{ "best_metric": null, "best_model_checkpoint": null, "epoch": 5.7006369426751595, "eval_steps": 500, "global_step": 450, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "completion_length": 247.125, "epoch": 0.050955414012738856, "grad_norm": 5.409891128540039, "kl": 2.317987572401762, "learning_rate": 8.57142857142857e-08, "loss": 0.0023, "reward": 17.20423835515976, "reward_std": 1.1507948003709316, "rewards/concensus_correctness_reward_func": 12.706375002861023, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.7556760907173157, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.023437537252903, "step": 4 }, { "completion_length": 228.625, "epoch": 0.10191082802547771, "grad_norm": 4.3909173011779785, "kl": 1.6623684950172901, "learning_rate": 1.714285714285714e-07, "loss": 0.0017, "reward": 12.55974566936493, "reward_std": 6.854554038494825, "rewards/concensus_correctness_reward_func": 8.75, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5599956843070686, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.1247500106692314, "step": 8 }, { "completion_length": 194.375, "epoch": 0.15286624203821655, "grad_norm": 2.985093832015991, "kl": 6.860304161906242, "learning_rate": 2.571428571428571e-07, "loss": 0.0069, "reward": 16.054461300373077, "reward_std": 9.667749413754791, "rewards/concensus_correctness_reward_func": 11.670499965548515, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.6383362491615117, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.2143750190734863, "step": 12 }, { "completion_length": 269.75, "epoch": 0.20382165605095542, "grad_norm": 3.6396210193634033, "kl": 5.6482887379825115, "learning_rate": 2.99984424553496e-07, "loss": 0.0056, "reward": 12.981057226657867, "reward_std": 6.896631011739373, "rewards/concensus_correctness_reward_func": 9.086749993264675, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.597869697958231, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.15625, "rewards/xmlcount_reward_func": 0.8901875242590904, "step": 16 }, { "completion_length": 213.6875, "epoch": 0.25477707006369427, "grad_norm": 2.7366411685943604, "kl": 8.909522596746683, "learning_rate": 2.998598403883547e-07, "loss": 0.0089, "reward": 11.70977608859539, "reward_std": 4.304964765789919, "rewards/concensus_correctness_reward_func": 7.7207499742507935, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.46571373101323843, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.179562509059906, "step": 20 }, { "completion_length": 250.625, "epoch": 0.3057324840764331, "grad_norm": 4.964949607849121, "kl": 5.63778755068779, "learning_rate": 2.9961077554357913e-07, "loss": 0.0056, "reward": 10.035876393318176, "reward_std": 6.212635695934296, "rewards/concensus_correctness_reward_func": 6.3186250030994415, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.7127514779567719, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.0670000314712524, "step": 24 }, { "completion_length": 279.0, "epoch": 0.35668789808917195, "grad_norm": 2.4961581230163574, "kl": 67.01326366513968, "learning_rate": 2.9923743690422306e-07, "loss": 0.067, "reward": 16.109780073165894, "reward_std": 5.918968662619591, "rewards/concensus_correctness_reward_func": 11.880625024437904, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.6405299380421638, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.1198750287294388, "step": 28 }, { "completion_length": 273.6875, "epoch": 0.40764331210191085, "grad_norm": 5.211795330047607, "kl": 5.373722583055496, "learning_rate": 2.9874013458303877e-07, "loss": 0.0054, "reward": 17.488071978092194, "reward_std": 4.414429775904864, "rewards/concensus_correctness_reward_func": 12.954499989748001, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.5903223338536918, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.15625, "rewards/xmlcount_reward_func": 1.162000022828579, "step": 32 }, { "completion_length": 241.6875, "epoch": 0.4585987261146497, "grad_norm": 1.7989726066589355, "kl": 5.3555001355707645, "learning_rate": 2.9811928166288275e-07, "loss": 0.0053, "reward": 15.544344544410706, "reward_std": 2.0546061204513535, "rewards/concensus_correctness_reward_func": 11.492375005036592, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.5364694236777723, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.1717500165104866, "step": 36 }, { "completion_length": 250.4375, "epoch": 0.5095541401273885, "grad_norm": 55.185237884521484, "kl": 47.606556087732315, "learning_rate": 2.973753938535892e-07, "loss": 0.0476, "reward": 12.839633375406265, "reward_std": 2.927978876978159, "rewards/concensus_correctness_reward_func": 8.834124997258186, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.5, "rewards/question_recreation_reward_func": 0.6650712452828884, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.1529375314712524, "step": 40 }, { "completion_length": 263.125, "epoch": 0.5605095541401274, "grad_norm": 3.3247878551483154, "kl": 9.563120489940047, "learning_rate": 2.9650908906359733e-07, "loss": 0.0096, "reward": 17.1003201007843, "reward_std": 4.631968731526285, "rewards/concensus_correctness_reward_func": 12.573749996721745, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.6720699109137058, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.104500025510788, "step": 44 }, { "completion_length": 290.75, "epoch": 0.6114649681528662, "grad_norm": 3.066596508026123, "kl": 12.11058185622096, "learning_rate": 2.955210868866872e-07, "loss": 0.0121, "reward": 19.856602907180786, "reward_std": 8.727525525609963, "rewards/concensus_correctness_reward_func": 15.0, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.625, "rewards/question_recreation_reward_func": 0.700602900236845, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 0.8122500255703926, "step": 48 }, { "completion_length": 229.4375, "epoch": 0.6624203821656051, "grad_norm": 2.433725118637085, "kl": 1.3885855339467525, "learning_rate": 2.944122080042512e-07, "loss": 0.0014, "reward": 14.267784029245377, "reward_std": 5.394140258431435, "rewards/concensus_correctness_reward_func": 10.306124992668629, "rewards/consensus_reward_func": 0.75, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.6182841658592224, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.46875, "rewards/xmlcount_reward_func": 1.2496250122785568, "step": 52 }, { "completion_length": 265.3125, "epoch": 0.7133757961783439, "grad_norm": 1.840900182723999, "kl": 6.538037680089474, "learning_rate": 2.93183373503597e-07, "loss": 0.0065, "reward": 16.03761088848114, "reward_std": 10.041991762816906, "rewards/concensus_correctness_reward_func": 11.560124978423119, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.6827985420823097, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1071875169873238, "step": 56 }, { "completion_length": 211.0, "epoch": 0.7643312101910829, "grad_norm": 1.3512978553771973, "kl": 7.67945366166532, "learning_rate": 2.91835604112849e-07, "loss": 0.0077, "reward": 19.18588936328888, "reward_std": 6.0852950513362885, "rewards/concensus_correctness_reward_func": 14.20750005543232, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6377642604056746, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.2468750029802322, "step": 60 }, { "completion_length": 209.0625, "epoch": 0.8152866242038217, "grad_norm": 2.5626749992370605, "kl": 2.9191525876522064, "learning_rate": 2.9037001935308313e-07, "loss": 0.0029, "reward": 14.56528091430664, "reward_std": 4.982788998633623, "rewards/concensus_correctness_reward_func": 10.215499997138977, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.7710309810936451, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.2037499845027924, "step": 64 }, { "completion_length": 215.3125, "epoch": 0.8662420382165605, "grad_norm": 3.383816719055176, "kl": 3.1076866053044796, "learning_rate": 2.8878783660839964e-07, "loss": 0.0031, "reward": 20.26084053516388, "reward_std": 4.580408729612827, "rewards/concensus_correctness_reward_func": 15.626375034451485, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.41534046828746796, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.1253750175237656, "step": 68 }, { "completion_length": 245.3125, "epoch": 0.9171974522292994, "grad_norm": 11.53940200805664, "kl": 36.37679213285446, "learning_rate": 2.8709037011470624e-07, "loss": 0.0364, "reward": 12.166111320257187, "reward_std": 7.469847652129829, "rewards/concensus_correctness_reward_func": 8.189249992370605, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6527361150365323, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.0428750179708004, "step": 72 }, { "completion_length": 252.6875, "epoch": 0.9681528662420382, "grad_norm": 1.9474271535873413, "kl": 5.756619825959206, "learning_rate": 2.852790298680516e-07, "loss": 0.0058, "reward": 18.147495985031128, "reward_std": 5.955903281457722, "rewards/concensus_correctness_reward_func": 13.75, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5002461355179548, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.084750011563301, "step": 76 }, { "completion_length": 231.14285714285714, "epoch": 1.0127388535031847, "grad_norm": 2.1662113666534424, "kl": 4.730595575911658, "learning_rate": 2.833553204534156e-07, "loss": 0.0041, "reward": 19.707179682595388, "reward_std": 4.576129041612148, "rewards/concensus_correctness_reward_func": 14.77628571646554, "rewards/consensus_reward_func": 1.5714285714285714, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.1428571428571428, "rewards/question_recreation_reward_func": 0.6548936026436942, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.32142857142857145, "rewards/xmlcount_reward_func": 1.2402857201439994, "step": 80 }, { "completion_length": 181.9375, "epoch": 1.0636942675159236, "grad_norm": 3.710115909576416, "kl": 7.103449460119009, "learning_rate": 2.8132083979493016e-07, "loss": 0.0071, "reward": 15.854183077812195, "reward_std": 8.889904107898474, "rewards/concensus_correctness_reward_func": 11.673750042915344, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5401829741895199, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.40625, "rewards/xmlcount_reward_func": 1.234000027179718, "step": 84 }, { "completion_length": 250.4375, "epoch": 1.1146496815286624, "grad_norm": 4.344290256500244, "kl": 3.607717391103506, "learning_rate": 2.7917727782856764e-07, "loss": 0.0036, "reward": 13.101991266012192, "reward_std": 6.616666306043044, "rewards/concensus_correctness_reward_func": 8.970250010490417, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.5953665748238564, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.1926250606775284, "step": 88 }, { "completion_length": 224.5625, "epoch": 1.1656050955414012, "grad_norm": 1.8617960214614868, "kl": 2.941370815038681, "learning_rate": 2.769264150983997e-07, "loss": 0.0029, "reward": 18.346871256828308, "reward_std": 3.0000894721597433, "rewards/concensus_correctness_reward_func": 13.75, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.7451213523745537, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.1017500162124634, "step": 92 }, { "completion_length": 275.3125, "epoch": 1.21656050955414, "grad_norm": 2.1597092151641846, "kl": 7.501682840287685, "learning_rate": 2.745701212775937e-07, "loss": 0.0075, "reward": 15.922978341579437, "reward_std": 10.099833406507969, "rewards/concensus_correctness_reward_func": 11.25, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6603539576753974, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.1063750311732292, "step": 96 }, { "completion_length": 232.375, "epoch": 1.267515923566879, "grad_norm": 2.969134569168091, "kl": 7.7300273440778255, "learning_rate": 2.7211035361537307e-07, "loss": 0.0077, "reward": 11.717751204967499, "reward_std": 7.577772298827767, "rewards/concensus_correctness_reward_func": 7.94712507724762, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.5, "rewards/question_recreation_reward_func": 0.5368763818405569, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.1400000303983688, "step": 100 }, { "completion_length": 226.0625, "epoch": 1.3184713375796178, "grad_norm": 1.5485366582870483, "kl": 2.664130464196205, "learning_rate": 2.695491553112336e-07, "loss": 0.0027, "reward": 20.976629972457886, "reward_std": 6.513700008392334, "rewards/concensus_correctness_reward_func": 16.460624933242798, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.4631927441805601, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.1778125315904617, "step": 104 }, { "completion_length": 202.1875, "epoch": 1.3694267515923566, "grad_norm": 5.035473346710205, "kl": 19.53648843243718, "learning_rate": 2.6688865381776554e-07, "loss": 0.0195, "reward": 16.44457733631134, "reward_std": 6.439986160025001, "rewards/concensus_correctness_reward_func": 11.878250032663345, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.6380774034187198, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.40625, "rewards/xmlcount_reward_func": 1.1470000073313713, "step": 108 }, { "completion_length": 238.375, "epoch": 1.4203821656050954, "grad_norm": 2.5546350479125977, "kl": 4.930681709200144, "learning_rate": 2.6413105907349006e-07, "loss": 0.0049, "reward": 19.794534921646118, "reward_std": 4.685703623574227, "rewards/concensus_correctness_reward_func": 15.0, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.7159727662801743, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.1723125129938126, "step": 112 }, { "completion_length": 213.0625, "epoch": 1.4713375796178343, "grad_norm": 2.9883830547332764, "kl": 3.1261193864047527, "learning_rate": 2.612786616671803e-07, "loss": 0.0031, "reward": 13.908211499452591, "reward_std": 4.823537867516279, "rewards/concensus_correctness_reward_func": 10.0, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.689586691558361, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1561250165104866, "step": 116 }, { "completion_length": 229.25, "epoch": 1.5222929936305731, "grad_norm": 2.6060760021209717, "kl": 93.09971876814961, "learning_rate": 2.583338309351894e-07, "loss": 0.0931, "reward": 14.722533643245697, "reward_std": 3.7422131042694673, "rewards/concensus_correctness_reward_func": 10.464500008150935, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.5823462381958961, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.14443751424551, "step": 120 }, { "completion_length": 241.9375, "epoch": 1.573248407643312, "grad_norm": 3.838078260421753, "kl": 6.129395470023155, "learning_rate": 2.55299012993368e-07, "loss": 0.0061, "reward": 18.527520298957825, "reward_std": 6.345299224369228, "rewards/concensus_correctness_reward_func": 14.059624962508678, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.5597703522071242, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 0.8456250205636024, "step": 124 }, { "completion_length": 228.1875, "epoch": 1.6242038216560508, "grad_norm": 2.3569176197052, "kl": 190.0281139537692, "learning_rate": 2.5217672870520407e-07, "loss": 0.19, "reward": 19.29222685098648, "reward_std": 0.9815351580036804, "rewards/concensus_correctness_reward_func": 15.0, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5401642136275768, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.0958125218749046, "step": 128 }, { "completion_length": 216.75, "epoch": 1.6751592356687897, "grad_norm": 1.5537621974945068, "kl": 2.767712377011776, "learning_rate": 2.489695715878747e-07, "loss": 0.0028, "reward": 18.718332588672638, "reward_std": 6.182364942971617, "rewards/concensus_correctness_reward_func": 13.75, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.625, "rewards/question_recreation_reward_func": 0.5509578062337823, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.1673750281333923, "step": 132 }, { "completion_length": 257.9375, "epoch": 1.7261146496815285, "grad_norm": 3.8648273944854736, "kl": 2.600251194089651, "learning_rate": 2.4568020565794784e-07, "loss": 0.0026, "reward": 10.30246677994728, "reward_std": 6.988724898546934, "rewards/concensus_correctness_reward_func": 6.4593749940395355, "rewards/consensus_reward_func": 0.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6528416611254215, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.0965000167489052, "step": 136 }, { "completion_length": 242.6875, "epoch": 1.7770700636942676, "grad_norm": 54.2701530456543, "kl": 31.665705926716328, "learning_rate": 2.4231136321852394e-07, "loss": 0.0317, "reward": 20.147151827812195, "reward_std": 7.880714102386264, "rewards/concensus_correctness_reward_func": 15.221750020980835, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6395266838371754, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.0983750149607658, "step": 140 }, { "completion_length": 229.6875, "epoch": 1.8280254777070064, "grad_norm": 2.66843581199646, "kl": 3.5122765973210335, "learning_rate": 2.3886584258965555e-07, "loss": 0.0035, "reward": 15.746762156486511, "reward_std": 5.882234873250127, "rewards/concensus_correctness_reward_func": 11.25, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.5443250723183155, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.1711875051259995, "step": 144 }, { "completion_length": 270.125, "epoch": 1.8789808917197452, "grad_norm": 1.622960090637207, "kl": 7.444879077374935, "learning_rate": 2.3534650578393037e-07, "loss": 0.0074, "reward": 13.791082382202148, "reward_std": 6.4896623231470585, "rewards/concensus_correctness_reward_func": 9.334750011563301, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.698457557708025, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.03912503272295, "step": 148 }, { "completion_length": 199.625, "epoch": 1.929936305732484, "grad_norm": 2.772538900375366, "kl": 1.7244556173682213, "learning_rate": 2.3175627612914788e-07, "loss": 0.0017, "reward": 18.972005277872086, "reward_std": 2.4147312035784125, "rewards/concensus_correctness_reward_func": 13.971750006079674, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.5098803527653217, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.40625, "rewards/xmlcount_reward_func": 1.2091249972581863, "step": 152 }, { "completion_length": 244.0625, "epoch": 1.980891719745223, "grad_norm": 2.382359504699707, "kl": 6.68058829754591, "learning_rate": 2.2809813584006504e-07, "loss": 0.0067, "reward": 18.733210682868958, "reward_std": 5.422195242252201, "rewards/concensus_correctness_reward_func": 14.069375038146973, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.5, "rewards/question_recreation_reward_func": 0.4836479816585779, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.0864375233650208, "step": 156 }, { "completion_length": 189.71428571428572, "epoch": 2.0254777070063694, "grad_norm": 4.063220977783203, "kl": 2.7896384043352946, "learning_rate": 2.243751235412272e-07, "loss": 0.0024, "reward": 19.006362915039062, "reward_std": 5.089762266193118, "rewards/concensus_correctness_reward_func": 14.781285745756966, "rewards/consensus_reward_func": 1.5714285714285714, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.8571428571428571, "rewards/question_recreation_reward_func": 0.515291694019522, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.2857142857142857, "rewards/xmlcount_reward_func": 0.9955000025885445, "step": 160 }, { "completion_length": 213.4375, "epoch": 2.0764331210191083, "grad_norm": 45.27822494506836, "kl": 18.627060689032078, "learning_rate": 2.2059033174294308e-07, "loss": 0.0186, "reward": 17.683844089508057, "reward_std": 10.830251511186361, "rewards/concensus_correctness_reward_func": 13.190500020980835, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.5491568585857749, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.162937492132187, "step": 164 }, { "completion_length": 233.25, "epoch": 2.127388535031847, "grad_norm": 10.763078689575195, "kl": 8.058194138109684, "learning_rate": 2.167469042724989e-07, "loss": 0.0081, "reward": 16.62453094124794, "reward_std": 2.8090486722067, "rewards/concensus_correctness_reward_func": 11.670500010251999, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.5, "rewards/question_recreation_reward_func": 0.6700308993458748, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.065250001847744, "step": 168 }, { "completion_length": 239.9375, "epoch": 2.178343949044586, "grad_norm": 2.4544873237609863, "kl": 4.6812648214399815, "learning_rate": 2.1284803366274665e-07, "loss": 0.0047, "reward": 10.69625610113144, "reward_std": 2.8211628943681717, "rewards/concensus_correctness_reward_func": 6.55375000089407, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.5454436782747507, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.2220625132322311, "step": 172 }, { "completion_length": 243.1875, "epoch": 2.229299363057325, "grad_norm": 2.8351073265075684, "kl": 11.15944654121995, "learning_rate": 2.0889695850023532e-07, "loss": 0.0111, "reward": 14.30355390906334, "reward_std": 3.1945501037698705, "rewards/concensus_correctness_reward_func": 10.470749974250793, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.5688038393855095, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.0452500134706497, "step": 176 }, { "completion_length": 227.25, "epoch": 2.2802547770700636, "grad_norm": 2.5390284061431885, "kl": 4.675673048943281, "learning_rate": 2.048969607350874e-07, "loss": 0.0047, "reward": 13.208377301692963, "reward_std": 6.394375938922167, "rewards/concensus_correctness_reward_func": 9.178874969482422, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5128772370517254, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.0478749945759773, "step": 180 }, { "completion_length": 241.5, "epoch": 2.3312101910828025, "grad_norm": 339.2472839355469, "kl": 120.68484226241708, "learning_rate": 2.008513629548557e-07, "loss": 0.1207, "reward": 12.759812444448471, "reward_std": 5.974757697433233, "rewards/concensus_correctness_reward_func": 9.136750057339668, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.5173125192523003, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.0745000019669533, "step": 184 }, { "completion_length": 266.1875, "epoch": 2.3821656050955413, "grad_norm": 9.494641304016113, "kl": 6.892231393605471, "learning_rate": 1.967635256246248e-07, "loss": 0.0069, "reward": 18.39091458916664, "reward_std": 3.120848624035716, "rewards/concensus_correctness_reward_func": 13.973124995827675, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.630602179095149, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.03718750923872, "step": 188 }, { "completion_length": 256.1875, "epoch": 2.43312101910828, "grad_norm": 3.3058788776397705, "kl": 3.385168194770813, "learning_rate": 1.9263684429564973e-07, "loss": 0.0034, "reward": 15.78951108455658, "reward_std": 9.866211914690211, "rewards/concensus_correctness_reward_func": 11.25, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.7607610449194908, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.0912500247359276, "step": 192 }, { "completion_length": 266.625, "epoch": 2.484076433121019, "grad_norm": 3.944892644882202, "kl": 8.661903837695718, "learning_rate": 1.8847474678485014e-07, "loss": 0.0087, "reward": 17.91079044342041, "reward_std": 7.754661478335038, "rewards/concensus_correctness_reward_func": 12.71512496471405, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.5, "rewards/question_recreation_reward_func": 0.666103033348918, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.2170625180006027, "step": 196 }, { "completion_length": 225.8125, "epoch": 2.535031847133758, "grad_norm": 2.1859211921691895, "kl": 4.23666650429368, "learning_rate": 1.8428069032750338e-07, "loss": 0.0042, "reward": 19.735592007637024, "reward_std": 7.996539369225502, "rewards/concensus_correctness_reward_func": 15.209874987602234, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.5920920763164759, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.1523750126361847, "step": 200 }, { "completion_length": 256.375, "epoch": 2.5859872611464967, "grad_norm": 6.963754653930664, "kl": 56.80678379908204, "learning_rate": 1.8005815870550113e-07, "loss": 0.0568, "reward": 16.95860594511032, "reward_std": 7.698275171220303, "rewards/concensus_correctness_reward_func": 12.922000050544739, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5784185063093901, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 0.9894375093281269, "step": 204 }, { "completion_length": 235.6875, "epoch": 2.6369426751592355, "grad_norm": 56.92842102050781, "kl": 21.92546533420682, "learning_rate": 1.7581065935355522e-07, "loss": 0.0219, "reward": 18.0313401222229, "reward_std": 5.557751599699259, "rewards/concensus_correctness_reward_func": 13.97724997997284, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.42977757239714265, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.15625, "rewards/xmlcount_reward_func": 1.0930624902248383, "step": 208 }, { "completion_length": 239.125, "epoch": 2.6878980891719744, "grad_norm": 4.275964736938477, "kl": 2.6714395098388195, "learning_rate": 1.715417204457562e-07, "loss": 0.0027, "reward": 18.481036722660065, "reward_std": 5.910495655843988, "rewards/concensus_correctness_reward_func": 13.918125033378601, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6785994963720441, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1968125253915787, "step": 212 }, { "completion_length": 238.1875, "epoch": 2.738853503184713, "grad_norm": 2.9757699966430664, "kl": 4.989309325814247, "learning_rate": 1.6725488796490503e-07, "loss": 0.0051, "reward": 16.237927079200745, "reward_std": 5.6823064510535914, "rewards/concensus_correctness_reward_func": 11.685875058174133, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.57005206332542, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.2320000380277634, "step": 216 }, { "completion_length": 210.75, "epoch": 2.789808917197452, "grad_norm": 6.644160270690918, "kl": 28.66787153482437, "learning_rate": 1.6295372275705162e-07, "loss": 0.0287, "reward": 15.902063801884651, "reward_std": 2.8661335185170174, "rewards/concensus_correctness_reward_func": 11.25, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.5894390102475882, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 0.9688750207424164, "step": 220 }, { "completion_length": 225.875, "epoch": 2.840764331210191, "grad_norm": 2.6777331829071045, "kl": 139.06324845552444, "learning_rate": 1.5864179757368764e-07, "loss": 0.1391, "reward": 17.266418665647507, "reward_std": 7.936541106551886, "rewards/concensus_correctness_reward_func": 12.5, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.7135435082018375, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.1778749972581863, "step": 224 }, { "completion_length": 223.4375, "epoch": 2.8917197452229297, "grad_norm": 5.439139366149902, "kl": 3.8189243748784065, "learning_rate": 1.5432269410404998e-07, "loss": 0.0038, "reward": 13.274927824735641, "reward_std": 2.9763048992026597, "rewards/concensus_correctness_reward_func": 9.352000012993813, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.6428653616458178, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.15625, "rewards/xmlcount_reward_func": 1.123812511563301, "step": 228 }, { "completion_length": 247.5, "epoch": 2.9426751592356686, "grad_norm": 170.2266845703125, "kl": 167.23719503730536, "learning_rate": 1.5e-07, "loss": 0.1672, "reward": 14.749447345733643, "reward_std": 7.275530036538839, "rewards/concensus_correctness_reward_func": 10.215250015258789, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.5158222373574972, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.2058750241994858, "step": 232 }, { "completion_length": 268.0, "epoch": 2.9936305732484074, "grad_norm": 1.7273995876312256, "kl": 5.3727107755839825, "learning_rate": 1.4567730589595006e-07, "loss": 0.0054, "reward": 19.68195617198944, "reward_std": 0.35719050047919154, "rewards/concensus_correctness_reward_func": 15.112999998033047, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6851438917219639, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.0713125243782997, "step": 236 }, { "completion_length": 256.64285714285717, "epoch": 3.038216560509554, "grad_norm": 110.74983978271484, "kl": 41.563892917973654, "learning_rate": 1.4135820242631235e-07, "loss": 0.0364, "reward": 15.650528294699532, "reward_std": 4.540684355156762, "rewards/concensus_correctness_reward_func": 11.661857196262904, "rewards/consensus_reward_func": 1.5714285714285714, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.7142857142857143, "rewards/question_recreation_reward_func": 0.49124259342040333, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.17857142857142858, "rewards/xmlcount_reward_func": 1.0331428689616067, "step": 240 }, { "completion_length": 241.625, "epoch": 3.089171974522293, "grad_norm": 2.8471550941467285, "kl": 26.10317723825574, "learning_rate": 1.3704627724294842e-07, "loss": 0.0261, "reward": 15.622052431106567, "reward_std": 6.783383652102202, "rewards/concensus_correctness_reward_func": 11.25, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6173020545393229, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.0985000133514404, "step": 244 }, { "completion_length": 237.625, "epoch": 3.140127388535032, "grad_norm": 4.9774088859558105, "kl": 2.7739184722304344, "learning_rate": 1.3274511203509498e-07, "loss": 0.0028, "reward": 16.195281475782394, "reward_std": 5.6568857207894325, "rewards/concensus_correctness_reward_func": 11.760749965906143, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6031564082950354, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.2063750326633453, "step": 248 }, { "completion_length": 232.25, "epoch": 3.1910828025477707, "grad_norm": 3.6411895751953125, "kl": 8.415857467800379, "learning_rate": 1.2845827955424378e-07, "loss": 0.0084, "reward": 21.606714069843292, "reward_std": 2.034520313842222, "rewards/concensus_correctness_reward_func": 16.477874994277954, "rewards/consensus_reward_func": 1.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.5895260944962502, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.1330625265836716, "step": 252 }, { "completion_length": 219.5625, "epoch": 3.2420382165605095, "grad_norm": 1.8258823156356812, "kl": 16.24698005989194, "learning_rate": 1.241893406464448e-07, "loss": 0.0162, "reward": 18.671132683753967, "reward_std": 5.8178422772325575, "rewards/concensus_correctness_reward_func": 13.984624981880188, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6623825784772635, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.2116250097751617, "step": 256 }, { "completion_length": 207.75, "epoch": 3.2929936305732483, "grad_norm": 4.799376010894775, "kl": 5.706813037395477, "learning_rate": 1.199418412944989e-07, "loss": 0.0058, "reward": 17.392418444156647, "reward_std": 4.783364934177371, "rewards/concensus_correctness_reward_func": 12.703624963760376, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.6976060681045055, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1786875426769257, "step": 260 }, { "completion_length": 222.8125, "epoch": 3.343949044585987, "grad_norm": 2.5605955123901367, "kl": 9.123412758111954, "learning_rate": 1.1571930967249665e-07, "loss": 0.0091, "reward": 14.08166092634201, "reward_std": 6.08101740386337, "rewards/concensus_correctness_reward_func": 9.220499992370605, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.6586607862263918, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.2337500303983688, "step": 264 }, { "completion_length": 289.375, "epoch": 3.394904458598726, "grad_norm": 1.0773470401763916, "kl": 7.33378242328763, "learning_rate": 1.1152525321514992e-07, "loss": 0.0073, "reward": 15.252871230244637, "reward_std": 2.67856452986598, "rewards/concensus_correctness_reward_func": 11.25, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5149340042844415, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.050437517464161, "step": 268 }, { "completion_length": 248.8125, "epoch": 3.445859872611465, "grad_norm": 3.0849928855895996, "kl": 4.085891917347908, "learning_rate": 1.0736315570435027e-07, "loss": 0.0041, "reward": 12.67966365814209, "reward_std": 6.439146310091019, "rewards/concensus_correctness_reward_func": 9.08587496727705, "rewards/consensus_reward_func": 0.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5481634996831417, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 0.9831250309944153, "step": 272 }, { "completion_length": 250.5, "epoch": 3.4968152866242037, "grad_norm": 2.536210536956787, "kl": 2.399098828434944, "learning_rate": 1.0323647437537522e-07, "loss": 0.0024, "reward": 14.057019233703613, "reward_std": 11.72206101194024, "rewards/concensus_correctness_reward_func": 10.0, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5792695693671703, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1652500182390213, "step": 276 }, { "completion_length": 223.6875, "epoch": 3.5477707006369426, "grad_norm": 8.720197677612305, "kl": 10.37247185409069, "learning_rate": 9.914863704514432e-08, "loss": 0.0104, "reward": 18.15606963634491, "reward_std": 7.864195089787245, "rewards/concensus_correctness_reward_func": 13.134375005960464, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.5, "rewards/question_recreation_reward_func": 0.6000700099393725, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.375, "rewards/xmlcount_reward_func": 1.171625018119812, "step": 280 }, { "completion_length": 241.5, "epoch": 3.5987261146496814, "grad_norm": 2.412242889404297, "kl": 5.458444844931364, "learning_rate": 9.510303926491262e-08, "loss": 0.0055, "reward": 15.639052152633667, "reward_std": 8.874945568852127, "rewards/concensus_correctness_reward_func": 11.855250000953674, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.5041148196905851, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.1859375089406967, "step": 284 }, { "completion_length": 262.0, "epoch": 3.6496815286624202, "grad_norm": 2.1715633869171143, "kl": 3.581681564450264, "learning_rate": 9.11030414997647e-08, "loss": 0.0036, "reward": 15.71645438671112, "reward_std": 5.47196826711297, "rewards/concensus_correctness_reward_func": 11.672625064849854, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.6106421127915382, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.2144375145435333, "step": 288 }, { "completion_length": 287.875, "epoch": 3.700636942675159, "grad_norm": 1.9456199407577515, "kl": 37.41057540290058, "learning_rate": 8.715196633725341e-08, "loss": 0.0374, "reward": 15.976769387722015, "reward_std": 6.066280534490943, "rewards/concensus_correctness_reward_func": 11.459874987602234, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.7050191890448332, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.15625, "rewards/xmlcount_reward_func": 1.1556250378489494, "step": 292 }, { "completion_length": 189.875, "epoch": 3.7515923566878984, "grad_norm": 4.28585147857666, "kl": 12.30507379770279, "learning_rate": 8.325309572750114e-08, "loss": 0.0123, "reward": 18.18193393945694, "reward_std": 6.495595270884223, "rewards/concensus_correctness_reward_func": 13.848375000059605, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.6111842412501574, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1598750203847885, "step": 296 }, { "completion_length": 209.5625, "epoch": 3.802547770700637, "grad_norm": 2.9255335330963135, "kl": 2.6841611191630363, "learning_rate": 7.940966825705694e-08, "loss": 0.0028, "reward": 18.69688767194748, "reward_std": 7.003051041887375, "rewards/concensus_correctness_reward_func": 13.75, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.6851999312639236, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1991875022649765, "step": 300 }, { "completion_length": 239.4375, "epoch": 3.853503184713376, "grad_norm": 1089.72265625, "kl": 959.4794948212802, "learning_rate": 7.562487645877278e-08, "loss": 0.9595, "reward": 15.485539972782135, "reward_std": 0.4287999947555363, "rewards/concensus_correctness_reward_func": 10.538875006139278, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.7834777049720287, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.1944375187158585, "step": 304 }, { "completion_length": 183.5, "epoch": 3.904458598726115, "grad_norm": 2.360800266265869, "kl": 7.300635326653719, "learning_rate": 7.1901864159935e-08, "loss": 0.0073, "reward": 20.999223351478577, "reward_std": 5.513432171428576, "rewards/concensus_correctness_reward_func": 16.687874913215637, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.4483482171781361, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.206750027835369, "step": 308 }, { "completion_length": 219.875, "epoch": 3.9554140127388537, "grad_norm": 1.964698314666748, "kl": 1.7256745472550392, "learning_rate": 6.824372387085212e-08, "loss": 0.0017, "reward": 13.001979649066925, "reward_std": 9.3370072809048, "rewards/concensus_correctness_reward_func": 9.080874972045422, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.5092292642220855, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.224374994635582, "step": 312 }, { "completion_length": 235.14285714285714, "epoch": 4.0, "grad_norm": 0.010738754644989967, "kl": 12.37380084821156, "learning_rate": 6.465349421606967e-08, "loss": 0.0109, "reward": 16.182638917650497, "reward_std": 4.973704278768439, "rewards/concensus_correctness_reward_func": 11.428571428571429, "rewards/consensus_reward_func": 1.7142857142857142, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.7142857142857143, "rewards/question_recreation_reward_func": 0.760781603200095, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.32142857142857145, "rewards/xmlcount_reward_func": 1.2432857240949358, "step": 316 }, { "completion_length": 222.625, "epoch": 4.050955414012739, "grad_norm": 3.1885955333709717, "kl": 10.466495990753174, "learning_rate": 6.113415741034446e-08, "loss": 0.0105, "reward": 18.242493152618408, "reward_std": 2.5054425531998277, "rewards/concensus_correctness_reward_func": 13.963750004768372, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.5736186504364014, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.1738750040531158, "step": 320 }, { "completion_length": 211.25, "epoch": 4.101910828025478, "grad_norm": 2.3097496032714844, "kl": 10.439931187778711, "learning_rate": 5.768863678147609e-08, "loss": 0.0104, "reward": 15.326348930597305, "reward_std": 4.878300238400698, "rewards/concensus_correctness_reward_func": 10.434500008821487, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.7187239453196526, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.0793750137090683, "step": 324 }, { "completion_length": 242.8125, "epoch": 4.1528662420382165, "grad_norm": 2.6428279876708984, "kl": 31.37735340744257, "learning_rate": 5.4319794342052186e-08, "loss": 0.0314, "reward": 13.825124025344849, "reward_std": 9.31145294936141, "rewards/concensus_correctness_reward_func": 9.380625009536743, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.5849994292948395, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.2032500058412552, "step": 328 }, { "completion_length": 185.875, "epoch": 4.203821656050955, "grad_norm": 8.287538528442383, "kl": 4.1289902329444885, "learning_rate": 5.103042841212535e-08, "loss": 0.0041, "reward": 23.004461526870728, "reward_std": 3.9006739165633917, "rewards/concensus_correctness_reward_func": 17.7224999666214, "rewards/consensus_reward_func": 1.75, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6572116315364838, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.218500018119812, "step": 332 }, { "completion_length": 221.75, "epoch": 4.254777070063694, "grad_norm": 148.15139770507812, "kl": 78.43285623937845, "learning_rate": 4.782327129479596e-08, "loss": 0.0784, "reward": 17.33521395921707, "reward_std": 4.419136416167021, "rewards/concensus_correctness_reward_func": 12.714375019073486, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.875, "rewards/question_recreation_reward_func": 0.6673390492796898, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.1097500286996365, "step": 336 }, { "completion_length": 229.9375, "epoch": 4.305732484076433, "grad_norm": 1.7480417490005493, "kl": 28.588999539613724, "learning_rate": 4.4700987006632047e-08, "loss": 0.0286, "reward": 22.500885009765625, "reward_std": 4.617068930529058, "rewards/concensus_correctness_reward_func": 17.5, "rewards/consensus_reward_func": 1.75, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.5405731201171875, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.2415625154972076, "step": 340 }, { "completion_length": 241.875, "epoch": 4.356687898089172, "grad_norm": 2.9678404331207275, "kl": 11.135519992560148, "learning_rate": 4.166616906481058e-08, "loss": 0.0111, "reward": 22.390071988105774, "reward_std": 4.402062423527241, "rewards/concensus_correctness_reward_func": 17.70799994468689, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.6001966744661331, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.0818749964237213, "step": 344 }, { "completion_length": 237.4375, "epoch": 4.407643312101911, "grad_norm": 4.497431755065918, "kl": 3.662317145615816, "learning_rate": 3.872133833281973e-08, "loss": 0.0037, "reward": 18.673893690109253, "reward_std": 5.760366526897997, "rewards/concensus_correctness_reward_func": 14.171375036239624, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.5415192926302552, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.086000021547079, "step": 348 }, { "completion_length": 243.5625, "epoch": 4.45859872611465, "grad_norm": 3.805427312850952, "kl": 9.271330177783966, "learning_rate": 3.586894092650992e-08, "loss": 0.0093, "reward": 9.116474032402039, "reward_std": 7.954511460848153, "rewards/concensus_correctness_reward_func": 5.309750065207481, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.5349741466343403, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.0842500105500221, "step": 352 }, { "completion_length": 216.0, "epoch": 4.509554140127388, "grad_norm": 1.2827595472335815, "kl": 586.344656933099, "learning_rate": 3.311134618223448e-08, "loss": 0.5863, "reward": 18.384202480316162, "reward_std": 6.0147824686719105, "rewards/concensus_correctness_reward_func": 13.971250057220459, "rewards/consensus_reward_func": 1.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.6950150849297643, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.375, "rewards/xmlcount_reward_func": 1.2179375290870667, "step": 356 }, { "completion_length": 238.25, "epoch": 4.560509554140127, "grad_norm": 2.5090322494506836, "kl": 14.409381460398436, "learning_rate": 3.045084468876636e-08, "loss": 0.0144, "reward": 11.096249029040337, "reward_std": 5.002050343900919, "rewards/concensus_correctness_reward_func": 7.652750000357628, "rewards/consensus_reward_func": 1.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.6521867662668228, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 0.9788125269114971, "step": 360 }, { "completion_length": 231.6875, "epoch": 4.611464968152866, "grad_norm": 5.940495014190674, "kl": 5.444742456078529, "learning_rate": 2.7889646384626974e-08, "loss": 0.0054, "reward": 16.203705072402954, "reward_std": 6.409488521516323, "rewards/concensus_correctness_reward_func": 11.46025002002716, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.7280802987515926, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.1716250255703926, "step": 364 }, { "completion_length": 261.9375, "epoch": 4.662420382165605, "grad_norm": 4.413754463195801, "kl": 40.07878900319338, "learning_rate": 2.5429878722406294e-08, "loss": 0.0401, "reward": 12.807837784290314, "reward_std": 5.907793074846268, "rewards/concensus_correctness_reward_func": 9.157750010490417, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.5562750669196248, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.0313125103712082, "step": 368 }, { "completion_length": 246.5625, "epoch": 4.713375796178344, "grad_norm": 2.676518201828003, "kl": 193.69527792930603, "learning_rate": 2.3073584901600297e-08, "loss": 0.1937, "reward": 15.810337275266647, "reward_std": 7.08709304407239, "rewards/concensus_correctness_reward_func": 11.445250004529953, "rewards/consensus_reward_func": 0.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.5, "rewards/question_recreation_reward_func": 0.7263994300737977, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.0761875212192535, "step": 372 }, { "completion_length": 220.375, "epoch": 4.764331210191083, "grad_norm": 15.561013221740723, "kl": 10.431547999382019, "learning_rate": 2.0822722171432373e-08, "loss": 0.0104, "reward": 16.888904064893723, "reward_std": 4.636547790840268, "rewards/concensus_correctness_reward_func": 12.990749955177307, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.42490432411432266, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.125, "rewards/xmlcount_reward_func": 1.0982500240206718, "step": 376 }, { "completion_length": 284.375, "epoch": 4.8152866242038215, "grad_norm": 2.221562385559082, "kl": 28.070011030882597, "learning_rate": 1.867916020506981e-08, "loss": 0.0281, "reward": 17.162397533655167, "reward_std": 4.522519879043102, "rewards/concensus_correctness_reward_func": 12.688500046730042, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.0, "rewards/question_recreation_reward_func": 0.600835133343935, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.0605625286698341, "step": 380 }, { "completion_length": 255.9375, "epoch": 4.86624203821656, "grad_norm": 3.4525911808013916, "kl": 13.200175400823355, "learning_rate": 1.6644679546584393e-08, "loss": 0.0132, "reward": 20.264204382896423, "reward_std": 3.887154497206211, "rewards/concensus_correctness_reward_func": 15.187625050544739, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.729454442858696, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.159624993801117, "step": 384 }, { "completion_length": 250.0625, "epoch": 4.917197452229299, "grad_norm": 1.8867841958999634, "kl": 8.001008473336697, "learning_rate": 1.4720970131948435e-08, "loss": 0.008, "reward": 18.74275043606758, "reward_std": 2.8640489764511585, "rewards/concensus_correctness_reward_func": 13.75, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6746251694858074, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.0681250244379044, "step": 388 }, { "completion_length": 244.5, "epoch": 4.968152866242038, "grad_norm": 3.219937801361084, "kl": 2.38198983669281, "learning_rate": 1.290962988529376e-08, "loss": 0.0024, "reward": 15.04402369260788, "reward_std": 4.118689671158791, "rewards/concensus_correctness_reward_func": 10.858500003814697, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.6415863204747438, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.2001875191926956, "step": 392 }, { "completion_length": 240.0, "epoch": 5.012738853503185, "grad_norm": 3.246076822280884, "kl": 16.680181052003586, "learning_rate": 1.1212163391600377e-08, "loss": 0.0146, "reward": 14.724290439060756, "reward_std": 6.1574646064213345, "rewards/concensus_correctness_reward_func": 10.47371428353446, "rewards/consensus_reward_func": 1.2857142857142858, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.2857142857142858, "rewards/question_recreation_reward_func": 0.5896476145301547, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.17857142857142858, "rewards/xmlcount_reward_func": 0.9109285984720502, "step": 396 }, { "completion_length": 235.25, "epoch": 5.063694267515924, "grad_norm": 2.46883487701416, "kl": 2.3545443788170815, "learning_rate": 9.629980646916868e-09, "loss": 0.0024, "reward": 16.084708213806152, "reward_std": 8.817421786487103, "rewards/concensus_correctness_reward_func": 11.687125086784363, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.4455205723643303, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.2020625323057175, "step": 400 }, { "completion_length": 247.125, "epoch": 5.114649681528663, "grad_norm": 2.4745118618011475, "kl": 2.9937894009053707, "learning_rate": 8.164395887150988e-09, "loss": 0.003, "reward": 14.124228060245514, "reward_std": 9.18977701291442, "rewards/concensus_correctness_reward_func": 9.378499984741211, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.720603097230196, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.2126250267028809, "step": 404 }, { "completion_length": 250.1875, "epoch": 5.165605095541402, "grad_norm": 2.0114660263061523, "kl": 21.8758534938097, "learning_rate": 6.816626496402977e-09, "loss": 0.0219, "reward": 13.668376445770264, "reward_std": 2.5443186461925507, "rewards/concensus_correctness_reward_func": 9.617874965071678, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.5688141155987978, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.0754375234246254, "step": 408 }, { "completion_length": 241.625, "epoch": 5.2165605095541405, "grad_norm": 2.3374829292297363, "kl": 3.71017612144351, "learning_rate": 5.5877919957487805e-09, "loss": 0.0037, "reward": 19.984151780605316, "reward_std": 4.235110512934625, "rewards/concensus_correctness_reward_func": 15.0, "rewards/consensus_reward_func": 1.75, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.8057144843041897, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.21875, "rewards/xmlcount_reward_func": 1.0846875384449959, "step": 412 }, { "completion_length": 218.5, "epoch": 5.267515923566879, "grad_norm": 1151.6732177734375, "kl": 297.7076738625765, "learning_rate": 4.478913113312771e-09, "loss": 0.2977, "reward": 13.16609039902687, "reward_std": 9.443016417324543, "rewards/concensus_correctness_reward_func": 8.948874950408936, "rewards/consensus_reward_func": 0.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.6226529218256474, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.40625, "rewards/xmlcount_reward_func": 1.1883125305175781, "step": 416 }, { "completion_length": 222.9375, "epoch": 5.318471337579618, "grad_norm": 2.7844438552856445, "kl": 9.570182114839554, "learning_rate": 3.490910936402691e-09, "loss": 0.0096, "reward": 18.053339332342148, "reward_std": 6.633316494524479, "rewards/concensus_correctness_reward_func": 13.75, "rewards/consensus_reward_func": 1.5, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.6550895310938358, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.1875, "rewards/xmlcount_reward_func": 1.2107500284910202, "step": 420 }, { "completion_length": 213.75, "epoch": 5.369426751592357, "grad_norm": 11.249397277832031, "kl": 17.067561771720648, "learning_rate": 2.6246061464108014e-09, "loss": 0.0171, "reward": 17.524891823530197, "reward_std": 4.37577617354691, "rewards/concensus_correctness_reward_func": 12.710250005126, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6043295841664076, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 1.2103125154972076, "step": 424 }, { "completion_length": 209.3125, "epoch": 5.420382165605096, "grad_norm": 3.837719202041626, "kl": 2.012119587510824, "learning_rate": 1.8807183371172484e-09, "loss": 0.002, "reward": 17.180835396051407, "reward_std": 4.515378940850496, "rewards/concensus_correctness_reward_func": 12.949124962091446, "rewards/consensus_reward_func": 0.75, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.5834604157134891, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.210750013589859, "step": 428 }, { "completion_length": 205.6875, "epoch": 5.471337579617835, "grad_norm": 2.5619990825653076, "kl": 13.116049729287624, "learning_rate": 1.2598654169611977e-09, "loss": 0.0131, "reward": 18.689254820346832, "reward_std": 5.8718895125202835, "rewards/concensus_correctness_reward_func": 13.973749995231628, "rewards/consensus_reward_func": 1.625, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.4517549993470311, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.28125, "rewards/xmlcount_reward_func": 1.2325000166893005, "step": 432 }, { "completion_length": 221.4375, "epoch": 5.522292993630574, "grad_norm": 4.5784711837768555, "kl": 13.171938437968493, "learning_rate": 7.625630957769357e-10, "loss": 0.0132, "reward": 12.938094705343246, "reward_std": 9.368400201201439, "rewards/concensus_correctness_reward_func": 9.259250115603209, "rewards/consensus_reward_func": 0.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.625, "rewards/question_recreation_reward_func": 0.6295947674661875, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.34375, "rewards/xmlcount_reward_func": 1.2055000215768814, "step": 436 }, { "completion_length": 235.4375, "epoch": 5.573248407643312, "grad_norm": 16.68247413635254, "kl": 12.497772630304098, "learning_rate": 3.8922445642086687e-10, "loss": 0.0125, "reward": 15.425481706857681, "reward_std": 2.6193624921143055, "rewards/concensus_correctness_reward_func": 11.479375004768372, "rewards/consensus_reward_func": 1.375, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.75, "rewards/question_recreation_reward_func": 0.4780444288626313, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.15625, "rewards/xmlcount_reward_func": 1.1868125051259995, "step": 440 }, { "completion_length": 259.8125, "epoch": 5.624203821656051, "grad_norm": 2.506514549255371, "kl": 2.1330077573657036, "learning_rate": 1.4015961164528522e-10, "loss": 0.0023, "reward": 20.30971536040306, "reward_std": 0.7802602792944526, "rewards/concensus_correctness_reward_func": 15.0, "rewards/consensus_reward_func": 1.875, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.375, "rewards/question_recreation_reward_func": 0.6319650914520025, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.1152500286698341, "step": 444 }, { "completion_length": 213.75, "epoch": 5.67515923566879, "grad_norm": 1.8979109525680542, "kl": 5.903007443994284, "learning_rate": 1.5575446504029198e-11, "loss": 0.0059, "reward": 15.859139263629913, "reward_std": 3.203619036823511, "rewards/concensus_correctness_reward_func": 11.463249996304512, "rewards/consensus_reward_func": 1.25, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.125, "rewards/question_recreation_reward_func": 0.637013990432024, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.3125, "rewards/xmlcount_reward_func": 1.0713750198483467, "step": 448 }, { "completion_length": 231.25, "epoch": 5.7006369426751595, "kl": 1.2166772335767746, "reward": 22.533211946487427, "reward_std": 5.739101491868496, "rewards/concensus_correctness_reward_func": 17.5, "rewards/consensus_reward_func": 1.75, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 1.25, "rewards/question_recreation_reward_func": 0.7859619110822678, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.25, "rewards/xmlcount_reward_func": 0.9972500205039978, "step": 450, "total_flos": 0.0, "train_loss": 0.03514481741314133, "train_runtime": 2046.9514, "train_samples_per_second": 0.879, "train_steps_per_second": 0.22 } ], "logging_steps": 4, "max_steps": 450, "num_input_tokens_seen": 0, "num_train_epochs": 6, "save_steps": 40, "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": 2, "trial_name": null, "trial_params": null }