{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 286,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 1.7241379310344825e-08,
      "logits/chosen": -2.46767258644104,
      "logits/rejected": -2.4595911502838135,
      "logps/chosen": -22.810455322265625,
      "logps/rejected": -22.677112579345703,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.7241379310344828e-07,
      "logits/chosen": -2.815382719039917,
      "logits/rejected": -2.7576191425323486,
      "logps/chosen": -19.62689971923828,
      "logps/rejected": -11.512568473815918,
      "loss": 0.6817,
      "rewards/accuracies": 0.6944444179534912,
      "rewards/chosen": -0.002836173167452216,
      "rewards/margins": 0.020304910838603973,
      "rewards/rejected": -0.023141082376241684,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.4482758620689656e-07,
      "logits/chosen": -2.760253429412842,
      "logits/rejected": -2.776731014251709,
      "logps/chosen": -18.500263214111328,
      "logps/rejected": -24.879409790039062,
      "loss": 0.3663,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -0.05979099124670029,
      "rewards/margins": 1.2453944683074951,
      "rewards/rejected": -1.3051855564117432,
      "step": 20
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.980544747081712e-07,
      "logits/chosen": -2.8509573936462402,
      "logits/rejected": -2.8971199989318848,
      "logps/chosen": -20.137048721313477,
      "logps/rejected": -61.92778778076172,
      "loss": 0.1236,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.16065621376037598,
      "rewards/margins": 4.86615514755249,
      "rewards/rejected": -5.026811599731445,
      "step": 30
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.785992217898833e-07,
      "logits/chosen": -2.792508602142334,
      "logits/rejected": -2.8558077812194824,
      "logps/chosen": -23.098939895629883,
      "logps/rejected": -88.87196350097656,
      "loss": 0.0921,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.44465622305870056,
      "rewards/margins": 7.166872978210449,
      "rewards/rejected": -7.611530303955078,
      "step": 40
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.591439688715953e-07,
      "logits/chosen": -2.8872122764587402,
      "logits/rejected": -2.8914027214050293,
      "logps/chosen": -21.817670822143555,
      "logps/rejected": -82.24397277832031,
      "loss": 0.0947,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.6026766896247864,
      "rewards/margins": 6.256636619567871,
      "rewards/rejected": -6.85931396484375,
      "step": 50
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.3968871595330735e-07,
      "logits/chosen": -2.8102235794067383,
      "logits/rejected": -2.867000102996826,
      "logps/chosen": -22.098957061767578,
      "logps/rejected": -106.98995208740234,
      "loss": 0.0684,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.6044292449951172,
      "rewards/margins": 8.731573104858398,
      "rewards/rejected": -9.336004257202148,
      "step": 60
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.202334630350194e-07,
      "logits/chosen": -2.838426113128662,
      "logits/rejected": -2.8441996574401855,
      "logps/chosen": -32.25341033935547,
      "logps/rejected": -112.56993103027344,
      "loss": 0.0467,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9181193113327026,
      "rewards/margins": 8.931884765625,
      "rewards/rejected": -9.850004196166992,
      "step": 70
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.007782101167315e-07,
      "logits/chosen": -2.7943294048309326,
      "logits/rejected": -2.81146240234375,
      "logps/chosen": -28.41353416442871,
      "logps/rejected": -115.3572769165039,
      "loss": 0.0531,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.8534424901008606,
      "rewards/margins": 9.425840377807617,
      "rewards/rejected": -10.27928352355957,
      "step": 80
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.813229571984435e-07,
      "logits/chosen": -2.6642768383026123,
      "logits/rejected": -2.695777416229248,
      "logps/chosen": -30.20281410217285,
      "logps/rejected": -115.2649917602539,
      "loss": 0.0493,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.9560843706130981,
      "rewards/margins": 9.298072814941406,
      "rewards/rejected": -10.254158020019531,
      "step": 90
    },
    {
      "epoch": 0.7,
      "learning_rate": 3.618677042801556e-07,
      "logits/chosen": -2.61277437210083,
      "logits/rejected": -2.7204833030700684,
      "logps/chosen": -33.576595306396484,
      "logps/rejected": -150.16848754882812,
      "loss": 0.0391,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -1.3517389297485352,
      "rewards/margins": 12.079813003540039,
      "rewards/rejected": -13.431551933288574,
      "step": 100
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.4241245136186767e-07,
      "logits/chosen": -2.7867584228515625,
      "logits/rejected": -2.7667887210845947,
      "logps/chosen": -24.368350982666016,
      "logps/rejected": -107.03987884521484,
      "loss": 0.0492,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.7897183895111084,
      "rewards/margins": 8.651564598083496,
      "rewards/rejected": -9.441283226013184,
      "step": 110
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.2295719844357975e-07,
      "logits/chosen": -2.6104767322540283,
      "logits/rejected": -2.675987482070923,
      "logps/chosen": -33.704322814941406,
      "logps/rejected": -135.02169799804688,
      "loss": 0.023,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -1.1443722248077393,
      "rewards/margins": 11.034332275390625,
      "rewards/rejected": -12.178705215454102,
      "step": 120
    },
    {
      "epoch": 0.91,
      "learning_rate": 3.0350194552529177e-07,
      "logits/chosen": -2.568190574645996,
      "logits/rejected": -2.6318023204803467,
      "logps/chosen": -30.85686683654785,
      "logps/rejected": -167.8513946533203,
      "loss": 0.0389,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -1.3852249383926392,
      "rewards/margins": 14.069249153137207,
      "rewards/rejected": -15.454473495483398,
      "step": 130
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.8404669260700384e-07,
      "logits/chosen": -2.5317270755767822,
      "logits/rejected": -2.593646764755249,
      "logps/chosen": -33.05012512207031,
      "logps/rejected": -199.79788208007812,
      "loss": 0.0281,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1634151935577393,
      "rewards/margins": 17.322010040283203,
      "rewards/rejected": -18.485424041748047,
      "step": 140
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.645914396887159e-07,
      "logits/chosen": -2.6736645698547363,
      "logits/rejected": -2.65024995803833,
      "logps/chosen": -32.11931610107422,
      "logps/rejected": -172.3567352294922,
      "loss": 0.011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.398026704788208,
      "rewards/margins": 14.626693725585938,
      "rewards/rejected": -16.02471923828125,
      "step": 150
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.45136186770428e-07,
      "logits/chosen": -2.6248397827148438,
      "logits/rejected": -2.6503357887268066,
      "logps/chosen": -35.05925369262695,
      "logps/rejected": -219.28378295898438,
      "loss": 0.0095,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.4844539165496826,
      "rewards/margins": 19.174663543701172,
      "rewards/rejected": -20.659120559692383,
      "step": 160
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.2568093385214007e-07,
      "logits/chosen": -2.6490864753723145,
      "logits/rejected": -2.6252989768981934,
      "logps/chosen": -31.650588989257812,
      "logps/rejected": -194.4810333251953,
      "loss": 0.0158,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9476087689399719,
      "rewards/margins": 17.060314178466797,
      "rewards/rejected": -18.007923126220703,
      "step": 170
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.0622568093385212e-07,
      "logits/chosen": -2.7093279361724854,
      "logits/rejected": -2.638579845428467,
      "logps/chosen": -32.82453155517578,
      "logps/rejected": -185.39230346679688,
      "loss": 0.0104,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.2247194051742554,
      "rewards/margins": 15.979866027832031,
      "rewards/rejected": -17.204586029052734,
      "step": 180
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.867704280155642e-07,
      "logits/chosen": -2.6887717247009277,
      "logits/rejected": -2.6667675971984863,
      "logps/chosen": -34.072509765625,
      "logps/rejected": -194.00869750976562,
      "loss": 0.0099,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.21720290184021,
      "rewards/margins": 16.749324798583984,
      "rewards/rejected": -17.96652603149414,
      "step": 190
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.6731517509727624e-07,
      "logits/chosen": -2.6063685417175293,
      "logits/rejected": -2.5614001750946045,
      "logps/chosen": -41.58473587036133,
      "logps/rejected": -219.73104858398438,
      "loss": 0.0121,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -1.829288125038147,
      "rewards/margins": 18.672809600830078,
      "rewards/rejected": -20.502098083496094,
      "step": 200
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.4785992217898832e-07,
      "logits/chosen": -2.7379167079925537,
      "logits/rejected": -2.674070119857788,
      "logps/chosen": -35.14494323730469,
      "logps/rejected": -182.973388671875,
      "loss": 0.0093,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3816758394241333,
      "rewards/margins": 15.531547546386719,
      "rewards/rejected": -16.913225173950195,
      "step": 210
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.2840466926070036e-07,
      "logits/chosen": -2.7067456245422363,
      "logits/rejected": -2.632633686065674,
      "logps/chosen": -34.483577728271484,
      "logps/rejected": -222.17990112304688,
      "loss": 0.0054,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9836966395378113,
      "rewards/margins": 19.769222259521484,
      "rewards/rejected": -20.752918243408203,
      "step": 220
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.0894941634241245e-07,
      "logits/chosen": -2.608846664428711,
      "logits/rejected": -2.604660749435425,
      "logps/chosen": -32.35722732543945,
      "logps/rejected": -246.7973175048828,
      "loss": 0.0076,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.2495921850204468,
      "rewards/margins": 22.08835220336914,
      "rewards/rejected": -23.33794403076172,
      "step": 230
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.949416342412451e-08,
      "logits/chosen": -2.607797622680664,
      "logits/rejected": -2.566305637359619,
      "logps/chosen": -27.041118621826172,
      "logps/rejected": -209.3827667236328,
      "loss": 0.0122,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9218173027038574,
      "rewards/margins": 18.870208740234375,
      "rewards/rejected": -19.79202651977539,
      "step": 240
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.003891050583658e-08,
      "logits/chosen": -2.518440008163452,
      "logits/rejected": -2.5331220626831055,
      "logps/chosen": -31.538745880126953,
      "logps/rejected": -247.35238647460938,
      "loss": 0.004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.2074544429779053,
      "rewards/margins": 22.2752742767334,
      "rewards/rejected": -23.48272705078125,
      "step": 250
    },
    {
      "epoch": 1.82,
      "learning_rate": 5.058365758754864e-08,
      "logits/chosen": -2.5735764503479004,
      "logits/rejected": -2.52424693107605,
      "logps/chosen": -28.21148681640625,
      "logps/rejected": -196.92958068847656,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0159205198287964,
      "rewards/margins": 17.290904998779297,
      "rewards/rejected": -18.306825637817383,
      "step": 260
    },
    {
      "epoch": 1.89,
      "learning_rate": 3.11284046692607e-08,
      "logits/chosen": -2.643744945526123,
      "logits/rejected": -2.626131772994995,
      "logps/chosen": -29.778423309326172,
      "logps/rejected": -234.2376251220703,
      "loss": 0.0074,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0768438577651978,
      "rewards/margins": 20.66169548034668,
      "rewards/rejected": -21.738544464111328,
      "step": 270
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.1673151750972762e-08,
      "logits/chosen": -2.467341899871826,
      "logits/rejected": -2.524524211883545,
      "logps/chosen": -29.820302963256836,
      "logps/rejected": -301.8487243652344,
      "loss": 0.0019,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9357167482376099,
      "rewards/margins": 27.8673152923584,
      "rewards/rejected": -28.80303382873535,
      "step": 280
    },
    {
      "epoch": 2.0,
      "step": 286,
      "total_flos": 0.0,
      "train_loss": 0.06552632141962543,
      "train_runtime": 3687.7187,
      "train_samples_per_second": 2.478,
      "train_steps_per_second": 0.078
    }
  ],
  "logging_steps": 10,
  "max_steps": 286,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}