{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 195,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.5000000000000004e-07,
      "logits/chosen": -2.530740737915039,
      "logits/rejected": -2.5715365409851074,
      "logps/chosen": -251.08152770996094,
      "logps/rejected": -311.3883972167969,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.5e-06,
      "logits/chosen": -2.6261370182037354,
      "logits/rejected": -2.549044370651245,
      "logps/chosen": -318.5334167480469,
      "logps/rejected": -310.1783447265625,
      "loss": 0.6928,
      "rewards/accuracies": 0.4791666567325592,
      "rewards/chosen": -0.008981944061815739,
      "rewards/margins": 0.0016753082163631916,
      "rewards/rejected": -0.010657250881195068,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-06,
      "logits/chosen": -2.5849127769470215,
      "logits/rejected": -2.5526692867279053,
      "logps/chosen": -310.12841796875,
      "logps/rejected": -311.95013427734375,
      "loss": 0.6892,
      "rewards/accuracies": 0.559374988079071,
      "rewards/chosen": -0.03645968809723854,
      "rewards/margins": 0.007315966300666332,
      "rewards/rejected": -0.04377565160393715,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.959823971496575e-06,
      "logits/chosen": -2.543229818344116,
      "logits/rejected": -2.4565789699554443,
      "logps/chosen": -312.2004699707031,
      "logps/rejected": -301.56103515625,
      "loss": 0.6783,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -0.03164006397128105,
      "rewards/margins": 0.040305931121110916,
      "rewards/rejected": -0.07194599509239197,
      "step": 30
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.8405871765993435e-06,
      "logits/chosen": -2.5465879440307617,
      "logits/rejected": -2.451957941055298,
      "logps/chosen": -331.6486511230469,
      "logps/rejected": -324.4945373535156,
      "loss": 0.6706,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -0.19176843762397766,
      "rewards/margins": 0.039799537509679794,
      "rewards/rejected": -0.23156797885894775,
      "step": 40
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.5046234130859375,
      "logits/rejected": -2.4880499839782715,
      "logps/chosen": -314.14642333984375,
      "logps/rejected": -340.1266174316406,
      "loss": 0.6585,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.15194456279277802,
      "rewards/margins": 0.08897710591554642,
      "rewards/rejected": -0.24092169106006622,
      "step": 50
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.382678665009028e-06,
      "logits/chosen": -2.4840145111083984,
      "logits/rejected": -2.402167558670044,
      "logps/chosen": -312.02252197265625,
      "logps/rejected": -343.81317138671875,
      "loss": 0.6559,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.19705083966255188,
      "rewards/margins": 0.08587387949228287,
      "rewards/rejected": -0.28292471170425415,
      "step": 60
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.058724504646834e-06,
      "logits/chosen": -2.4141929149627686,
      "logits/rejected": -2.388474941253662,
      "logps/chosen": -305.43341064453125,
      "logps/rejected": -347.254150390625,
      "loss": 0.645,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.2647872567176819,
      "rewards/margins": 0.14936725795269012,
      "rewards/rejected": -0.4141544699668884,
      "step": 70
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.684671656182497e-06,
      "logits/chosen": -2.4685239791870117,
      "logits/rejected": -2.3869450092315674,
      "logps/chosen": -293.1957092285156,
      "logps/rejected": -310.21990966796875,
      "loss": 0.6438,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.19762961566448212,
      "rewards/margins": 0.15006476640701294,
      "rewards/rejected": -0.34769436717033386,
      "step": 80
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.272542485937369e-06,
      "logits/chosen": -2.374202013015747,
      "logits/rejected": -2.352567195892334,
      "logps/chosen": -288.92327880859375,
      "logps/rejected": -301.8702392578125,
      "loss": 0.6483,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -0.17927215993404388,
      "rewards/margins": 0.11449799686670303,
      "rewards/rejected": -0.2937701642513275,
      "step": 90
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.294431447982788,
      "logits/rejected": -2.223918914794922,
      "logps/chosen": -330.19061279296875,
      "logps/rejected": -370.58306884765625,
      "loss": 0.635,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.3646013140678406,
      "rewards/margins": 0.20234891772270203,
      "rewards/rejected": -0.566950261592865,
      "step": 100
    },
    {
      "epoch": 0.51,
      "eval_logits/chosen": -2.2727713584899902,
      "eval_logits/rejected": -2.1430881023406982,
      "eval_logps/chosen": -348.326904296875,
      "eval_logps/rejected": -358.062255859375,
      "eval_loss": 0.6302204728126526,
      "eval_rewards/accuracies": 0.6439999938011169,
      "eval_rewards/chosen": -0.49924182891845703,
      "eval_rewards/margins": 0.18303123116493225,
      "eval_rewards/rejected": -0.6822730302810669,
      "eval_runtime": 384.1011,
      "eval_samples_per_second": 5.207,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.3878379241237136e-06,
      "logits/chosen": -2.1476898193359375,
      "logits/rejected": -2.064382314682007,
      "logps/chosen": -344.08465576171875,
      "logps/rejected": -359.03826904296875,
      "loss": 0.6318,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": -0.5136152505874634,
      "rewards/margins": 0.19320404529571533,
      "rewards/rejected": -0.7068192362785339,
      "step": 110
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9436976651092143e-06,
      "logits/chosen": -2.1169896125793457,
      "logits/rejected": -2.0298428535461426,
      "logps/chosen": -342.7489318847656,
      "logps/rejected": -361.8150634765625,
      "loss": 0.6217,
      "rewards/accuracies": 0.7093750238418579,
      "rewards/chosen": -0.423992782831192,
      "rewards/margins": 0.21997280418872833,
      "rewards/rejected": -0.6439656019210815,
      "step": 120
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5174374208651913e-06,
      "logits/chosen": -1.9802496433258057,
      "logits/rejected": -1.8583825826644897,
      "logps/chosen": -341.1569519042969,
      "logps/rejected": -375.6815490722656,
      "loss": 0.6046,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.4307962954044342,
      "rewards/margins": 0.2760288119316101,
      "rewards/rejected": -0.7068250775337219,
      "step": 130
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.122757546369744e-06,
      "logits/chosen": -1.9369285106658936,
      "logits/rejected": -1.7146968841552734,
      "logps/chosen": -376.5108642578125,
      "logps/rejected": -381.532470703125,
      "loss": 0.6116,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": -0.41971555352211,
      "rewards/margins": 0.23398633301258087,
      "rewards/rejected": -0.653701901435852,
      "step": 140
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -1.7080371379852295,
      "logits/rejected": -1.5588016510009766,
      "logps/chosen": -338.8884582519531,
      "logps/rejected": -386.35211181640625,
      "loss": 0.6146,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.4495311677455902,
      "rewards/margins": 0.23771929740905762,
      "rewards/rejected": -0.6872504353523254,
      "step": 150
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.774575140626317e-07,
      "logits/chosen": -1.5616605281829834,
      "logits/rejected": -1.4405885934829712,
      "logps/chosen": -346.27801513671875,
      "logps/rejected": -376.26641845703125,
      "loss": 0.6089,
      "rewards/accuracies": 0.6781250238418579,
      "rewards/chosen": -0.5472201108932495,
      "rewards/margins": 0.22093167901039124,
      "rewards/rejected": -0.7681517004966736,
      "step": 160
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4757783024395244e-07,
      "logits/chosen": -1.4499655961990356,
      "logits/rejected": -1.4005109071731567,
      "logps/chosen": -340.2152099609375,
      "logps/rejected": -365.9745788574219,
      "loss": 0.6172,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": -0.5039272904396057,
      "rewards/margins": 0.2527262270450592,
      "rewards/rejected": -0.7566534876823425,
      "step": 170
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.00928482603669e-08,
      "logits/chosen": -1.6221641302108765,
      "logits/rejected": -1.3607091903686523,
      "logps/chosen": -371.7226867675781,
      "logps/rejected": -376.9884338378906,
      "loss": 0.6106,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.49880844354629517,
      "rewards/margins": 0.2649240493774414,
      "rewards/rejected": -0.7637324929237366,
      "step": 180
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.006426501190233e-08,
      "logits/chosen": -1.5273410081863403,
      "logits/rejected": -1.343459129333496,
      "logps/chosen": -345.67236328125,
      "logps/rejected": -376.15081787109375,
      "loss": 0.6117,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.5101321339607239,
      "rewards/margins": 0.24796441197395325,
      "rewards/rejected": -0.7580965757369995,
      "step": 190
    },
    {
      "epoch": 1.0,
      "step": 195,
      "total_flos": 0.0,
      "train_loss": 0.6390555693553044,
      "train_runtime": 8832.9066,
      "train_samples_per_second": 2.83,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 195,
  "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
}