{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 100,
  "global_step": 192,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010416666666666666,
      "grad_norm": 0.6369225577240899,
      "learning_rate": 4.4661867031221926e-08,
      "logits/chosen": -2.590585231781006,
      "logits/rejected": -2.5664222240448,
      "logps/chosen": -80.29847717285156,
      "logps/rejected": -53.10200881958008,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.10416666666666667,
      "grad_norm": 0.5699844472823067,
      "learning_rate": 4.4661867031221926e-07,
      "logits/chosen": -2.5559396743774414,
      "logits/rejected": -2.5378711223602295,
      "logps/chosen": -87.90701293945312,
      "logps/rejected": -81.000732421875,
      "loss": 0.6931,
      "rewards/accuracies": 0.1944444477558136,
      "rewards/chosen": 7.878588803578168e-06,
      "rewards/margins": -2.865187343559228e-05,
      "rewards/rejected": 3.653046223917045e-05,
      "step": 10
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 0.5220617909669454,
      "learning_rate": 8.932373406244385e-07,
      "logits/chosen": -2.6017892360687256,
      "logits/rejected": -2.5524039268493652,
      "logps/chosen": -102.29816436767578,
      "logps/rejected": -88.93941497802734,
      "loss": 0.6929,
      "rewards/accuracies": 0.32499998807907104,
      "rewards/chosen": 0.0010181397665292025,
      "rewards/margins": 0.00010430384281789884,
      "rewards/rejected": 0.0009138360619544983,
      "step": 20
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.8025622814810965,
      "learning_rate": 8.85808141328471e-07,
      "logits/chosen": -2.4462294578552246,
      "logits/rejected": -2.4619638919830322,
      "logps/chosen": -66.71989440917969,
      "logps/rejected": -76.12188720703125,
      "loss": 0.6922,
      "rewards/accuracies": 0.32499998807907104,
      "rewards/chosen": 0.0006937496364116669,
      "rewards/margins": 0.001850384520366788,
      "rewards/rejected": -0.0011566350003704429,
      "step": 30
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 0.8299819796102439,
      "learning_rate": 8.63767702844543e-07,
      "logits/chosen": -2.4784650802612305,
      "logits/rejected": -2.4647819995880127,
      "logps/chosen": -74.95284271240234,
      "logps/rejected": -78.62578582763672,
      "loss": 0.6904,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": -0.002713928697630763,
      "rewards/margins": 0.004938088357448578,
      "rewards/rejected": -0.007652017287909985,
      "step": 40
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 1.7129675129738506,
      "learning_rate": 8.278492807257787e-07,
      "logits/chosen": -2.275789260864258,
      "logits/rejected": -2.291548252105713,
      "logps/chosen": -55.84526824951172,
      "logps/rejected": -68.4447250366211,
      "loss": 0.6892,
      "rewards/accuracies": 0.20000000298023224,
      "rewards/chosen": -0.004781405441462994,
      "rewards/margins": 0.004440293647348881,
      "rewards/rejected": -0.009221700020134449,
      "step": 50
    },
    {
      "epoch": 0.625,
      "grad_norm": 1.6503123985933317,
      "learning_rate": 7.792478322547763e-07,
      "logits/chosen": -2.007650136947632,
      "logits/rejected": -1.9700886011123657,
      "logps/chosen": -114.91951751708984,
      "logps/rejected": -133.2468719482422,
      "loss": 0.6865,
      "rewards/accuracies": 0.30000001192092896,
      "rewards/chosen": -0.03467016667127609,
      "rewards/margins": 0.017464537173509598,
      "rewards/rejected": -0.05213470384478569,
      "step": 60
    },
    {
      "epoch": 0.7291666666666666,
      "grad_norm": 2.2984298301869166,
      "learning_rate": 7.195802618323811e-07,
      "logits/chosen": -0.7167578935623169,
      "logits/rejected": -0.5914499163627625,
      "logps/chosen": -169.179443359375,
      "logps/rejected": -187.8536376953125,
      "loss": 0.6802,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": -0.07043690979480743,
      "rewards/margins": 0.02859547734260559,
      "rewards/rejected": -0.09903239458799362,
      "step": 70
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 7.779524343348389,
      "learning_rate": 6.508316287570068e-07,
      "logits/chosen": 0.3923811912536621,
      "logits/rejected": 0.710164487361908,
      "logps/chosen": -222.629638671875,
      "logps/rejected": -283.06634521484375,
      "loss": 0.669,
      "rewards/accuracies": 0.3687500059604645,
      "rewards/chosen": -0.13247501850128174,
      "rewards/margins": 0.06450365483760834,
      "rewards/rejected": -0.19697865843772888,
      "step": 80
    },
    {
      "epoch": 0.9375,
      "grad_norm": 6.759384309065668,
      "learning_rate": 5.752891069888697e-07,
      "logits/chosen": 1.1880756616592407,
      "logits/rejected": 1.3526252508163452,
      "logps/chosen": -208.9159393310547,
      "logps/rejected": -350.1947021484375,
      "loss": 0.6679,
      "rewards/accuracies": 0.26249998807907104,
      "rewards/chosen": -0.14983566105365753,
      "rewards/margins": 0.12745192646980286,
      "rewards/rejected": -0.2772876024246216,
      "step": 90
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 6.64561814006091,
      "learning_rate": 4.954658939683863e-07,
      "logits/chosen": 1.6723073720932007,
      "logits/rejected": 2.053896427154541,
      "logps/chosen": -224.61581420898438,
      "logps/rejected": -340.3245849609375,
      "loss": 0.66,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": -0.14611998200416565,
      "rewards/margins": 0.11334720999002457,
      "rewards/rejected": -0.2594671845436096,
      "step": 100
    },
    {
      "epoch": 1.0416666666666667,
      "eval_logits/chosen": 2.222659111022949,
      "eval_logits/rejected": 2.3998076915740967,
      "eval_logps/chosen": -231.40858459472656,
      "eval_logps/rejected": -327.22540283203125,
      "eval_loss": 0.6709679961204529,
      "eval_rewards/accuracies": 0.3154761791229248,
      "eval_rewards/chosen": -0.15175169706344604,
      "eval_rewards/margins": 0.08624394983053207,
      "eval_rewards/rejected": -0.2379956692457199,
      "eval_runtime": 113.6716,
      "eval_samples_per_second": 17.595,
      "eval_steps_per_second": 0.554,
      "step": 100
    },
    {
      "epoch": 1.1458333333333333,
      "grad_norm": 10.14024135671265,
      "learning_rate": 4.140175999393427e-07,
      "logits/chosen": 2.7885522842407227,
      "logits/rejected": 2.5952858924865723,
      "logps/chosen": -217.1932830810547,
      "logps/rejected": -423.316162109375,
      "loss": 0.6344,
      "rewards/accuracies": 0.375,
      "rewards/chosen": -0.14682665467262268,
      "rewards/margins": 0.18375235795974731,
      "rewards/rejected": -0.3305789828300476,
      "step": 110
    },
    {
      "epoch": 1.25,
      "grad_norm": 5.687581380009954,
      "learning_rate": 3.3365389939087466e-07,
      "logits/chosen": 2.281104564666748,
      "logits/rejected": 2.7267863750457764,
      "logps/chosen": -382.77508544921875,
      "logps/rejected": -542.299560546875,
      "loss": 0.6531,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.2657240629196167,
      "rewards/margins": 0.1684594750404358,
      "rewards/rejected": -0.43418359756469727,
      "step": 120
    },
    {
      "epoch": 1.3541666666666667,
      "grad_norm": 8.59716391066745,
      "learning_rate": 2.5704838385518876e-07,
      "logits/chosen": 2.97074556350708,
      "logits/rejected": 3.06956148147583,
      "logps/chosen": -262.03167724609375,
      "logps/rejected": -430.7930603027344,
      "loss": 0.628,
      "rewards/accuracies": 0.36250001192092896,
      "rewards/chosen": -0.16968943178653717,
      "rewards/margins": 0.16052642464637756,
      "rewards/rejected": -0.3302158713340759,
      "step": 130
    },
    {
      "epoch": 1.4583333333333333,
      "grad_norm": 11.108890574756282,
      "learning_rate": 1.8674961513641657e-07,
      "logits/chosen": 3.3152003288269043,
      "logits/rejected": 3.317988634109497,
      "logps/chosen": -273.95831298828125,
      "logps/rejected": -546.5203247070312,
      "loss": 0.6343,
      "rewards/accuracies": 0.3499999940395355,
      "rewards/chosen": -0.21727445721626282,
      "rewards/margins": 0.2482767552137375,
      "rewards/rejected": -0.4655512273311615,
      "step": 140
    },
    {
      "epoch": 1.5625,
      "grad_norm": 6.0584348916117925,
      "learning_rate": 1.250963381092756e-07,
      "logits/chosen": 3.313689708709717,
      "logits/rejected": 3.6598522663116455,
      "logps/chosen": -273.2975769042969,
      "logps/rejected": -551.4356689453125,
      "loss": 0.6207,
      "rewards/accuracies": 0.38749998807907104,
      "rewards/chosen": -0.19452792406082153,
      "rewards/margins": 0.2676007151603699,
      "rewards/rejected": -0.4621286988258362,
      "step": 150
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 11.564568090628018,
      "learning_rate": 7.413967384294965e-08,
      "logits/chosen": 3.5584030151367188,
      "logits/rejected": 3.723942279815674,
      "logps/chosen": -357.8267517089844,
      "logps/rejected": -687.1414794921875,
      "loss": 0.6181,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.2677861154079437,
      "rewards/margins": 0.31169238686561584,
      "rewards/rejected": -0.5794785022735596,
      "step": 160
    },
    {
      "epoch": 1.7708333333333335,
      "grad_norm": 12.348100258027504,
      "learning_rate": 3.557488157963567e-08,
      "logits/chosen": 4.085668087005615,
      "logits/rejected": 4.25767183303833,
      "logps/chosen": -461.731689453125,
      "logps/rejected": -780.9080810546875,
      "loss": 0.6175,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": -0.3737877607345581,
      "rewards/margins": 0.30115869641304016,
      "rewards/rejected": -0.6749464273452759,
      "step": 170
    },
    {
      "epoch": 1.875,
      "grad_norm": 14.884569591942372,
      "learning_rate": 1.0684959754358576e-08,
      "logits/chosen": 3.8572006225585938,
      "logits/rejected": 4.063995361328125,
      "logps/chosen": -347.04864501953125,
      "logps/rejected": -606.2640380859375,
      "loss": 0.617,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.26613375544548035,
      "rewards/margins": 0.248983234167099,
      "rewards/rejected": -0.5151170492172241,
      "step": 180
    },
    {
      "epoch": 1.9791666666666665,
      "grad_norm": 7.915820342970624,
      "learning_rate": 2.979623737803134e-10,
      "logits/chosen": 3.7023074626922607,
      "logits/rejected": 3.886380434036255,
      "logps/chosen": -341.0790710449219,
      "logps/rejected": -703.4373779296875,
      "loss": 0.6085,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": -0.24430128931999207,
      "rewards/margins": 0.3483408987522125,
      "rewards/rejected": -0.5926421880722046,
      "step": 190
    },
    {
      "epoch": 2.0,
      "step": 192,
      "total_flos": 0.0,
      "train_loss": 0.6542075766871372,
      "train_runtime": 2362.1376,
      "train_samples_per_second": 5.176,
      "train_steps_per_second": 0.081
    }
  ],
  "logging_steps": 10,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}