      "loss": 0.5475,
      "rewards/accuracies": 0.7906249761581421,
      "rewards/chosen": -0.36117881536483765,
      "rewards/margins": 0.44570302963256836,
      "rewards/rejected": -0.8068817853927612,
      "step": 140
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.2565987432367032e-07,
      "logits/chosen": -2.070026159286499,
      "logits/rejected": -2.1689717769622803,
      "logps/chosen": -349.19305419921875,
      "logps/rejected": -467.4447326660156,
      "loss": 0.541,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.32100656628608704,
      "rewards/margins": 0.48250946402549744,
      "rewards/rejected": -0.8035160303115845,
      "step": 150
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4067554877743861e-08,
      "logits/chosen": -2.0285654067993164,
      "logits/rejected": -2.1381633281707764,
      "logps/chosen": -329.7977294921875,
      "logps/rejected": -446.18780517578125,
      "loss": 0.5574,
      "rewards/accuracies": 0.778124988079071,
      "rewards/chosen": -0.29442131519317627,
      "rewards/margins": 0.48129719495773315,
      "rewards/rejected": -0.7757185697555542,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 165,
      "total_flos": 0.0,
      "train_loss": 0.5940336465835572,
      "train_runtime": 8165.5058,
      "train_samples_per_second": 2.588,
      "train_steps_per_second": 0.02
    }
  ],
  "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
}