{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 468,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.0638297872340426e-07,
      "logits/chosen": 0.1359557956457138,
      "logits/rejected": 0.030706744641065598,
      "logps/chosen": -736.0869140625,
      "logps/rejected": -613.6344604492188,
      "loss": 2.0331,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0638297872340427e-06,
      "logits/chosen": 0.11680144816637039,
      "logits/rejected": 0.26637858152389526,
      "logps/chosen": -546.4848022460938,
      "logps/rejected": -597.5820922851562,
      "loss": 2.1606,
      "rewards/accuracies": 0.4027777910232544,
      "rewards/chosen": 0.00018841998826246709,
      "rewards/margins": 0.0001322307507507503,
      "rewards/rejected": 5.6189212045865133e-05,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.1276595744680853e-06,
      "logits/chosen": 0.16349577903747559,
      "logits/rejected": 0.2681581676006317,
      "logps/chosen": -604.7147827148438,
      "logps/rejected": -649.5513916015625,
      "loss": 2.0961,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": -0.0001730217190925032,
      "rewards/margins": -4.76564237033017e-05,
      "rewards/rejected": -0.0001253652444574982,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.191489361702128e-06,
      "logits/chosen": 0.1495944708585739,
      "logits/rejected": 0.19113227725028992,
      "logps/chosen": -594.8287353515625,
      "logps/rejected": -588.0198974609375,
      "loss": 2.124,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": -0.000811461650300771,
      "rewards/margins": -7.172574896685546e-06,
      "rewards/rejected": -0.0008042889530770481,
      "step": 30
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.255319148936171e-06,
      "logits/chosen": 0.16839194297790527,
      "logits/rejected": 0.23641005158424377,
      "logps/chosen": -574.6544799804688,
      "logps/rejected": -631.8447875976562,
      "loss": 2.1986,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.0023893495090305805,
      "rewards/margins": 0.001175779732875526,
      "rewards/rejected": -0.0035651293583214283,
      "step": 40
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.999373573764188e-06,
      "logits/chosen": 0.14030753076076508,
      "logits/rejected": 0.22485153377056122,
      "logps/chosen": -613.718505859375,
      "logps/rejected": -637.7477416992188,
      "loss": 2.1858,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.01074314396828413,
      "rewards/margins": 0.004234404768794775,
      "rewards/rejected": -0.014977550134062767,
      "step": 50
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.988245838331339e-06,
      "logits/chosen": 0.16629436612129211,
      "logits/rejected": 0.1657789647579193,
      "logps/chosen": -639.8953857421875,
      "logps/rejected": -674.060302734375,
      "loss": 2.1366,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.03131238371133804,
      "rewards/margins": 0.011484434828162193,
      "rewards/rejected": -0.04279681295156479,
      "step": 60
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.963268819535228e-06,
      "logits/chosen": 0.10423990339040756,
      "logits/rejected": 0.11754454672336578,
      "logps/chosen": -634.0193481445312,
      "logps/rejected": -738.2115478515625,
      "loss": 2.1147,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.06819052994251251,
      "rewards/margins": 0.03033190593123436,
      "rewards/rejected": -0.09852243959903717,
      "step": 70
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.9245815365216115e-06,
      "logits/chosen": 0.13063886761665344,
      "logits/rejected": 0.18018567562103271,
      "logps/chosen": -764.638671875,
      "logps/rejected": -714.2967529296875,
      "loss": 2.1622,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.12814757227897644,
      "rewards/margins": 0.030911359935998917,
      "rewards/rejected": -0.15905895829200745,
      "step": 80
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.872399318152594e-06,
      "logits/chosen": 0.032799430191516876,
      "logits/rejected": 0.08405763655900955,
      "logps/chosen": -795.3159790039062,
      "logps/rejected": -893.0309448242188,
      "loss": 2.056,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.20986966788768768,
      "rewards/margins": 0.08644559979438782,
      "rewards/rejected": -0.2963152527809143,
      "step": 90
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.807012604511542e-06,
      "logits/chosen": 0.07495725154876709,
      "logits/rejected": 0.16169913113117218,
      "logps/chosen": -909.8048706054688,
      "logps/rejected": -995.3268432617188,
      "loss": 2.0738,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.2920338809490204,
      "rewards/margins": 0.09413494169712067,
      "rewards/rejected": -0.38616880774497986,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.728785330347771e-06,
      "logits/chosen": 0.16292598843574524,
      "logits/rejected": 0.1960914433002472,
      "logps/chosen": -904.22509765625,
      "logps/rejected": -912.8228759765625,
      "loss": 1.9508,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.2555181682109833,
      "rewards/margins": 0.05460430309176445,
      "rewards/rejected": -0.3101224899291992,
      "step": 110
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.63815289945858e-06,
      "logits/chosen": 0.10713698714971542,
      "logits/rejected": 0.20941510796546936,
      "logps/chosen": -819.4172973632812,
      "logps/rejected": -1036.814208984375,
      "loss": 1.9156,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.26556941866874695,
      "rewards/margins": 0.15225186944007874,
      "rewards/rejected": -0.41782131791114807,
      "step": 120
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.535619761282989e-06,
      "logits/chosen": 0.14809398353099823,
      "logits/rejected": 0.18617282807826996,
      "logps/chosen": -886.8990478515625,
      "logps/rejected": -1030.2890625,
      "loss": 1.9636,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.32516202330589294,
      "rewards/margins": 0.13631486892700195,
      "rewards/rejected": -0.4614768922328949,
      "step": 130
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.42175660319555e-06,
      "logits/chosen": 0.18207255005836487,
      "logits/rejected": 0.18847420811653137,
      "logps/chosen": -927.9578247070312,
      "logps/rejected": -1096.78955078125,
      "loss": 1.8634,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3230741322040558,
      "rewards/margins": 0.20067179203033447,
      "rewards/rejected": -0.5237459540367126,
      "step": 140
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297197174127619e-06,
      "logits/chosen": 0.17518748342990875,
      "logits/rejected": 0.22532868385314941,
      "logps/chosen": -905.5735473632812,
      "logps/rejected": -1151.804443359375,
      "loss": 1.8103,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3360939621925354,
      "rewards/margins": 0.21287350356578827,
      "rewards/rejected": -0.5489674210548401,
      "step": 150
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.162634757195418e-06,
      "logits/chosen": 0.2002106010913849,
      "logits/rejected": 0.19682054221630096,
      "logps/chosen": -903.08935546875,
      "logps/rejected": -1042.5498046875,
      "loss": 1.9037,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.3114651143550873,
      "rewards/margins": 0.1639038622379303,
      "rewards/rejected": -0.4753689765930176,
      "step": 160
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.018818310967843e-06,
      "logits/chosen": 0.20454064011573792,
      "logits/rejected": 0.2340986430644989,
      "logps/chosen": -755.2179565429688,
      "logps/rejected": -950.396484375,
      "loss": 1.9158,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.21088416874408722,
      "rewards/margins": 0.1926594078540802,
      "rewards/rejected": -0.4035435616970062,
      "step": 170
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.866548300851254e-06,
      "logits/chosen": 0.17594218254089355,
      "logits/rejected": 0.19749297201633453,
      "logps/chosen": -861.9845581054688,
      "logps/rejected": -1095.2757568359375,
      "loss": 1.8464,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.26217204332351685,
      "rewards/margins": 0.22928395867347717,
      "rewards/rejected": -0.49145597219467163,
      "step": 180
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.706672243793271e-06,
      "logits/chosen": 0.2364441156387329,
      "logits/rejected": 0.29772695899009705,
      "logps/chosen": -868.7154541015625,
      "logps/rejected": -1106.079345703125,
      "loss": 1.8101,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.27426981925964355,
      "rewards/margins": 0.2382505238056183,
      "rewards/rejected": -0.5125203132629395,
      "step": 190
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.5400799911032357e-06,
      "logits/chosen": 0.22934266924858093,
      "logits/rejected": 0.2707999646663666,
      "logps/chosen": -949.2401123046875,
      "logps/rejected": -1226.9296875,
      "loss": 1.7955,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.3255212903022766,
      "rewards/margins": 0.26814141869544983,
      "rewards/rejected": -0.5936626195907593,
      "step": 200
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3676987756445894e-06,
      "logits/chosen": 0.20278143882751465,
      "logits/rejected": 0.2634762227535248,
      "logps/chosen": -825.6008911132812,
      "logps/rejected": -1062.2701416015625,
      "loss": 1.8656,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.2474614679813385,
      "rewards/margins": 0.25558072328567505,
      "rewards/rejected": -0.5030421614646912,
      "step": 210
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1904880509659397e-06,
      "logits/chosen": 0.2307882010936737,
      "logits/rejected": 0.2746850252151489,
      "logps/chosen": -902.4068603515625,
      "logps/rejected": -1159.0933837890625,
      "loss": 1.7356,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.28473055362701416,
      "rewards/margins": 0.2488754689693451,
      "rewards/rejected": -0.5336060523986816,
      "step": 220
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0094341510955697e-06,
      "logits/chosen": 0.15470555424690247,
      "logits/rejected": 0.24446210265159607,
      "logps/chosen": -932.212890625,
      "logps/rejected": -1326.384033203125,
      "loss": 1.8039,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.2982610762119293,
      "rewards/margins": 0.38883471488952637,
      "rewards/rejected": -0.6870958209037781,
      "step": 230
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.825544800722376e-06,
      "logits/chosen": 0.1649903804063797,
      "logits/rejected": 0.2850923240184784,
      "logps/chosen": -914.9806518554688,
      "logps/rejected": -1192.943359375,
      "loss": 1.8769,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.33413922786712646,
      "rewards/margins": 0.2569636404514313,
      "rewards/rejected": -0.5911028981208801,
      "step": 240
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.639843506318899e-06,
      "logits/chosen": 0.2421075850725174,
      "logits/rejected": 0.23027054965496063,
      "logps/chosen": -815.9312744140625,
      "logps/rejected": -1075.3148193359375,
      "loss": 1.9302,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.2732352614402771,
      "rewards/margins": 0.20585504174232483,
      "rewards/rejected": -0.47909030318260193,
      "step": 250
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4533638594248094e-06,
      "logits/chosen": 0.23280009627342224,
      "logits/rejected": 0.27932989597320557,
      "logps/chosen": -820.83447265625,
      "logps/rejected": -1066.0859375,
      "loss": 1.8994,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.24443455040454865,
      "rewards/margins": 0.23904922604560852,
      "rewards/rejected": -0.4834837317466736,
      "step": 260
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2671437837980943e-06,
      "logits/chosen": 0.20527169108390808,
      "logits/rejected": 0.1906355619430542,
      "logps/chosen": -802.508544921875,
      "logps/rejected": -1058.9862060546875,
      "loss": 1.7706,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.23849380016326904,
      "rewards/margins": 0.23252327740192413,
      "rewards/rejected": -0.47101712226867676,
      "step": 270
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.082219758453629e-06,
      "logits/chosen": 0.18224522471427917,
      "logits/rejected": 0.23733553290367126,
      "logps/chosen": -854.2337646484375,
      "logps/rejected": -1095.759765625,
      "loss": 1.7181,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.26597216725349426,
      "rewards/margins": 0.22445416450500488,
      "rewards/rejected": -0.4904263913631439,
      "step": 280
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.899621048743019e-06,
      "logits/chosen": 0.1843213140964508,
      "logits/rejected": 0.2989790141582489,
      "logps/chosen": -902.1618041992188,
      "logps/rejected": -1187.279541015625,
      "loss": 1.7856,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.3390468955039978,
      "rewards/margins": 0.2693897485733032,
      "rewards/rejected": -0.608436644077301,
      "step": 290
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7203639775848423e-06,
      "logits/chosen": 0.15538965165615082,
      "logits/rejected": 0.25810253620147705,
      "logps/chosen": -898.4368896484375,
      "logps/rejected": -1099.8011474609375,
      "loss": 1.8812,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.32740405201911926,
      "rewards/margins": 0.21446183323860168,
      "rewards/rejected": -0.5418658256530762,
      "step": 300
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5454462687309445e-06,
      "logits/chosen": 0.17961570620536804,
      "logits/rejected": 0.22687454521656036,
      "logps/chosen": -870.6505737304688,
      "logps/rejected": -1096.16015625,
      "loss": 1.8067,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.2986447811126709,
      "rewards/margins": 0.21215729415416718,
      "rewards/rejected": -0.5108019709587097,
      "step": 310
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.3758414935535147e-06,
      "logits/chosen": 0.18137860298156738,
      "logits/rejected": 0.24465851485729218,
      "logps/chosen": -902.7379150390625,
      "logps/rejected": -1196.3099365234375,
      "loss": 1.68,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.3003231883049011,
      "rewards/margins": 0.2799701988697052,
      "rewards/rejected": -0.5802933573722839,
      "step": 320
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2124936522614622e-06,
      "logits/chosen": 0.18053703010082245,
      "logits/rejected": 0.19177524745464325,
      "logps/chosen": -886.50341796875,
      "logps/rejected": -1139.5965576171875,
      "loss": 1.7893,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.3074926435947418,
      "rewards/margins": 0.2617393434047699,
      "rewards/rejected": -0.5692320466041565,
      "step": 330
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0563119197063934e-06,
      "logits/chosen": 0.20963367819786072,
      "logits/rejected": 0.23218393325805664,
      "logps/chosen": -865.2020263671875,
      "logps/rejected": -1143.264404296875,
      "loss": 1.7222,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.2907507121562958,
      "rewards/margins": 0.2580210566520691,
      "rewards/rejected": -0.5487717390060425,
      "step": 340
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.081655850224449e-07,
      "logits/chosen": 0.16453954577445984,
      "logits/rejected": 0.18646371364593506,
      "logps/chosen": -938.1325073242188,
      "logps/rejected": -1226.00634765625,
      "loss": 1.7215,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.3565717339515686,
      "rewards/margins": 0.2777550220489502,
      "rewards/rejected": -0.6343267560005188,
      "step": 350
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.688792132653111e-07,
      "logits/chosen": 0.14916151762008667,
      "logits/rejected": 0.23365116119384766,
      "logps/chosen": -938.5538940429688,
      "logps/rejected": -1271.653076171875,
      "loss": 1.7912,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.3168650269508362,
      "rewards/margins": 0.3130674362182617,
      "rewards/rejected": -0.6299324035644531,
      "step": 360
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.392280559802341e-07,
      "logits/chosen": 0.20630326867103577,
      "logits/rejected": 0.19222524762153625,
      "logps/chosen": -968.6492309570312,
      "logps/rejected": -1210.966552734375,
      "loss": 1.7648,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.35769984126091003,
      "rewards/margins": 0.22969777882099152,
      "rewards/rejected": -0.5873976945877075,
      "step": 370
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.199337362431792e-07,
      "logits/chosen": 0.23185975849628448,
      "logits/rejected": 0.12383759021759033,
      "logps/chosen": -913.8956298828125,
      "logps/rejected": -1156.3277587890625,
      "loss": 1.764,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.3383910357952118,
      "rewards/margins": 0.2371567189693451,
      "rewards/rejected": -0.5755477547645569,
      "step": 380
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1166023219176176e-07,
      "logits/chosen": 0.17365404963493347,
      "logits/rejected": 0.22495949268341064,
      "logps/chosen": -965.8720703125,
      "logps/rejected": -1141.7669677734375,
      "loss": 1.7503,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.3551119267940521,
      "rewards/margins": 0.21352621912956238,
      "rewards/rejected": -0.5686381459236145,
      "step": 390
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.150101814011136e-07,
      "logits/chosen": 0.12874256074428558,
      "logits/rejected": 0.17315654456615448,
      "logps/chosen": -858.0895385742188,
      "logps/rejected": -1191.4886474609375,
      "loss": 1.7221,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.2957189381122589,
      "rewards/margins": 0.26001372933387756,
      "rewards/rejected": -0.5557327270507812,
      "step": 400
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.3052152667409289e-07,
      "logits/chosen": 0.16493479907512665,
      "logits/rejected": 0.17930065095424652,
      "logps/chosen": -871.76025390625,
      "logps/rejected": -1177.62353515625,
      "loss": 1.8162,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.28667283058166504,
      "rewards/margins": 0.30985671281814575,
      "rewards/rejected": -0.5965295433998108,
      "step": 410
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5866452191498488e-07,
      "logits/chosen": 0.165872260928154,
      "logits/rejected": 0.2134593427181244,
      "logps/chosen": -985.79248046875,
      "logps/rejected": -1185.5306396484375,
      "loss": 1.8081,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.38030657172203064,
      "rewards/margins": 0.19018930196762085,
      "rewards/rejected": -0.5704958438873291,
      "step": 420
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.983911475163727e-08,
      "logits/chosen": 0.1297980695962906,
      "logits/rejected": 0.2198360413312912,
      "logps/chosen": -843.9484252929688,
      "logps/rejected": -1071.7698974609375,
      "loss": 1.816,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.28988003730773926,
      "rewards/margins": 0.22120681405067444,
      "rewards/rejected": -0.5110868811607361,
      "step": 430
    },
    {
      "epoch": 0.94,
      "learning_rate": 5.437272047405712e-08,
      "logits/chosen": 0.14703018963336945,
      "logits/rejected": 0.26481714844703674,
      "logps/chosen": -841.8990478515625,
      "logps/rejected": -1089.636474609375,
      "loss": 1.7983,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.32132893800735474,
      "rewards/margins": 0.21181070804595947,
      "rewards/rejected": -0.5331395864486694,
      "step": 440
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.251839967945535e-08,
      "logits/chosen": 0.1110246405005455,
      "logits/rejected": 0.1886615753173828,
      "logps/chosen": -919.92333984375,
      "logps/rejected": -1245.7840576171875,
      "loss": 1.6896,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.30925479531288147,
      "rewards/margins": 0.329929918050766,
      "rewards/rejected": -0.6391846537590027,
      "step": 450
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.453449766758933e-09,
      "logits/chosen": 0.13569632172584534,
      "logits/rejected": 0.18673357367515564,
      "logps/chosen": -879.3952026367188,
      "logps/rejected": -1108.7191162109375,
      "loss": 1.7726,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.35073980689048767,
      "rewards/margins": 0.19130484759807587,
      "rewards/rejected": -0.5420446991920471,
      "step": 460
    },
    {
      "epoch": 1.0,
      "step": 468,
      "total_flos": 0.0,
      "train_loss": 1.4258099234002268,
      "train_runtime": 12564.9924,
      "train_samples_per_second": 2.388,
      "train_steps_per_second": 0.037
    }
  ],
  "logging_steps": 10,
  "max_steps": 468,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}