{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 478,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02092050209205021,
      "grad_norm": 1.1945245265960693,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -0.30859375,
      "logits/rejected": -0.2890625,
      "logps/chosen": -312.0,
      "logps/rejected": -288.0,
      "loss": 0.692,
      "rewards/accuracies": 0.26249998807907104,
      "rewards/chosen": 0.0007476806640625,
      "rewards/margins": 0.0005035400390625,
      "rewards/rejected": 0.000240325927734375,
      "step": 10
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 1.1291463375091553,
      "learning_rate": 8.333333333333333e-07,
      "logits/chosen": -0.330078125,
      "logits/rejected": -0.294921875,
      "logps/chosen": -302.0,
      "logps/rejected": -268.0,
      "loss": 0.6921,
      "rewards/accuracies": 0.3531250059604645,
      "rewards/chosen": -0.00098419189453125,
      "rewards/margins": 0.000614166259765625,
      "rewards/rejected": -0.00160980224609375,
      "step": 20
    },
    {
      "epoch": 0.06276150627615062,
      "grad_norm": 1.1083910465240479,
      "learning_rate": 9.995691082675907e-07,
      "logits/chosen": -0.22265625,
      "logits/rejected": -0.2314453125,
      "logps/chosen": -338.0,
      "logps/rejected": -306.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0006103515625,
      "rewards/margins": 0.005645751953125,
      "rewards/rejected": -0.006256103515625,
      "step": 30
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 1.1081091165542603,
      "learning_rate": 9.969385700404345e-07,
      "logits/chosen": -0.35546875,
      "logits/rejected": -0.34765625,
      "logps/chosen": -328.0,
      "logps/rejected": -300.0,
      "loss": 0.6882,
      "rewards/accuracies": 0.5718749761581421,
      "rewards/chosen": 0.003509521484375,
      "rewards/margins": 0.011474609375,
      "rewards/rejected": -0.0079345703125,
      "step": 40
    },
    {
      "epoch": 0.10460251046025104,
      "grad_norm": 1.1411256790161133,
      "learning_rate": 9.91929453572245e-07,
      "logits/chosen": -0.259765625,
      "logits/rejected": -0.279296875,
      "logps/chosen": -342.0,
      "logps/rejected": -284.0,
      "loss": 0.6843,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": 0.0086669921875,
      "rewards/margins": 0.02197265625,
      "rewards/rejected": -0.0133056640625,
      "step": 50
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 1.1798174381256104,
      "learning_rate": 9.845657348152955e-07,
      "logits/chosen": -0.33203125,
      "logits/rejected": -0.28515625,
      "logps/chosen": -330.0,
      "logps/rejected": -272.0,
      "loss": 0.6789,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.0038909912109375,
      "rewards/margins": 0.033203125,
      "rewards/rejected": -0.029296875,
      "step": 60
    },
    {
      "epoch": 0.14644351464435146,
      "grad_norm": 1.3184058666229248,
      "learning_rate": 9.748826599393632e-07,
      "logits/chosen": -0.396484375,
      "logits/rejected": -0.3046875,
      "logps/chosen": -282.0,
      "logps/rejected": -268.0,
      "loss": 0.675,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.000690460205078125,
      "rewards/margins": 0.043212890625,
      "rewards/rejected": -0.0439453125,
      "step": 70
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 1.1700127124786377,
      "learning_rate": 9.629265766272291e-07,
      "logits/chosen": -0.49609375,
      "logits/rejected": -0.404296875,
      "logps/chosen": -302.0,
      "logps/rejected": -290.0,
      "loss": 0.6704,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.005828857421875,
      "rewards/margins": 0.03955078125,
      "rewards/rejected": -0.04541015625,
      "step": 80
    },
    {
      "epoch": 0.18828451882845187,
      "grad_norm": 1.171791672706604,
      "learning_rate": 9.487547122331964e-07,
      "logits/chosen": -0.4609375,
      "logits/rejected": -0.412109375,
      "logps/chosen": -314.0,
      "logps/rejected": -308.0,
      "loss": 0.6674,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.014404296875,
      "rewards/margins": 0.05859375,
      "rewards/rejected": -0.0732421875,
      "step": 90
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 1.2314943075180054,
      "learning_rate": 9.324348998664548e-07,
      "logits/chosen": -0.453125,
      "logits/rejected": -0.41796875,
      "logps/chosen": -312.0,
      "logps/rejected": -294.0,
      "loss": 0.6568,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.01043701171875,
      "rewards/margins": 0.083984375,
      "rewards/rejected": -0.0947265625,
      "step": 100
    },
    {
      "epoch": 0.2301255230125523,
      "grad_norm": 1.1103886365890503,
      "learning_rate": 9.140452537103941e-07,
      "logits/chosen": -0.55859375,
      "logits/rejected": -0.5625,
      "logps/chosen": -330.0,
      "logps/rejected": -300.0,
      "loss": 0.6515,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": -0.045166015625,
      "rewards/margins": 0.0888671875,
      "rewards/rejected": -0.1337890625,
      "step": 110
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 1.180451512336731,
      "learning_rate": 8.936737951319275e-07,
      "logits/chosen": -0.609375,
      "logits/rejected": -0.5625,
      "logps/chosen": -342.0,
      "logps/rejected": -316.0,
      "loss": 0.6445,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.06494140625,
      "rewards/margins": 0.1513671875,
      "rewards/rejected": -0.2158203125,
      "step": 120
    },
    {
      "epoch": 0.2719665271966527,
      "grad_norm": 1.2995760440826416,
      "learning_rate": 8.714180313704489e-07,
      "logits/chosen": -0.7109375,
      "logits/rejected": -0.578125,
      "logps/chosen": -340.0,
      "logps/rejected": -328.0,
      "loss": 0.6395,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.111328125,
      "rewards/margins": 0.1591796875,
      "rewards/rejected": -0.26953125,
      "step": 130
    },
    {
      "epoch": 0.2928870292887029,
      "grad_norm": 1.2812036275863647,
      "learning_rate": 8.473844888230064e-07,
      "logits/chosen": -0.6875,
      "logits/rejected": -0.64453125,
      "logps/chosen": -332.0,
      "logps/rejected": -290.0,
      "loss": 0.6381,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.1845703125,
      "rewards/margins": 0.158203125,
      "rewards/rejected": -0.34375,
      "step": 140
    },
    {
      "epoch": 0.3138075313807531,
      "grad_norm": 1.8293964862823486,
      "learning_rate": 8.216882031596096e-07,
      "logits/chosen": -0.73046875,
      "logits/rejected": -0.6953125,
      "logps/chosen": -340.0,
      "logps/rejected": -322.0,
      "loss": 0.6355,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": -0.220703125,
      "rewards/margins": 0.17578125,
      "rewards/rejected": -0.396484375,
      "step": 150
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 1.3217679262161255,
      "learning_rate": 7.944521687092142e-07,
      "logits/chosen": -0.87109375,
      "logits/rejected": -0.8046875,
      "logps/chosen": -344.0,
      "logps/rejected": -320.0,
      "loss": 0.6231,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.21875,
      "rewards/margins": 0.240234375,
      "rewards/rejected": -0.458984375,
      "step": 160
    },
    {
      "epoch": 0.35564853556485354,
      "grad_norm": 1.493200659751892,
      "learning_rate": 7.658067497518772e-07,
      "logits/chosen": -0.94140625,
      "logits/rejected": -0.92578125,
      "logps/chosen": -324.0,
      "logps/rejected": -308.0,
      "loss": 0.6278,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.193359375,
      "rewards/margins": 0.1865234375,
      "rewards/rejected": -0.380859375,
      "step": 170
    },
    {
      "epoch": 0.37656903765690375,
      "grad_norm": 1.3974004983901978,
      "learning_rate": 7.358890565349105e-07,
      "logits/chosen": -0.90234375,
      "logits/rejected": -0.96484375,
      "logps/chosen": -340.0,
      "logps/rejected": -320.0,
      "loss": 0.6182,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.2333984375,
      "rewards/margins": 0.216796875,
      "rewards/rejected": -0.451171875,
      "step": 180
    },
    {
      "epoch": 0.39748953974895396,
      "grad_norm": 1.4906997680664062,
      "learning_rate": 7.048422889997115e-07,
      "logits/chosen": -0.96484375,
      "logits/rejected": -0.91796875,
      "logps/chosen": -342.0,
      "logps/rejected": -330.0,
      "loss": 0.632,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.2578125,
      "rewards/margins": 0.212890625,
      "rewards/rejected": -0.470703125,
      "step": 190
    },
    {
      "epoch": 0.41841004184100417,
      "grad_norm": 1.4964449405670166,
      "learning_rate": 6.72815051360494e-07,
      "logits/chosen": -1.03125,
      "logits/rejected": -1.015625,
      "logps/chosen": -324.0,
      "logps/rejected": -314.0,
      "loss": 0.6232,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.263671875,
      "rewards/margins": 0.1962890625,
      "rewards/rejected": -0.458984375,
      "step": 200
    },
    {
      "epoch": 0.4393305439330544,
      "grad_norm": 1.7729754447937012,
      "learning_rate": 6.399606408156687e-07,
      "logits/chosen": -1.046875,
      "logits/rejected": -1.0078125,
      "logps/chosen": -316.0,
      "logps/rejected": -316.0,
      "loss": 0.6014,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.2470703125,
      "rewards/margins": 0.2431640625,
      "rewards/rejected": -0.490234375,
      "step": 210
    },
    {
      "epoch": 0.4602510460251046,
      "grad_norm": 1.4692357778549194,
      "learning_rate": 6.064363137964225e-07,
      "logits/chosen": -1.125,
      "logits/rejected": -0.9765625,
      "logps/chosen": -348.0,
      "logps/rejected": -358.0,
      "loss": 0.6114,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.267578125,
      "rewards/margins": 0.30859375,
      "rewards/rejected": -0.578125,
      "step": 220
    },
    {
      "epoch": 0.4811715481171548,
      "grad_norm": 1.377569317817688,
      "learning_rate": 5.724025332645793e-07,
      "logits/chosen": -1.0625,
      "logits/rejected": -1.0390625,
      "logps/chosen": -350.0,
      "logps/rejected": -344.0,
      "loss": 0.6098,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.2890625,
      "rewards/margins": 0.234375,
      "rewards/rejected": -0.5234375,
      "step": 230
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 1.5865265130996704,
      "learning_rate": 5.380222006625179e-07,
      "logits/chosen": -1.1171875,
      "logits/rejected": -1.078125,
      "logps/chosen": -334.0,
      "logps/rejected": -328.0,
      "loss": 0.6252,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.26953125,
      "rewards/margins": 0.263671875,
      "rewards/rejected": -0.53125,
      "step": 240
    },
    {
      "epoch": 0.5230125523012552,
      "grad_norm": 1.5621623992919922,
      "learning_rate": 5.034598761913916e-07,
      "logits/chosen": -1.140625,
      "logits/rejected": -1.1015625,
      "logps/chosen": -318.0,
      "logps/rejected": -318.0,
      "loss": 0.6285,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.248046875,
      "rewards/margins": 0.2314453125,
      "rewards/rejected": -0.478515625,
      "step": 250
    },
    {
      "epoch": 0.5439330543933054,
      "grad_norm": 1.3323726654052734,
      "learning_rate": 4.688809911497609e-07,
      "logits/chosen": -1.125,
      "logits/rejected": -1.109375,
      "logps/chosen": -346.0,
      "logps/rejected": -350.0,
      "loss": 0.6079,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.240234375,
      "rewards/margins": 0.30859375,
      "rewards/rejected": -0.546875,
      "step": 260
    },
    {
      "epoch": 0.5648535564853556,
      "grad_norm": 1.4896010160446167,
      "learning_rate": 4.344510561027498e-07,
      "logits/chosen": -1.15625,
      "logits/rejected": -1.15625,
      "logps/chosen": -346.0,
      "logps/rejected": -330.0,
      "loss": 0.5974,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.271484375,
      "rewards/margins": 0.310546875,
      "rewards/rejected": -0.58203125,
      "step": 270
    },
    {
      "epoch": 0.5857740585774058,
      "grad_norm": 1.4263241291046143,
      "learning_rate": 4.003348686717949e-07,
      "logits/chosen": -1.1953125,
      "logits/rejected": -1.1171875,
      "logps/chosen": -338.0,
      "logps/rejected": -356.0,
      "loss": 0.6141,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.27734375,
      "rewards/margins": 0.306640625,
      "rewards/rejected": -0.58203125,
      "step": 280
    },
    {
      "epoch": 0.606694560669456,
      "grad_norm": 1.510299563407898,
      "learning_rate": 3.666957247368757e-07,
      "logits/chosen": -1.109375,
      "logits/rejected": -1.078125,
      "logps/chosen": -338.0,
      "logps/rejected": -322.0,
      "loss": 0.5992,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.2451171875,
      "rewards/margins": 0.345703125,
      "rewards/rejected": -0.58984375,
      "step": 290
    },
    {
      "epoch": 0.6276150627615062,
      "grad_norm": 1.8962136507034302,
      "learning_rate": 3.3369463682677234e-07,
      "logits/chosen": -1.1875,
      "logits/rejected": -1.1484375,
      "logps/chosen": -362.0,
      "logps/rejected": -362.0,
      "loss": 0.6158,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.267578125,
      "rewards/margins": 0.32421875,
      "rewards/rejected": -0.59375,
      "step": 300
    },
    {
      "epoch": 0.6485355648535565,
      "grad_norm": 1.5493396520614624,
      "learning_rate": 3.014895634385014e-07,
      "logits/chosen": -1.109375,
      "logits/rejected": -1.140625,
      "logps/chosen": -352.0,
      "logps/rejected": -340.0,
      "loss": 0.601,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.306640625,
      "rewards/margins": 0.2333984375,
      "rewards/rejected": -0.5390625,
      "step": 310
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 1.5511579513549805,
      "learning_rate": 2.7023465297476424e-07,
      "logits/chosen": -1.15625,
      "logits/rejected": -1.15625,
      "logps/chosen": -354.0,
      "logps/rejected": -358.0,
      "loss": 0.595,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.283203125,
      "rewards/margins": 0.326171875,
      "rewards/rejected": -0.609375,
      "step": 320
    },
    {
      "epoch": 0.6903765690376569,
      "grad_norm": 1.7871730327606201,
      "learning_rate": 2.4007950591826913e-07,
      "logits/chosen": -1.1328125,
      "logits/rejected": -1.125,
      "logps/chosen": -350.0,
      "logps/rejected": -352.0,
      "loss": 0.612,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.330078125,
      "rewards/margins": 0.30078125,
      "rewards/rejected": -0.6328125,
      "step": 330
    },
    {
      "epoch": 0.7112970711297071,
      "grad_norm": 1.6344778537750244,
      "learning_rate": 2.1116845877450805e-07,
      "logits/chosen": -1.1953125,
      "logits/rejected": -1.171875,
      "logps/chosen": -330.0,
      "logps/rejected": -316.0,
      "loss": 0.6125,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.32421875,
      "rewards/margins": 0.255859375,
      "rewards/rejected": -0.58203125,
      "step": 340
    },
    {
      "epoch": 0.7322175732217573,
      "grad_norm": 1.6645337343215942,
      "learning_rate": 1.8363989321036577e-07,
      "logits/chosen": -1.296875,
      "logits/rejected": -1.203125,
      "logps/chosen": -332.0,
      "logps/rejected": -340.0,
      "loss": 0.6038,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.333984375,
      "rewards/margins": 0.26953125,
      "rewards/rejected": -0.6015625,
      "step": 350
    },
    {
      "epoch": 0.7531380753138075,
      "grad_norm": 1.5749666690826416,
      "learning_rate": 1.5762557369534708e-07,
      "logits/chosen": -1.15625,
      "logits/rejected": -1.1875,
      "logps/chosen": -322.0,
      "logps/rejected": -348.0,
      "loss": 0.6019,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.38671875,
      "rewards/margins": 0.265625,
      "rewards/rejected": -0.65234375,
      "step": 360
    },
    {
      "epoch": 0.7740585774058577,
      "grad_norm": 1.582518458366394,
      "learning_rate": 1.332500168157748e-07,
      "logits/chosen": -1.15625,
      "logits/rejected": -1.203125,
      "logps/chosen": -360.0,
      "logps/rejected": -348.0,
      "loss": 0.6147,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.412109375,
      "rewards/margins": 0.25390625,
      "rewards/rejected": -0.6640625,
      "step": 370
    },
    {
      "epoch": 0.7949790794979079,
      "grad_norm": 1.690048098564148,
      "learning_rate": 1.1062989528071681e-07,
      "logits/chosen": -1.1953125,
      "logits/rejected": -1.1796875,
      "logps/chosen": -364.0,
      "logps/rejected": -378.0,
      "loss": 0.6108,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.349609375,
      "rewards/margins": 0.2890625,
      "rewards/rejected": -0.63671875,
      "step": 380
    },
    {
      "epoch": 0.8158995815899581,
      "grad_norm": 1.6943994760513306,
      "learning_rate": 8.987347947234192e-08,
      "logits/chosen": -1.109375,
      "logits/rejected": -1.1328125,
      "logps/chosen": -328.0,
      "logps/rejected": -324.0,
      "loss": 0.6044,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.3203125,
      "rewards/margins": 0.322265625,
      "rewards/rejected": -0.64453125,
      "step": 390
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 1.6517119407653809,
      "learning_rate": 7.108011921370727e-08,
      "logits/chosen": -1.1875,
      "logits/rejected": -1.1640625,
      "logps/chosen": -386.0,
      "logps/rejected": -382.0,
      "loss": 0.6041,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.375,
      "rewards/margins": 0.328125,
      "rewards/rejected": -0.703125,
      "step": 400
    },
    {
      "epoch": 0.8577405857740585,
      "grad_norm": 1.6941229104995728,
      "learning_rate": 5.433976823447262e-08,
      "logits/chosen": -1.203125,
      "logits/rejected": -1.1015625,
      "logps/chosen": -340.0,
      "logps/rejected": -352.0,
      "loss": 0.6026,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.392578125,
      "rewards/margins": 0.333984375,
      "rewards/rejected": -0.7265625,
      "step": 410
    },
    {
      "epoch": 0.8786610878661087,
      "grad_norm": 1.6011600494384766,
      "learning_rate": 3.973255361067346e-08,
      "logits/chosen": -1.2265625,
      "logits/rejected": -1.1875,
      "logps/chosen": -362.0,
      "logps/rejected": -352.0,
      "loss": 0.5924,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.34375,
      "rewards/margins": 0.2890625,
      "rewards/rejected": -0.6328125,
      "step": 420
    },
    {
      "epoch": 0.899581589958159,
      "grad_norm": 1.7107675075531006,
      "learning_rate": 2.732839223940914e-08,
      "logits/chosen": -1.1484375,
      "logits/rejected": -1.1953125,
      "logps/chosen": -356.0,
      "logps/rejected": -366.0,
      "loss": 0.5962,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.373046875,
      "rewards/margins": 0.3671875,
      "rewards/rejected": -0.7421875,
      "step": 430
    },
    {
      "epoch": 0.9205020920502092,
      "grad_norm": 1.5065439939498901,
      "learning_rate": 1.7186656184179473e-08,
      "logits/chosen": -1.203125,
      "logits/rejected": -1.2109375,
      "logps/chosen": -364.0,
      "logps/rejected": -374.0,
      "loss": 0.6021,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.361328125,
      "rewards/margins": 0.373046875,
      "rewards/rejected": -0.734375,
      "step": 440
    },
    {
      "epoch": 0.9414225941422594,
      "grad_norm": 1.560874104499817,
      "learning_rate": 9.355888492680153e-09,
      "logits/chosen": -1.1328125,
      "logits/rejected": -1.1015625,
      "logps/chosen": -340.0,
      "logps/rejected": -346.0,
      "loss": 0.5957,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.337890625,
      "rewards/margins": 0.357421875,
      "rewards/rejected": -0.6953125,
      "step": 450
    },
    {
      "epoch": 0.9623430962343096,
      "grad_norm": 1.5713697671890259,
      "learning_rate": 3.873570847285012e-09,
      "logits/chosen": -1.234375,
      "logits/rejected": -1.2109375,
      "logps/chosen": -356.0,
      "logps/rejected": -356.0,
      "loss": 0.6004,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.40234375,
      "rewards/margins": 0.30078125,
      "rewards/rejected": -0.703125,
      "step": 460
    },
    {
      "epoch": 0.9832635983263598,
      "grad_norm": 1.5760387182235718,
      "learning_rate": 7.65944160348142e-10,
      "logits/chosen": -1.2109375,
      "logits/rejected": -1.140625,
      "logps/chosen": -378.0,
      "logps/rejected": -374.0,
      "loss": 0.6114,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.3515625,
      "rewards/margins": 0.330078125,
      "rewards/rejected": -0.6796875,
      "step": 470
    }
  ],
  "logging_steps": 10,
  "max_steps": 478,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "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": 16,
  "trial_name": null,
  "trial_params": null
}