diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,9393 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "eval_steps": 500, + "global_step": 585, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "all_correct": 0.625, + "all_wrong": 0.0, + "completion_length": 360.09375, + "epoch": 0.0017094017094017094, + "grad_norm": 1.9826301268041722, + "kl": 0.0, + "learning_rate": 4.999963950687845e-07, + "loss": 0.0, + "reward": 2.6583333015441895, + "reward_std": 0.8285301923751831, + "rewards/accuracy_reward": 1.7989583015441895, + "rewards/format_reward": 0.53125, + "step": 1, + "temporal_rewards": 0.75 + }, + { + "all_correct": 0.75, + "all_wrong": 0.0, + "completion_length": 460.71875, + "epoch": 0.003418803418803419, + "grad_norm": 1.783036944219682, + "kl": 0.00041484832763671875, + "learning_rate": 4.999855803791026e-07, + "loss": 0.0, + "reward": 3.078125, + "reward_std": 0.714574933052063, + "rewards/accuracy_reward": 2.090625047683716, + "rewards/format_reward": 0.65625, + "step": 2, + "temporal_rewards": 0.75 + }, + { + "all_correct": 0.75, + "all_wrong": 0.0, + "completion_length": 458.53125, + "epoch": 0.005128205128205128, + "grad_norm": 1.9140893572712068, + "kl": 0.004611968994140625, + "learning_rate": 4.999675562428436e-07, + "loss": 0.0002, + "reward": 3.1812500953674316, + "reward_std": 0.6964144110679626, + "rewards/accuracy_reward": 2.0531249046325684, + "rewards/format_reward": 0.78125, + "step": 3, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.625, + "all_wrong": 0.125, + "completion_length": 342.34375, + "epoch": 0.006837606837606838, + "grad_norm": 1.6435177595568693, + "kl": 0.0047607421875, + "learning_rate": 4.99942323179814e-07, + "loss": 0.0002, + "reward": 3.0593748092651367, + "reward_std": 0.4561102092266083, + "rewards/accuracy_reward": 1.834375023841858, + "rewards/format_reward": 0.9375, + "step": 4, + "temporal_rewards": 0.625 + }, + { + "all_correct": 0.75, + "all_wrong": 0.0, + "completion_length": 369.84375, + "epoch": 0.008547008547008548, + "grad_norm": 1.9133739405680183, + "kl": 0.004207611083984375, + "learning_rate": 4.999098819177214e-07, + "loss": 0.0002, + "reward": 3.379166603088379, + "reward_std": 0.6761210560798645, + "rewards/accuracy_reward": 2.0791666507720947, + "rewards/format_reward": 0.9375, + "step": 5, + "temporal_rewards": 0.75 + }, + { + "all_correct": 0.625, + "all_wrong": 0.0, + "completion_length": 351.4375, + "epoch": 0.010256410256410256, + "grad_norm": 1.7373006031428762, + "kl": 0.008636474609375, + "learning_rate": 4.998702333921537e-07, + "loss": 0.0003, + "reward": 3.2947916984558105, + "reward_std": 0.7285541296005249, + "rewards/accuracy_reward": 1.988541603088379, + "rewards/format_reward": 1.0, + "step": 6, + "temporal_rewards": 0.75 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 330.90625, + "epoch": 0.011965811965811967, + "grad_norm": 1.81484083295842, + "kl": 0.00798797607421875, + "learning_rate": 4.998233787465529e-07, + "loss": 0.0003, + "reward": 3.910416603088379, + "reward_std": 0.41429048776626587, + "rewards/accuracy_reward": 2.519791603088379, + "rewards/format_reward": 1.0, + "step": 7, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.75, + "all_wrong": 0.125, + "completion_length": 337.34375, + "epoch": 0.013675213675213675, + "grad_norm": 1.5267078479800362, + "kl": 0.00785064697265625, + "learning_rate": 4.99769319332181e-07, + "loss": 0.0003, + "reward": 3.3531250953674316, + "reward_std": 0.3145015239715576, + "rewards/accuracy_reward": 2.034374952316284, + "rewards/format_reward": 1.0, + "step": 8, + "temporal_rewards": 0.75 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 335.0625, + "epoch": 0.015384615384615385, + "grad_norm": 1.685582407194424, + "kl": 0.00934600830078125, + "learning_rate": 4.997080567080816e-07, + "loss": 0.0004, + "reward": 3.6895830631256104, + "reward_std": 0.36123037338256836, + "rewards/accuracy_reward": 2.3427083492279053, + "rewards/format_reward": 0.96875, + "step": 9, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 353.875, + "epoch": 0.017094017094017096, + "grad_norm": 1.5329400178924226, + "kl": 0.0093841552734375, + "learning_rate": 4.996395926410354e-07, + "loss": 0.0004, + "reward": 3.886458396911621, + "reward_std": 0.3707513213157654, + "rewards/accuracy_reward": 2.4739582538604736, + "rewards/format_reward": 0.96875, + "step": 10, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 351.0, + "epoch": 0.018803418803418803, + "grad_norm": 1.549766282148727, + "kl": 0.0126953125, + "learning_rate": 4.995639291055083e-07, + "loss": 0.0005, + "reward": 4.042708396911621, + "reward_std": 0.2290562391281128, + "rewards/accuracy_reward": 2.586458206176758, + "rewards/format_reward": 1.0, + "step": 11, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.75, + "all_wrong": 0.0, + "completion_length": 358.1875, + "epoch": 0.020512820512820513, + "grad_norm": 1.496305479107766, + "kl": 0.016632080078125, + "learning_rate": 4.994810682835951e-07, + "loss": 0.0007, + "reward": 3.6510417461395264, + "reward_std": 0.5492165088653564, + "rewards/accuracy_reward": 2.3135416507720947, + "rewards/format_reward": 0.96875, + "step": 12, + "temporal_rewards": 0.75 + }, + { + "all_correct": 0.75, + "all_wrong": 0.0, + "completion_length": 368.4375, + "epoch": 0.022222222222222223, + "grad_norm": 1.541763401661613, + "kl": 0.0149078369140625, + "learning_rate": 4.99391012564956e-07, + "loss": 0.0006, + "reward": 3.7291665077209473, + "reward_std": 0.4612278342247009, + "rewards/accuracy_reward": 2.351041793823242, + "rewards/format_reward": 0.96875, + "step": 13, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 374.09375, + "epoch": 0.023931623931623933, + "grad_norm": 1.548862572081062, + "kl": 0.0169219970703125, + "learning_rate": 4.992937645467487e-07, + "loss": 0.0007, + "reward": 3.936458110809326, + "reward_std": 0.32494786381721497, + "rewards/accuracy_reward": 2.5614583492279053, + "rewards/format_reward": 0.96875, + "step": 14, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 374.21875, + "epoch": 0.02564102564102564, + "grad_norm": 1.3211685700244122, + "kl": 0.0189056396484375, + "learning_rate": 4.991893270335525e-07, + "loss": 0.0008, + "reward": 3.8302083015441895, + "reward_std": 0.275448203086853, + "rewards/accuracy_reward": 2.386458396911621, + "rewards/format_reward": 1.0, + "step": 15, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 390.65625, + "epoch": 0.02735042735042735, + "grad_norm": 1.0827777389563216, + "kl": 0.0201568603515625, + "learning_rate": 4.990777030372877e-07, + "loss": 0.0008, + "reward": 3.918750047683716, + "reward_std": 0.27283355593681335, + "rewards/accuracy_reward": 2.46875, + "rewards/format_reward": 1.0, + "step": 16, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 396.5, + "epoch": 0.02905982905982906, + "grad_norm": 1.4719325483552113, + "kl": 0.0209808349609375, + "learning_rate": 4.989588957771289e-07, + "loss": 0.0008, + "reward": 4.057291030883789, + "reward_std": 0.38743653893470764, + "rewards/accuracy_reward": 2.613541603088379, + "rewards/format_reward": 1.0, + "step": 17, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.15625, + "epoch": 0.03076923076923077, + "grad_norm": 1.3897844926570597, + "kl": 0.0229339599609375, + "learning_rate": 4.988329086794122e-07, + "loss": 0.0009, + "reward": 4.282291412353516, + "reward_std": 0.2059958279132843, + "rewards/accuracy_reward": 2.8260416984558105, + "rewards/format_reward": 1.0, + "step": 18, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 394.75, + "epoch": 0.03247863247863248, + "grad_norm": 1.4025810108116477, + "kl": 0.0246124267578125, + "learning_rate": 4.98699745377536e-07, + "loss": 0.001, + "reward": 3.6979165077209473, + "reward_std": 0.22571580111980438, + "rewards/accuracy_reward": 2.285416603088379, + "rewards/format_reward": 1.0, + "step": 19, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 395.375, + "epoch": 0.03418803418803419, + "grad_norm": 1.3422743689560948, + "kl": 0.0244293212890625, + "learning_rate": 4.98559409711857e-07, + "loss": 0.001, + "reward": 4.1895833015441895, + "reward_std": 0.34896954894065857, + "rewards/accuracy_reward": 2.764583110809326, + "rewards/format_reward": 1.0, + "step": 20, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.5625, + "epoch": 0.035897435897435895, + "grad_norm": 1.3706326401655777, + "kl": 0.0267791748046875, + "learning_rate": 4.984119057295782e-07, + "loss": 0.0011, + "reward": 4.2083330154418945, + "reward_std": 0.21551749110221863, + "rewards/accuracy_reward": 2.714583396911621, + "rewards/format_reward": 1.0, + "step": 21, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 416.375, + "epoch": 0.037606837606837605, + "grad_norm": 1.3430039575658776, + "kl": 0.0264739990234375, + "learning_rate": 4.982572376846336e-07, + "loss": 0.0011, + "reward": 4.012499809265137, + "reward_std": 0.36676347255706787, + "rewards/accuracy_reward": 2.5687501430511475, + "rewards/format_reward": 1.0, + "step": 22, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 414.5, + "epoch": 0.039316239316239315, + "grad_norm": 1.0704324370355776, + "kl": 0.027374267578125, + "learning_rate": 4.980954100375641e-07, + "loss": 0.0011, + "reward": 3.9510416984558105, + "reward_std": 0.21451421082019806, + "rewards/accuracy_reward": 2.476041793823242, + "rewards/format_reward": 1.0, + "step": 23, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.75, + "all_wrong": 0.125, + "completion_length": 385.6875, + "epoch": 0.041025641025641026, + "grad_norm": 1.319994732732891, + "kl": 0.0298309326171875, + "learning_rate": 4.979264274553905e-07, + "loss": 0.0012, + "reward": 3.832291603088379, + "reward_std": 0.3102354407310486, + "rewards/accuracy_reward": 2.4510416984558105, + "rewards/format_reward": 1.0, + "step": 24, + "temporal_rewards": 0.75 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 417.15625, + "epoch": 0.042735042735042736, + "grad_norm": 1.2440324072815687, + "kl": 0.025665283203125, + "learning_rate": 4.977502948114771e-07, + "loss": 0.001, + "reward": 4.2677083015441895, + "reward_std": 0.11905767768621445, + "rewards/accuracy_reward": 2.7989585399627686, + "rewards/format_reward": 1.0, + "step": 25, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 397.78125, + "epoch": 0.044444444444444446, + "grad_norm": 1.3286663602585649, + "kl": 0.02685546875, + "learning_rate": 4.975670171853925e-07, + "loss": 0.0011, + "reward": 3.9354166984558105, + "reward_std": 0.29135236144065857, + "rewards/accuracy_reward": 2.522916793823242, + "rewards/format_reward": 1.0, + "step": 26, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 429.0, + "epoch": 0.046153846153846156, + "grad_norm": 1.301323110823831, + "kl": 0.0247650146484375, + "learning_rate": 4.973765998627628e-07, + "loss": 0.001, + "reward": 3.8374998569488525, + "reward_std": 0.3257061839103699, + "rewards/accuracy_reward": 2.4000000953674316, + "rewards/format_reward": 1.0, + "step": 27, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.28125, + "epoch": 0.04786324786324787, + "grad_norm": 1.2554920983763502, + "kl": 0.029266357421875, + "learning_rate": 4.971790483351185e-07, + "loss": 0.0012, + "reward": 4.2864580154418945, + "reward_std": 0.131798654794693, + "rewards/accuracy_reward": 2.7989583015441895, + "rewards/format_reward": 1.0, + "step": 28, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.75, + "all_wrong": 0.0, + "completion_length": 394.0625, + "epoch": 0.04957264957264957, + "grad_norm": 0.9339974580407308, + "kl": 0.0288543701171875, + "learning_rate": 4.969743682997371e-07, + "loss": 0.0012, + "reward": 3.6687498092651367, + "reward_std": 0.45044267177581787, + "rewards/accuracy_reward": 2.25, + "rewards/format_reward": 1.0, + "step": 29, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.4375, + "epoch": 0.05128205128205128, + "grad_norm": 1.2262654726579296, + "kl": 0.0308685302734375, + "learning_rate": 4.967625656594781e-07, + "loss": 0.0012, + "reward": 4.2395830154418945, + "reward_std": 0.16444088518619537, + "rewards/accuracy_reward": 2.7708332538604736, + "rewards/format_reward": 1.0, + "step": 30, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.25, + "epoch": 0.05299145299145299, + "grad_norm": 1.2573879496841676, + "kl": 0.0293731689453125, + "learning_rate": 4.965436465226134e-07, + "loss": 0.0012, + "reward": 4.016666412353516, + "reward_std": 0.14875400066375732, + "rewards/accuracy_reward": 2.535416603088379, + "rewards/format_reward": 1.0, + "step": 31, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 368.6875, + "epoch": 0.0547008547008547, + "grad_norm": 1.2564985420337802, + "kl": 0.030120849609375, + "learning_rate": 4.963176172026501e-07, + "loss": 0.0012, + "reward": 3.9968748092651367, + "reward_std": 0.3774029016494751, + "rewards/accuracy_reward": 2.559375047683716, + "rewards/format_reward": 1.0, + "step": 32, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 424.0625, + "epoch": 0.05641025641025641, + "grad_norm": 0.9694026424422826, + "kl": 0.033355712890625, + "learning_rate": 4.960844842181495e-07, + "loss": 0.0013, + "reward": 4.047916412353516, + "reward_std": 0.3180779814720154, + "rewards/accuracy_reward": 2.616666555404663, + "rewards/format_reward": 1.0, + "step": 33, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.03125, + "epoch": 0.05811965811965812, + "grad_norm": 1.1514296295303623, + "kl": 0.03143310546875, + "learning_rate": 4.958442542925385e-07, + "loss": 0.0013, + "reward": 3.9666666984558105, + "reward_std": 0.14849570393562317, + "rewards/accuracy_reward": 2.4791667461395264, + "rewards/format_reward": 1.0, + "step": 34, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 362.71875, + "epoch": 0.05982905982905983, + "grad_norm": 1.1335397647489873, + "kl": 0.0335693359375, + "learning_rate": 4.955969343539162e-07, + "loss": 0.0013, + "reward": 4.120833396911621, + "reward_std": 0.1538814902305603, + "rewards/accuracy_reward": 2.6583333015441895, + "rewards/format_reward": 1.0, + "step": 35, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 385.40625, + "epoch": 0.06153846153846154, + "grad_norm": 1.3288920319551032, + "kl": 0.03155517578125, + "learning_rate": 4.953425315348533e-07, + "loss": 0.0013, + "reward": 4.229166507720947, + "reward_std": 0.2994784116744995, + "rewards/accuracy_reward": 2.7760415077209473, + "rewards/format_reward": 1.0, + "step": 36, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 407.46875, + "epoch": 0.06324786324786325, + "grad_norm": 1.3491529940946705, + "kl": 0.0291900634765625, + "learning_rate": 4.950810531721873e-07, + "loss": 0.0012, + "reward": 4.1583333015441895, + "reward_std": 0.08171668648719788, + "rewards/accuracy_reward": 2.727083206176758, + "rewards/format_reward": 1.0, + "step": 37, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.375, + "epoch": 0.06495726495726496, + "grad_norm": 1.4603865780151248, + "kl": 0.02972412109375, + "learning_rate": 4.948125068068102e-07, + "loss": 0.0012, + "reward": 4.2864580154418945, + "reward_std": 0.11952318251132965, + "rewards/accuracy_reward": 2.811458110809326, + "rewards/format_reward": 1.0, + "step": 38, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 476.25, + "epoch": 0.06666666666666667, + "grad_norm": 1.3332368328871798, + "kl": 0.02838134765625, + "learning_rate": 4.945369001834514e-07, + "loss": 0.0011, + "reward": 4.363541603088379, + "reward_std": 0.16119801998138428, + "rewards/accuracy_reward": 2.8885416984558105, + "rewards/format_reward": 1.0, + "step": 39, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.90625, + "epoch": 0.06837606837606838, + "grad_norm": 1.1501096034585943, + "kl": 0.0298004150390625, + "learning_rate": 4.942542412504542e-07, + "loss": 0.0012, + "reward": 4.179166316986084, + "reward_std": 0.09968569874763489, + "rewards/accuracy_reward": 2.6854166984558105, + "rewards/format_reward": 1.0, + "step": 40, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.78125, + "epoch": 0.07008547008547009, + "grad_norm": 1.3385150586201982, + "kl": 0.0303802490234375, + "learning_rate": 4.939645381595469e-07, + "loss": 0.0012, + "reward": 4.415625095367432, + "reward_std": 0.12376689910888672, + "rewards/accuracy_reward": 2.9468750953674316, + "rewards/format_reward": 1.0, + "step": 41, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.125, + "epoch": 0.07179487179487179, + "grad_norm": 1.5092483520373827, + "kl": 0.032257080078125, + "learning_rate": 4.93667799265607e-07, + "loss": 0.0013, + "reward": 4.179166793823242, + "reward_std": 0.24446402490139008, + "rewards/accuracy_reward": 2.7291665077209473, + "rewards/format_reward": 1.0, + "step": 42, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 413.6875, + "epoch": 0.0735042735042735, + "grad_norm": 0.8198637140300333, + "kl": 0.0333251953125, + "learning_rate": 4.933640331264209e-07, + "loss": 0.0013, + "reward": 4.356249809265137, + "reward_std": 0.02693380042910576, + "rewards/accuracy_reward": 2.862499952316284, + "rewards/format_reward": 1.0, + "step": 43, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.34375, + "epoch": 0.07521367521367521, + "grad_norm": 1.362239639644523, + "kl": 0.03790283203125, + "learning_rate": 4.930532485024371e-07, + "loss": 0.0015, + "reward": 4.239583492279053, + "reward_std": 0.15847893059253693, + "rewards/accuracy_reward": 2.7645833492279053, + "rewards/format_reward": 1.0, + "step": 44, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.34375, + "epoch": 0.07692307692307693, + "grad_norm": 1.348031471721414, + "kl": 0.03143310546875, + "learning_rate": 4.92735454356513e-07, + "loss": 0.0013, + "reward": 4.290624618530273, + "reward_std": 0.0891985222697258, + "rewards/accuracy_reward": 2.796875, + "rewards/format_reward": 1.0, + "step": 45, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.8125, + "epoch": 0.07863247863247863, + "grad_norm": 1.0447875433549505, + "kl": 0.0330810546875, + "learning_rate": 4.924106598536569e-07, + "loss": 0.0013, + "reward": 4.339583396911621, + "reward_std": 0.0612565279006958, + "rewards/accuracy_reward": 2.858333110809326, + "rewards/format_reward": 1.0, + "step": 46, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.03125, + "epoch": 0.08034188034188035, + "grad_norm": 1.1567990099669652, + "kl": 0.03277587890625, + "learning_rate": 4.920788743607635e-07, + "loss": 0.0013, + "reward": 4.3927083015441895, + "reward_std": 0.13107708096504211, + "rewards/accuracy_reward": 2.9114582538604736, + "rewards/format_reward": 1.0, + "step": 47, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.8125, + "epoch": 0.08205128205128205, + "grad_norm": 1.0369718009838986, + "kl": 0.03460693359375, + "learning_rate": 4.917401074463441e-07, + "loss": 0.0014, + "reward": 4.4270830154418945, + "reward_std": 0.13362948596477509, + "rewards/accuracy_reward": 2.9270834922790527, + "rewards/format_reward": 1.0, + "step": 48, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.25, + "epoch": 0.08376068376068375, + "grad_norm": 0.7093588876792561, + "kl": 0.037353515625, + "learning_rate": 4.913943688802497e-07, + "loss": 0.0015, + "reward": 3.8843750953674316, + "reward_std": 0.1348297894001007, + "rewards/accuracy_reward": 2.390625, + "rewards/format_reward": 1.0, + "step": 49, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 415.625, + "epoch": 0.08547008547008547, + "grad_norm": 0.9797316320763304, + "kl": 0.035736083984375, + "learning_rate": 4.910416686333906e-07, + "loss": 0.0014, + "reward": 3.855208396911621, + "reward_std": 0.03958335518836975, + "rewards/accuracy_reward": 2.3614583015441895, + "rewards/format_reward": 1.0, + "step": 50, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 402.71875, + "epoch": 0.08717948717948718, + "grad_norm": 0.7267487162340365, + "kl": 0.034515380859375, + "learning_rate": 4.906820168774477e-07, + "loss": 0.0014, + "reward": 4.229166507720947, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.7291665077209473, + "rewards/format_reward": 1.0, + "step": 51, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 408.65625, + "epoch": 0.08888888888888889, + "grad_norm": 0.9894251393442087, + "kl": 0.0360107421875, + "learning_rate": 4.903154239845797e-07, + "loss": 0.0014, + "reward": 4.327083587646484, + "reward_std": 0.09055028110742569, + "rewards/accuracy_reward": 2.8270833492279053, + "rewards/format_reward": 1.0, + "step": 52, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.21875, + "epoch": 0.0905982905982906, + "grad_norm": 1.2134526914387425, + "kl": 0.0322265625, + "learning_rate": 4.899419005271241e-07, + "loss": 0.0013, + "reward": 4.413541793823242, + "reward_std": 0.1331278383731842, + "rewards/accuracy_reward": 2.9197916984558105, + "rewards/format_reward": 1.0, + "step": 53, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 428.34375, + "epoch": 0.09230769230769231, + "grad_norm": 1.1445171148845454, + "kl": 0.035369873046875, + "learning_rate": 4.895614572772916e-07, + "loss": 0.0014, + "reward": 4.2208333015441895, + "reward_std": 0.23445481061935425, + "rewards/accuracy_reward": 2.745833396911621, + "rewards/format_reward": 1.0, + "step": 54, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 404.46875, + "epoch": 0.09401709401709402, + "grad_norm": 0.9479258360176128, + "kl": 0.036468505859375, + "learning_rate": 4.891741052068563e-07, + "loss": 0.0015, + "reward": 4.464583396911621, + "reward_std": 0.05090419948101044, + "rewards/accuracy_reward": 2.977083206176758, + "rewards/format_reward": 1.0, + "step": 55, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 419.21875, + "epoch": 0.09572649572649573, + "grad_norm": 1.185262142708624, + "kl": 0.0364990234375, + "learning_rate": 4.887798554868387e-07, + "loss": 0.0015, + "reward": 4.202083110809326, + "reward_std": 0.08526714146137238, + "rewards/accuracy_reward": 2.7333333492279053, + "rewards/format_reward": 1.0, + "step": 56, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 427.78125, + "epoch": 0.09743589743589744, + "grad_norm": 0.6149452604974177, + "kl": 0.033416748046875, + "learning_rate": 4.883787194871841e-07, + "loss": 0.0013, + "reward": 4.359375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.871875047683716, + "rewards/format_reward": 1.0, + "step": 57, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.4375, + "epoch": 0.09914529914529914, + "grad_norm": 1.1440282896265581, + "kl": 0.039398193359375, + "learning_rate": 4.879707087764336e-07, + "loss": 0.0016, + "reward": 4.335416793823242, + "reward_std": 0.06786316633224487, + "rewards/accuracy_reward": 2.835416793823242, + "rewards/format_reward": 1.0, + "step": 58, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.34375, + "epoch": 0.10085470085470086, + "grad_norm": 1.1667079878189268, + "kl": 0.0361328125, + "learning_rate": 4.875558351213917e-07, + "loss": 0.0014, + "reward": 4.196874618530273, + "reward_std": 0.07863453030586243, + "rewards/accuracy_reward": 2.7093749046325684, + "rewards/format_reward": 1.0, + "step": 59, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 402.3125, + "epoch": 0.10256410256410256, + "grad_norm": 0.7257762494267078, + "kl": 0.04315185546875, + "learning_rate": 4.871341104867864e-07, + "loss": 0.0017, + "reward": 4.106249809265137, + "reward_std": 0.03750002384185791, + "rewards/accuracy_reward": 2.606250047683716, + "rewards/format_reward": 1.0, + "step": 60, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 390.375, + "epoch": 0.10427350427350428, + "grad_norm": 1.246730170179853, + "kl": 0.038177490234375, + "learning_rate": 4.86705547034924e-07, + "loss": 0.0015, + "reward": 4.362500190734863, + "reward_std": 0.18642225861549377, + "rewards/accuracy_reward": 2.9124999046325684, + "rewards/format_reward": 1.0, + "step": 61, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.5625, + "epoch": 0.10598290598290598, + "grad_norm": 0.757303226225468, + "kl": 0.0382080078125, + "learning_rate": 4.862701571253386e-07, + "loss": 0.0015, + "reward": 4.3645830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.8645832538604736, + "rewards/format_reward": 1.0, + "step": 62, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.84375, + "epoch": 0.1076923076923077, + "grad_norm": 0.7830222478805806, + "kl": 0.0361328125, + "learning_rate": 4.858279533144357e-07, + "loss": 0.0014, + "reward": 4.366666793823242, + "reward_std": 0.016666710376739502, + "rewards/accuracy_reward": 2.866666793823242, + "rewards/format_reward": 1.0, + "step": 63, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.0, + "epoch": 0.1094017094017094, + "grad_norm": 0.7856721207258216, + "kl": 0.037872314453125, + "learning_rate": 4.853789483551299e-07, + "loss": 0.0015, + "reward": 4.328125, + "reward_std": 0.07114279270172119, + "rewards/accuracy_reward": 2.828125, + "rewards/format_reward": 1.0, + "step": 64, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.6875, + "epoch": 0.1111111111111111, + "grad_norm": 0.5081050946401335, + "kl": 0.040435791015625, + "learning_rate": 4.849231551964771e-07, + "loss": 0.0016, + "reward": 4.213541507720947, + "reward_std": 0.059839192777872086, + "rewards/accuracy_reward": 2.7135415077209473, + "rewards/format_reward": 1.0, + "step": 65, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.4375, + "epoch": 0.11282051282051282, + "grad_norm": 0.7425251071608985, + "kl": 0.041046142578125, + "learning_rate": 4.844605869833011e-07, + "loss": 0.0016, + "reward": 4.2239580154418945, + "reward_std": 0.05208335071802139, + "rewards/accuracy_reward": 2.7239584922790527, + "rewards/format_reward": 1.0, + "step": 66, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.59375, + "epoch": 0.11452991452991453, + "grad_norm": 1.7762232281747519, + "kl": 0.03778076171875, + "learning_rate": 4.839912570558147e-07, + "loss": 0.0015, + "reward": 4.457291603088379, + "reward_std": 0.08541667461395264, + "rewards/accuracy_reward": 2.9635417461395264, + "rewards/format_reward": 1.0, + "step": 67, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.96875, + "epoch": 0.11623931623931624, + "grad_norm": 1.0171107252467595, + "kl": 0.039794921875, + "learning_rate": 4.835151789492348e-07, + "loss": 0.0016, + "reward": 4.327083587646484, + "reward_std": 0.06022907793521881, + "rewards/accuracy_reward": 2.8333334922790527, + "rewards/format_reward": 1.0, + "step": 68, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.0, + "epoch": 0.11794871794871795, + "grad_norm": 0.8418226397769828, + "kl": 0.038909912109375, + "learning_rate": 4.830323663933919e-07, + "loss": 0.0016, + "reward": 4.474999904632568, + "reward_std": 0.034150637686252594, + "rewards/accuracy_reward": 2.981250047683716, + "rewards/format_reward": 1.0, + "step": 69, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 358.40625, + "epoch": 0.11965811965811966, + "grad_norm": 1.2733531515130359, + "kl": 0.036468505859375, + "learning_rate": 4.825428333123346e-07, + "loss": 0.0015, + "reward": 4.4375, + "reward_std": 0.06786306202411652, + "rewards/accuracy_reward": 2.968750238418579, + "rewards/format_reward": 1.0, + "step": 70, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.875, + "epoch": 0.12136752136752137, + "grad_norm": 1.1124941217119864, + "kl": 0.039459228515625, + "learning_rate": 4.820465938239273e-07, + "loss": 0.0016, + "reward": 4.077083110809326, + "reward_std": 0.09583336114883423, + "rewards/accuracy_reward": 2.5833332538604736, + "rewards/format_reward": 1.0, + "step": 71, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.0625, + "epoch": 0.12307692307692308, + "grad_norm": 0.7611058988515808, + "kl": 0.037384033203125, + "learning_rate": 4.815436622394441e-07, + "loss": 0.0015, + "reward": 4.3333330154418945, + "reward_std": 0.07452815771102905, + "rewards/accuracy_reward": 2.8333330154418945, + "rewards/format_reward": 1.0, + "step": 72, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 401.3125, + "epoch": 0.12478632478632479, + "grad_norm": 0.7580960727925862, + "kl": 0.03826904296875, + "learning_rate": 4.810340530631549e-07, + "loss": 0.0015, + "reward": 4.196874618530273, + "reward_std": 0.042802631855010986, + "rewards/accuracy_reward": 2.7093749046325684, + "rewards/format_reward": 1.0, + "step": 73, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.8125, + "epoch": 0.1264957264957265, + "grad_norm": 1.0374945029632214, + "kl": 0.036468505859375, + "learning_rate": 4.805177809919081e-07, + "loss": 0.0015, + "reward": 4.40625, + "reward_std": 0.13188013434410095, + "rewards/accuracy_reward": 2.9124999046325684, + "rewards/format_reward": 1.0, + "step": 74, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.5, + "epoch": 0.1282051282051282, + "grad_norm": 1.4163910687642582, + "kl": 0.0360107421875, + "learning_rate": 4.799948609147061e-07, + "loss": 0.0014, + "reward": 4.358333110809326, + "reward_std": 0.20628556609153748, + "rewards/accuracy_reward": 2.9083333015441895, + "rewards/format_reward": 0.96875, + "step": 75, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 376.75, + "epoch": 0.12991452991452992, + "grad_norm": 0.9270824940096077, + "kl": 0.03839111328125, + "learning_rate": 4.794653079122759e-07, + "loss": 0.0015, + "reward": 3.981250047683716, + "reward_std": 0.3762458562850952, + "rewards/accuracy_reward": 2.53125, + "rewards/format_reward": 1.0, + "step": 76, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.625, + "epoch": 0.13162393162393163, + "grad_norm": 0.9270824940096077, + "kl": 0.036865234375, + "learning_rate": 4.794653079122759e-07, + "loss": 0.0015, + "reward": 4.4666666984558105, + "reward_std": 0.053745806217193604, + "rewards/accuracy_reward": 2.972916603088379, + "rewards/format_reward": 1.0, + "step": 77, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 365.625, + "epoch": 0.13333333333333333, + "grad_norm": 1.1283643584028937, + "kl": 0.03497314453125, + "learning_rate": 4.789291372566351e-07, + "loss": 0.0014, + "reward": 4.336458206176758, + "reward_std": 0.06416243314743042, + "rewards/accuracy_reward": 2.8489584922790527, + "rewards/format_reward": 1.0, + "step": 78, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.78125, + "epoch": 0.13504273504273503, + "grad_norm": 1.0511802041200045, + "kl": 0.03485107421875, + "learning_rate": 4.783863644106502e-07, + "loss": 0.0014, + "reward": 4.3385419845581055, + "reward_std": 0.05041898041963577, + "rewards/accuracy_reward": 2.851041793823242, + "rewards/format_reward": 1.0, + "step": 79, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.0625, + "epoch": 0.13675213675213677, + "grad_norm": 1.1818496933235974, + "kl": 0.0379638671875, + "learning_rate": 4.778370050275913e-07, + "loss": 0.0015, + "reward": 4.327083110809326, + "reward_std": 0.055192895233631134, + "rewards/accuracy_reward": 2.8458333015441895, + "rewards/format_reward": 1.0, + "step": 80, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.21875, + "epoch": 0.13846153846153847, + "grad_norm": 0.9747451365148834, + "kl": 0.032684326171875, + "learning_rate": 4.772810749506809e-07, + "loss": 0.0013, + "reward": 4.213541507720947, + "reward_std": 0.05453469604253769, + "rewards/accuracy_reward": 2.7197914123535156, + "rewards/format_reward": 1.0, + "step": 81, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.21875, + "epoch": 0.14017094017094017, + "grad_norm": 0.5883256070908801, + "kl": 0.031829833984375, + "learning_rate": 4.767185902126363e-07, + "loss": 0.0013, + "reward": 4.368750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 2.875, + "rewards/format_reward": 1.0, + "step": 82, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 365.8125, + "epoch": 0.14188034188034188, + "grad_norm": 1.0703457007134716, + "kl": 0.038055419921875, + "learning_rate": 4.7614956703520804e-07, + "loss": 0.0015, + "reward": 4.228124618530273, + "reward_std": 0.15567252039909363, + "rewards/accuracy_reward": 2.734375, + "rewards/format_reward": 1.0, + "step": 83, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 359.59375, + "epoch": 0.14358974358974358, + "grad_norm": 1.2506164162586872, + "kl": 0.03045654296875, + "learning_rate": 4.755740218287112e-07, + "loss": 0.0012, + "reward": 4.460416793823242, + "reward_std": 0.06860043108463287, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 84, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.9375, + "epoch": 0.1452991452991453, + "grad_norm": 0.7246056285956093, + "kl": 0.0301971435546875, + "learning_rate": 4.74991971191553e-07, + "loss": 0.0012, + "reward": 4.366666793823242, + "reward_std": 0.016666710376739502, + "rewards/accuracy_reward": 2.866666555404663, + "rewards/format_reward": 1.0, + "step": 85, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.53125, + "epoch": 0.147008547008547, + "grad_norm": 1.1668199192545485, + "kl": 0.0360107421875, + "learning_rate": 4.7440343190975353e-07, + "loss": 0.0014, + "reward": 4.4395833015441895, + "reward_std": 0.0936460942029953, + "rewards/accuracy_reward": 2.9583334922790527, + "rewards/format_reward": 1.0, + "step": 86, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.75, + "epoch": 0.14871794871794872, + "grad_norm": 0.9704193227590118, + "kl": 0.03289794921875, + "learning_rate": 4.738084209564617e-07, + "loss": 0.0013, + "reward": 4.3177080154418945, + "reward_std": 0.08490978181362152, + "rewards/accuracy_reward": 2.8302083015441895, + "rewards/format_reward": 1.0, + "step": 87, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 361.1875, + "epoch": 0.15042735042735042, + "grad_norm": 1.1259678958812716, + "kl": 0.0352783203125, + "learning_rate": 4.73206955491466e-07, + "loss": 0.0014, + "reward": 4.262499809265137, + "reward_std": 0.09999996423721313, + "rewards/accuracy_reward": 2.78125, + "rewards/format_reward": 1.0, + "step": 88, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.15625, + "epoch": 0.15213675213675212, + "grad_norm": 1.1425644306497342, + "kl": 0.035125732421875, + "learning_rate": 4.7259905286069954e-07, + "loss": 0.0014, + "reward": 4.217708587646484, + "reward_std": 0.05812295526266098, + "rewards/accuracy_reward": 2.7177083492279053, + "rewards/format_reward": 1.0, + "step": 89, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.21875, + "epoch": 0.15384615384615385, + "grad_norm": 0.9074640334431748, + "kl": 0.03875732421875, + "learning_rate": 4.7198473059573974e-07, + "loss": 0.0015, + "reward": 4.353124618530273, + "reward_std": 0.04375004023313522, + "rewards/accuracy_reward": 2.859375, + "rewards/format_reward": 1.0, + "step": 90, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.125, + "epoch": 0.15555555555555556, + "grad_norm": 0.7003982378481962, + "kl": 0.032257080078125, + "learning_rate": 4.7136400641330245e-07, + "loss": 0.0013, + "reward": 4.480208396911621, + "reward_std": 0.03958336263895035, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 91, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.71875, + "epoch": 0.15726495726495726, + "grad_norm": 0.875300468522747, + "kl": 0.033599853515625, + "learning_rate": 4.707368982147317e-07, + "loss": 0.0013, + "reward": 4.135416507720947, + "reward_std": 0.10416668653488159, + "rewards/accuracy_reward": 2.6354165077209473, + "rewards/format_reward": 1.0, + "step": 92, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.4375, + "epoch": 0.15897435897435896, + "grad_norm": 1.3612654393709456, + "kl": 0.03582763671875, + "learning_rate": 4.7010342408548287e-07, + "loss": 0.0014, + "reward": 4.120833396911621, + "reward_std": 0.17232877016067505, + "rewards/accuracy_reward": 2.664583206176758, + "rewards/format_reward": 1.0, + "step": 93, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.28125, + "epoch": 0.1606837606837607, + "grad_norm": 0.5201581820004612, + "kl": 0.037445068359375, + "learning_rate": 4.6946360229460114e-07, + "loss": 0.0015, + "reward": 4.34375, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.84375, + "rewards/format_reward": 1.0, + "step": 94, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.8125, + "epoch": 0.1623931623931624, + "grad_norm": 0.681909030921276, + "kl": 0.034759521484375, + "learning_rate": 4.6881745129419493e-07, + "loss": 0.0014, + "reward": 4.340624809265137, + "reward_std": 0.06875002384185791, + "rewards/accuracy_reward": 2.840625047683716, + "rewards/format_reward": 1.0, + "step": 95, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 376.03125, + "epoch": 0.1641025641025641, + "grad_norm": 1.1733277547649887, + "kl": 0.0390625, + "learning_rate": 4.6816498971890357e-07, + "loss": 0.0016, + "reward": 4.235416412353516, + "reward_std": 0.2851067781448364, + "rewards/accuracy_reward": 2.7916667461395264, + "rewards/format_reward": 1.0, + "step": 96, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.25, + "epoch": 0.1658119658119658, + "grad_norm": 1.2608932989375246, + "kl": 0.037841796875, + "learning_rate": 4.675062363853598e-07, + "loss": 0.0015, + "reward": 4.422916412353516, + "reward_std": 0.146120086312294, + "rewards/accuracy_reward": 2.922916889190674, + "rewards/format_reward": 1.0, + "step": 97, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.21875, + "epoch": 0.1675213675213675, + "grad_norm": 0.9647820810612151, + "kl": 0.034912109375, + "learning_rate": 4.668412102916473e-07, + "loss": 0.0014, + "reward": 4.3302083015441895, + "reward_std": 0.04382467269897461, + "rewards/accuracy_reward": 2.855208396911621, + "rewards/format_reward": 1.0, + "step": 98, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.125, + "epoch": 0.16923076923076924, + "grad_norm": 1.2632460448619276, + "kl": 0.036651611328125, + "learning_rate": 4.661699306167527e-07, + "loss": 0.0015, + "reward": 4.460416793823242, + "reward_std": 0.06192883849143982, + "rewards/accuracy_reward": 2.972916603088379, + "rewards/format_reward": 1.0, + "step": 99, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.25, + "epoch": 0.17094017094017094, + "grad_norm": 0.8583887047833437, + "kl": 0.0360107421875, + "learning_rate": 4.6549241672001225e-07, + "loss": 0.0014, + "reward": 4.269791603088379, + "reward_std": 0.08541667461395264, + "rewards/accuracy_reward": 2.7760415077209473, + "rewards/format_reward": 1.0, + "step": 100, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.5, + "epoch": 0.17264957264957265, + "grad_norm": 1.1101859505758365, + "kl": 0.03741455078125, + "learning_rate": 4.648086881405542e-07, + "loss": 0.0015, + "reward": 3.9375, + "reward_std": 0.30000001192092896, + "rewards/accuracy_reward": 2.481250047683716, + "rewards/format_reward": 1.0, + "step": 101, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 378.625, + "epoch": 0.17435897435897435, + "grad_norm": 1.247638864286834, + "kl": 0.045440673828125, + "learning_rate": 4.6411876459673425e-07, + "loss": 0.0018, + "reward": 4.053124904632568, + "reward_std": 0.3468588590621948, + "rewards/accuracy_reward": 2.621875047683716, + "rewards/format_reward": 1.0, + "step": 102, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.90625, + "epoch": 0.17606837606837608, + "grad_norm": 0.5504424525596744, + "kl": 0.039276123046875, + "learning_rate": 4.634226659855681e-07, + "loss": 0.0016, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 103, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.09375, + "epoch": 0.17777777777777778, + "grad_norm": 0.10526479007567158, + "kl": 0.03863525390625, + "learning_rate": 4.6272041238215624e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 104, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 397.59375, + "epoch": 0.1794871794871795, + "grad_norm": 0.7304258125371498, + "kl": 0.036346435546875, + "learning_rate": 4.6201202403910643e-07, + "loss": 0.0015, + "reward": 4.356249809265137, + "reward_std": 0.2124999761581421, + "rewards/accuracy_reward": 2.90625, + "rewards/format_reward": 1.0, + "step": 105, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.3125, + "epoch": 0.1811965811965812, + "grad_norm": 1.3375206675543057, + "kl": 0.0443115234375, + "learning_rate": 4.612975213859487e-07, + "loss": 0.0018, + "reward": 4.321874618530273, + "reward_std": 0.22130155563354492, + "rewards/accuracy_reward": 2.8343749046325684, + "rewards/format_reward": 1.0, + "step": 106, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.0, + "epoch": 0.1829059829059829, + "grad_norm": 1.395717497711829, + "kl": 0.042724609375, + "learning_rate": 4.6057692502854615e-07, + "loss": 0.0017, + "reward": 4.465624809265137, + "reward_std": 0.06875002384185791, + "rewards/accuracy_reward": 2.965625047683716, + "rewards/format_reward": 1.0, + "step": 107, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 407.03125, + "epoch": 0.18461538461538463, + "grad_norm": 0.4761809503163303, + "kl": 0.0377197265625, + "learning_rate": 4.5985025574850147e-07, + "loss": 0.0015, + "reward": 4.4979166984558105, + "reward_std": 0.004166662693023682, + "rewards/accuracy_reward": 2.9979166984558105, + "rewards/format_reward": 1.0, + "step": 108, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 409.375, + "epoch": 0.18632478632478633, + "grad_norm": 0.7405328781008026, + "kl": 0.039459228515625, + "learning_rate": 4.591175345025566e-07, + "loss": 0.0016, + "reward": 4.4270830154418945, + "reward_std": 0.09300213307142258, + "rewards/accuracy_reward": 2.9270834922790527, + "rewards/format_reward": 1.0, + "step": 109, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.28125, + "epoch": 0.18803418803418803, + "grad_norm": 0.4422606573426667, + "kl": 0.04058837890625, + "learning_rate": 4.5837878242198936e-07, + "loss": 0.0016, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 110, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.78125, + "epoch": 0.18974358974358974, + "grad_norm": 0.7932812380553261, + "kl": 0.04156494140625, + "learning_rate": 4.576340208120029e-07, + "loss": 0.0017, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 111, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.625, + "epoch": 0.19145299145299147, + "grad_norm": 0.9371190698328027, + "kl": 0.042938232421875, + "learning_rate": 4.568832711511125e-07, + "loss": 0.0017, + "reward": 4.453125, + "reward_std": 0.08570331335067749, + "rewards/accuracy_reward": 2.965625047683716, + "rewards/format_reward": 1.0, + "step": 112, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 417.09375, + "epoch": 0.19316239316239317, + "grad_norm": 0.9352268527879245, + "kl": 0.039825439453125, + "learning_rate": 4.56126555090525e-07, + "loss": 0.0016, + "reward": 4.378125190734863, + "reward_std": 0.2437499761581421, + "rewards/accuracy_reward": 2.8968749046325684, + "rewards/format_reward": 1.0, + "step": 113, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 411.0625, + "epoch": 0.19487179487179487, + "grad_norm": 1.0061361213200832, + "kl": 0.048492431640625, + "learning_rate": 4.5536389445351543e-07, + "loss": 0.0019, + "reward": 4.324999809265137, + "reward_std": 0.22499999403953552, + "rewards/accuracy_reward": 2.8687500953674316, + "rewards/format_reward": 0.96875, + "step": 114, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.40625, + "epoch": 0.19658119658119658, + "grad_norm": 0.7838565061379232, + "kl": 0.037445068359375, + "learning_rate": 4.5459531123479673e-07, + "loss": 0.0015, + "reward": 4.407291889190674, + "reward_std": 0.1695673167705536, + "rewards/accuracy_reward": 2.913541555404663, + "rewards/format_reward": 1.0, + "step": 115, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.3125, + "epoch": 0.19829059829059828, + "grad_norm": 0.7434533516049402, + "kl": 0.05072021484375, + "learning_rate": 4.5382082759988605e-07, + "loss": 0.002, + "reward": 4.337500095367432, + "reward_std": 0.07500002533197403, + "rewards/accuracy_reward": 2.8375000953674316, + "rewards/format_reward": 1.0, + "step": 116, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 399.3125, + "epoch": 0.2, + "grad_norm": 1.0066718401998942, + "kl": 0.040252685546875, + "learning_rate": 4.530404658844653e-07, + "loss": 0.0016, + "reward": 4.273958206176758, + "reward_std": 0.22102615237236023, + "rewards/accuracy_reward": 2.8177080154418945, + "rewards/format_reward": 1.0, + "step": 117, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.5625, + "epoch": 0.20170940170940171, + "grad_norm": 1.2338773011379278, + "kl": 0.040985107421875, + "learning_rate": 4.5225424859373684e-07, + "loss": 0.0016, + "reward": 4.435416221618652, + "reward_std": 0.12916666269302368, + "rewards/accuracy_reward": 2.9479165077209473, + "rewards/format_reward": 1.0, + "step": 118, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.6875, + "epoch": 0.20341880341880342, + "grad_norm": 1.2606475694446262, + "kl": 0.044189453125, + "learning_rate": 4.5146219840177475e-07, + "loss": 0.0018, + "reward": 4.474999904632568, + "reward_std": 0.050000011920928955, + "rewards/accuracy_reward": 2.987499952316284, + "rewards/format_reward": 1.0, + "step": 119, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 410.34375, + "epoch": 0.20512820512820512, + "grad_norm": 1.0930625377031935, + "kl": 0.034515380859375, + "learning_rate": 4.506643381508707e-07, + "loss": 0.0014, + "reward": 4.4635419845581055, + "reward_std": 0.07291674613952637, + "rewards/accuracy_reward": 2.9635417461395264, + "rewards/format_reward": 1.0, + "step": 120, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 408.71875, + "epoch": 0.20683760683760682, + "grad_norm": 1.1859878555606564, + "kl": 0.041656494140625, + "learning_rate": 4.498606908508753e-07, + "loss": 0.0017, + "reward": 4.470832824707031, + "reward_std": 0.058333415538072586, + "rewards/accuracy_reward": 2.9708335399627686, + "rewards/format_reward": 1.0, + "step": 121, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 412.5, + "epoch": 0.20854700854700856, + "grad_norm": 0.854920065845963, + "kl": 0.036956787109375, + "learning_rate": 4.490512796785344e-07, + "loss": 0.0015, + "reward": 4.475000381469727, + "reward_std": 0.050000011920928955, + "rewards/accuracy_reward": 2.981250047683716, + "rewards/format_reward": 1.0, + "step": 122, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 404.125, + "epoch": 0.21025641025641026, + "grad_norm": 1.2913323108581982, + "kl": 0.04022216796875, + "learning_rate": 4.4823612797682087e-07, + "loss": 0.0016, + "reward": 4.4083333015441895, + "reward_std": 0.10176850110292435, + "rewards/accuracy_reward": 2.914583206176758, + "rewards/format_reward": 1.0, + "step": 123, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 408.75, + "epoch": 0.21196581196581196, + "grad_norm": 1.2109364265824558, + "kl": 0.042877197265625, + "learning_rate": 4.474152592542612e-07, + "loss": 0.0017, + "reward": 4.381249904632568, + "reward_std": 0.22619642317295074, + "rewards/accuracy_reward": 2.887500047683716, + "rewards/format_reward": 1.0, + "step": 124, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 404.03125, + "epoch": 0.21367521367521367, + "grad_norm": 0.7902156024479733, + "kl": 0.0445556640625, + "learning_rate": 4.4658869718425774e-07, + "loss": 0.0018, + "reward": 4.4864583015441895, + "reward_std": 0.027083376422524452, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 125, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 402.03125, + "epoch": 0.2153846153846154, + "grad_norm": 1.0575353123534599, + "kl": 0.044830322265625, + "learning_rate": 4.457564656044056e-07, + "loss": 0.0018, + "reward": 4.453125, + "reward_std": 0.06609740853309631, + "rewards/accuracy_reward": 2.9656248092651367, + "rewards/format_reward": 1.0, + "step": 126, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.375, + "epoch": 0.2170940170940171, + "grad_norm": 1.0839676629160238, + "kl": 0.043975830078125, + "learning_rate": 4.4491858851580553e-07, + "loss": 0.0018, + "reward": 4.472916603088379, + "reward_std": 0.05416667461395264, + "rewards/accuracy_reward": 2.9854166507720947, + "rewards/format_reward": 1.0, + "step": 127, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.09375, + "epoch": 0.2188034188034188, + "grad_norm": 1.4967525172984775, + "kl": 0.043731689453125, + "learning_rate": 4.4407509008237196e-07, + "loss": 0.0018, + "reward": 4.415625095367432, + "reward_std": 0.12062296271324158, + "rewards/accuracy_reward": 2.921875, + "rewards/format_reward": 1.0, + "step": 128, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.65625, + "epoch": 0.2205128205128205, + "grad_norm": 1.156753232859856, + "kl": 0.045379638671875, + "learning_rate": 4.4322599463013545e-07, + "loss": 0.0018, + "reward": 4.4510416984558105, + "reward_std": 0.07613600790500641, + "rewards/accuracy_reward": 2.9510416984558105, + "rewards/format_reward": 1.0, + "step": 129, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.375, + "epoch": 0.2222222222222222, + "grad_norm": 0.7981149668305919, + "kl": 0.04595947265625, + "learning_rate": 4.4237132664654147e-07, + "loss": 0.0018, + "reward": 4.368750095367432, + "reward_std": 0.1907489001750946, + "rewards/accuracy_reward": 2.875, + "rewards/format_reward": 1.0, + "step": 130, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 397.0625, + "epoch": 0.22393162393162394, + "grad_norm": 0.8783345359959521, + "kl": 0.046112060546875, + "learning_rate": 4.415111107797445e-07, + "loss": 0.0018, + "reward": 4.3729166984558105, + "reward_std": 0.17916670441627502, + "rewards/accuracy_reward": 2.910416841506958, + "rewards/format_reward": 1.0, + "step": 131, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.375, + "epoch": 0.22564102564102564, + "grad_norm": 0.7538049450716533, + "kl": 0.047943115234375, + "learning_rate": 4.4064537183789675e-07, + "loss": 0.0019, + "reward": 4.487500190734863, + "reward_std": 0.01821785606443882, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 132, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.78125, + "epoch": 0.22735042735042735, + "grad_norm": 1.1842869997989454, + "kl": 0.04400634765625, + "learning_rate": 4.397741347884328e-07, + "loss": 0.0018, + "reward": 4.432291507720947, + "reward_std": 0.0720747858285904, + "rewards/accuracy_reward": 2.976041793823242, + "rewards/format_reward": 1.0, + "step": 133, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.5625, + "epoch": 0.22905982905982905, + "grad_norm": 1.1785282734794618, + "kl": 0.05255126953125, + "learning_rate": 4.3889742475735e-07, + "loss": 0.0021, + "reward": 4.413541316986084, + "reward_std": 0.17291665077209473, + "rewards/accuracy_reward": 2.9322915077209473, + "rewards/format_reward": 1.0, + "step": 134, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.40625, + "epoch": 0.23076923076923078, + "grad_norm": 0.8559474662788366, + "kl": 0.0535888671875, + "learning_rate": 4.38015267028483e-07, + "loss": 0.0021, + "reward": 4.487500190734863, + "reward_std": 0.025000017136335373, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 135, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 384.65625, + "epoch": 0.23247863247863249, + "grad_norm": 1.229565072425211, + "kl": 0.05072021484375, + "learning_rate": 4.3712768704277524e-07, + "loss": 0.002, + "reward": 4.3072919845581055, + "reward_std": 0.3026726245880127, + "rewards/accuracy_reward": 2.851041793823242, + "rewards/format_reward": 1.0, + "step": 136, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.46875, + "epoch": 0.2341880341880342, + "grad_norm": 1.0543514102617846, + "kl": 0.051483154296875, + "learning_rate": 4.3623471039754525e-07, + "loss": 0.0021, + "reward": 4.4822916984558105, + "reward_std": 0.035416703671216965, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 137, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 375.9375, + "epoch": 0.2358974358974359, + "grad_norm": 0.9569773264572374, + "kl": 0.0533447265625, + "learning_rate": 4.3533636284574796e-07, + "loss": 0.0021, + "reward": 4.259374618530273, + "reward_std": 0.1562500149011612, + "rewards/accuracy_reward": 2.809375047683716, + "rewards/format_reward": 1.0, + "step": 138, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.71875, + "epoch": 0.2376068376068376, + "grad_norm": 0.9824411737431323, + "kl": 0.055755615234375, + "learning_rate": 4.3443267029523254e-07, + "loss": 0.0022, + "reward": 4.480208396911621, + "reward_std": 0.03958336263895035, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 139, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.59375, + "epoch": 0.23931623931623933, + "grad_norm": 1.0420267070552791, + "kl": 0.048675537109375, + "learning_rate": 4.335236588079948e-07, + "loss": 0.0019, + "reward": 4.477083206176758, + "reward_std": 0.04583332687616348, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 140, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.34375, + "epoch": 0.24102564102564103, + "grad_norm": 0.7563242853625025, + "kl": 0.04864501953125, + "learning_rate": 4.326093545994258e-07, + "loss": 0.0019, + "reward": 4.493750095367432, + "reward_std": 0.01250004768371582, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 141, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.625, + "epoch": 0.24273504273504273, + "grad_norm": 0.8629864588588781, + "kl": 0.047576904296875, + "learning_rate": 4.316897840375558e-07, + "loss": 0.0019, + "reward": 4.472916603088379, + "reward_std": 0.05416667461395264, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 142, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.40625, + "epoch": 0.24444444444444444, + "grad_norm": 0.8960359847522277, + "kl": 0.047332763671875, + "learning_rate": 4.307649736422938e-07, + "loss": 0.0019, + "reward": 4.358333110809326, + "reward_std": 0.03333333879709244, + "rewards/accuracy_reward": 2.8645834922790527, + "rewards/format_reward": 1.0, + "step": 143, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.15625, + "epoch": 0.24615384615384617, + "grad_norm": 0.7057007703502248, + "kl": 0.045379638671875, + "learning_rate": 4.2983495008466273e-07, + "loss": 0.0018, + "reward": 4.454166412353516, + "reward_std": 0.08357105404138565, + "rewards/accuracy_reward": 2.991666793823242, + "rewards/format_reward": 0.96875, + "step": 144, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.625, + "epoch": 0.24786324786324787, + "grad_norm": 1.1400069761424125, + "kl": 0.0487060546875, + "learning_rate": 4.2889974018603024e-07, + "loss": 0.0019, + "reward": 4.257291793823242, + "reward_std": 0.20642347633838654, + "rewards/accuracy_reward": 2.7760415077209473, + "rewards/format_reward": 1.0, + "step": 145, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.375, + "epoch": 0.24957264957264957, + "grad_norm": 0.5078458420795325, + "kl": 0.045196533203125, + "learning_rate": 4.279593709173351e-07, + "loss": 0.0018, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 146, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 401.03125, + "epoch": 0.2512820512820513, + "grad_norm": 0.9708960933046578, + "kl": 0.048797607421875, + "learning_rate": 4.2701386939830964e-07, + "loss": 0.002, + "reward": 4.426041603088379, + "reward_std": 0.1076958030462265, + "rewards/accuracy_reward": 2.9697916507720947, + "rewards/format_reward": 0.96875, + "step": 147, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.28125, + "epoch": 0.252991452991453, + "grad_norm": 0.7150957515941181, + "kl": 0.047882080078125, + "learning_rate": 4.2606326289669737e-07, + "loss": 0.0019, + "reward": 4.465624809265137, + "reward_std": 0.06875002384185791, + "rewards/accuracy_reward": 2.965625047683716, + "rewards/format_reward": 1.0, + "step": 148, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.59375, + "epoch": 0.2547008547008547, + "grad_norm": 0.940492080718513, + "kl": 0.047332763671875, + "learning_rate": 4.251075788274666e-07, + "loss": 0.0019, + "reward": 4.4791669845581055, + "reward_std": 0.036383602768182755, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 149, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.65625, + "epoch": 0.2564102564102564, + "grad_norm": 1.1493825902639314, + "kl": 0.04644775390625, + "learning_rate": 4.241468447520201e-07, + "loss": 0.0019, + "reward": 4.480208396911621, + "reward_std": 0.03958338499069214, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 150, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 398.28125, + "epoch": 0.25811965811965815, + "grad_norm": 1.459476642452973, + "kl": 0.043121337890625, + "learning_rate": 4.2318108837739986e-07, + "loss": 0.0017, + "reward": 4.234375, + "reward_std": 0.41515326499938965, + "rewards/accuracy_reward": 2.7906250953674316, + "rewards/format_reward": 1.0, + "step": 151, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 411.5, + "epoch": 0.25982905982905985, + "grad_norm": 1.110652347817343, + "kl": 0.050811767578125, + "learning_rate": 4.222103375554883e-07, + "loss": 0.002, + "reward": 4.445833206176758, + "reward_std": 0.06399547308683395, + "rewards/accuracy_reward": 2.952083110809326, + "rewards/format_reward": 1.0, + "step": 152, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.125, + "epoch": 0.26153846153846155, + "grad_norm": 1.0785519824350434, + "kl": 0.0518798828125, + "learning_rate": 4.21234620282205e-07, + "loss": 0.0021, + "reward": 4.2552080154418945, + "reward_std": 0.1653914898633957, + "rewards/accuracy_reward": 2.7552084922790527, + "rewards/format_reward": 1.0, + "step": 153, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 406.75, + "epoch": 0.26324786324786326, + "grad_norm": 1.0721439274563058, + "kl": 0.04229736328125, + "learning_rate": 4.2025396469669926e-07, + "loss": 0.0017, + "reward": 4.454166412353516, + "reward_std": 0.07500007003545761, + "rewards/accuracy_reward": 2.9541664123535156, + "rewards/format_reward": 1.0, + "step": 154, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 404.46875, + "epoch": 0.26495726495726496, + "grad_norm": 0.9988184997997093, + "kl": 0.043060302734375, + "learning_rate": 4.1926839908053847e-07, + "loss": 0.0017, + "reward": 4.479166507720947, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.9854166507720947, + "rewards/format_reward": 1.0, + "step": 155, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.5625, + "epoch": 0.26666666666666666, + "grad_norm": 0.7831640694765154, + "kl": 0.045196533203125, + "learning_rate": 4.182779518568925e-07, + "loss": 0.0018, + "reward": 4.4197916984558105, + "reward_std": 0.0699336901307106, + "rewards/accuracy_reward": 2.9322915077209473, + "rewards/format_reward": 1.0, + "step": 156, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 391.125, + "epoch": 0.26837606837606837, + "grad_norm": 0.9630033697673518, + "kl": 0.046722412109375, + "learning_rate": 4.172826515897145e-07, + "loss": 0.0019, + "reward": 4.3125, + "reward_std": 0.24396438896656036, + "rewards/accuracy_reward": 2.828125, + "rewards/format_reward": 1.0, + "step": 157, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.0625, + "epoch": 0.27008547008547007, + "grad_norm": 1.17921986026744, + "kl": 0.041473388671875, + "learning_rate": 4.1628252698291643e-07, + "loss": 0.0017, + "reward": 4.467708587646484, + "reward_std": 0.064583420753479, + "rewards/accuracy_reward": 2.9677083492279053, + "rewards/format_reward": 1.0, + "step": 158, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.125, + "epoch": 0.2717948717948718, + "grad_norm": 1.0249267708010807, + "kl": 0.042877197265625, + "learning_rate": 4.1527760687954154e-07, + "loss": 0.0017, + "reward": 4.4739580154418945, + "reward_std": 0.05208338797092438, + "rewards/accuracy_reward": 2.9739582538604736, + "rewards/format_reward": 1.0, + "step": 159, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.34375, + "epoch": 0.27350427350427353, + "grad_norm": 0.77719285849163, + "kl": 0.041015625, + "learning_rate": 4.142679202609327e-07, + "loss": 0.0016, + "reward": 4.354166507720947, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.8541667461395264, + "rewards/format_reward": 1.0, + "step": 160, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.25, + "epoch": 0.27521367521367524, + "grad_norm": 0.7752980246081294, + "kl": 0.039459228515625, + "learning_rate": 4.132534962458962e-07, + "loss": 0.0016, + "reward": 4.4270830154418945, + "reward_std": 0.06305919587612152, + "rewards/accuracy_reward": 2.9270834922790527, + "rewards/format_reward": 1.0, + "step": 161, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 375.75, + "epoch": 0.27692307692307694, + "grad_norm": 1.1993765246662056, + "kl": 0.047943115234375, + "learning_rate": 4.122343640898627e-07, + "loss": 0.0019, + "reward": 4.353125095367432, + "reward_std": 0.20582912862300873, + "rewards/accuracy_reward": 2.903125047683716, + "rewards/format_reward": 1.0, + "step": 162, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.03125, + "epoch": 0.27863247863247864, + "grad_norm": 0.59014945249797, + "kl": 0.041290283203125, + "learning_rate": 4.112105531840426e-07, + "loss": 0.0017, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 163, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.625, + "epoch": 0.28034188034188035, + "grad_norm": 0.5356413668339389, + "kl": 0.040740966796875, + "learning_rate": 4.101820930545791e-07, + "loss": 0.0016, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 164, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.96875, + "epoch": 0.28205128205128205, + "grad_norm": 0.13294410521207053, + "kl": 0.04388427734375, + "learning_rate": 4.0914901336169636e-07, + "loss": 0.0018, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 165, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.40625, + "epoch": 0.28376068376068375, + "grad_norm": 0.5764716736438367, + "kl": 0.0455322265625, + "learning_rate": 4.081113438988443e-07, + "loss": 0.0018, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 166, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.625, + "epoch": 0.28547008547008546, + "grad_norm": 1.465826987702493, + "kl": 0.045806884765625, + "learning_rate": 4.0706911459183915e-07, + "loss": 0.0018, + "reward": 4.4041666984558105, + "reward_std": 0.17083334922790527, + "rewards/accuracy_reward": 2.910416603088379, + "rewards/format_reward": 1.0, + "step": 167, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.78125, + "epoch": 0.28717948717948716, + "grad_norm": 0.5935301744550797, + "kl": 0.047821044921875, + "learning_rate": 4.060223554980007e-07, + "loss": 0.0019, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 168, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 494.09375, + "epoch": 0.28888888888888886, + "grad_norm": 0.8204426630893709, + "kl": 0.045166015625, + "learning_rate": 4.049710968052851e-07, + "loss": 0.0018, + "reward": 4.272916793823242, + "reward_std": 0.2933366894721985, + "rewards/accuracy_reward": 2.8541667461395264, + "rewards/format_reward": 0.96875, + "step": 169, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.21875, + "epoch": 0.2905982905982906, + "grad_norm": 0.6213443899630995, + "kl": 0.04119873046875, + "learning_rate": 4.039153688314145e-07, + "loss": 0.0016, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 170, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.34375, + "epoch": 0.2923076923076923, + "grad_norm": 0.7861053846314547, + "kl": 0.041229248046875, + "learning_rate": 4.0285520202300304e-07, + "loss": 0.0016, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 171, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 367.75, + "epoch": 0.294017094017094, + "grad_norm": 1.2307384278158895, + "kl": 0.04510498046875, + "learning_rate": 4.017906269546778e-07, + "loss": 0.0018, + "reward": 4.4354166984558105, + "reward_std": 0.10271236300468445, + "rewards/accuracy_reward": 2.960416793823242, + "rewards/format_reward": 1.0, + "step": 172, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.53125, + "epoch": 0.29572649572649573, + "grad_norm": 0.9676974134756903, + "kl": 0.046539306640625, + "learning_rate": 4.0072167432819804e-07, + "loss": 0.0019, + "reward": 4.477083206176758, + "reward_std": 0.036256518214941025, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 173, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.96875, + "epoch": 0.29743589743589743, + "grad_norm": 1.2274057316899676, + "kl": 0.04266357421875, + "learning_rate": 3.996483749715693e-07, + "loss": 0.0017, + "reward": 4.465624809265137, + "reward_std": 0.04625225439667702, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 174, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.46875, + "epoch": 0.29914529914529914, + "grad_norm": 1.2412182079168792, + "kl": 0.03826904296875, + "learning_rate": 3.9857075983815435e-07, + "loss": 0.0015, + "reward": 4.462499618530273, + "reward_std": 0.06443369388580322, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 175, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.0, + "epoch": 0.30085470085470084, + "grad_norm": 0.609179641547218, + "kl": 0.04107666015625, + "learning_rate": 3.974888600057807e-07, + "loss": 0.0016, + "reward": 4.478125095367432, + "reward_std": 0.029536345973610878, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 176, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.90625, + "epoch": 0.30256410256410254, + "grad_norm": 1.190308130875681, + "kl": 0.042327880859375, + "learning_rate": 3.964027066758442e-07, + "loss": 0.0017, + "reward": 4.469791412353516, + "reward_std": 0.0508398711681366, + "rewards/accuracy_reward": 2.9822916984558105, + "rewards/format_reward": 1.0, + "step": 177, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.375, + "epoch": 0.30427350427350425, + "grad_norm": 0.8508866922383489, + "kl": 0.041595458984375, + "learning_rate": 3.9531233117240916e-07, + "loss": 0.0017, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 178, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.6875, + "epoch": 0.305982905982906, + "grad_norm": 1.0986022812237966, + "kl": 0.0469970703125, + "learning_rate": 3.942177649413051e-07, + "loss": 0.0019, + "reward": 4.416666507720947, + "reward_std": 0.16666670143604279, + "rewards/accuracy_reward": 2.9166667461395264, + "rewards/format_reward": 1.0, + "step": 179, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.28125, + "epoch": 0.3076923076923077, + "grad_norm": 0.9041233903377462, + "kl": 0.039581298828125, + "learning_rate": 3.931190395492198e-07, + "loss": 0.0016, + "reward": 4.428125381469727, + "reward_std": 0.14375001192092896, + "rewards/accuracy_reward": 2.934375047683716, + "rewards/format_reward": 1.0, + "step": 180, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.5, + "epoch": 0.3094017094017094, + "grad_norm": 1.0826453818165487, + "kl": 0.040740966796875, + "learning_rate": 3.920161866827889e-07, + "loss": 0.0016, + "reward": 4.436458587646484, + "reward_std": 0.1042187437415123, + "rewards/accuracy_reward": 2.9427082538604736, + "rewards/format_reward": 1.0, + "step": 181, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 414.375, + "epoch": 0.3111111111111111, + "grad_norm": 0.904258739786522, + "kl": 0.0513916015625, + "learning_rate": 3.909092381476824e-07, + "loss": 0.0021, + "reward": 4.332291603088379, + "reward_std": 0.24280627071857452, + "rewards/accuracy_reward": 2.8697915077209473, + "rewards/format_reward": 1.0, + "step": 182, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 411.875, + "epoch": 0.3128205128205128, + "grad_norm": 0.42269488554491963, + "kl": 0.045135498046875, + "learning_rate": 3.8979822586768666e-07, + "loss": 0.0018, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 183, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 418.53125, + "epoch": 0.3145299145299145, + "grad_norm": 0.7604195134349618, + "kl": 0.0445556640625, + "learning_rate": 3.886831818837847e-07, + "loss": 0.0018, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 184, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 415.90625, + "epoch": 0.3162393162393162, + "grad_norm": 1.1493790993662951, + "kl": 0.036895751953125, + "learning_rate": 3.875641383532313e-07, + "loss": 0.0015, + "reward": 4.392707824707031, + "reward_std": 0.20166242122650146, + "rewards/accuracy_reward": 2.9114584922790527, + "rewards/format_reward": 1.0, + "step": 185, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 429.375, + "epoch": 0.31794871794871793, + "grad_norm": 0.4992140699455253, + "kl": 0.039215087890625, + "learning_rate": 3.864411275486261e-07, + "loss": 0.0016, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 186, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 432.4375, + "epoch": 0.31965811965811963, + "grad_norm": 0.7979965621901377, + "kl": 0.03582763671875, + "learning_rate": 3.8531418185698286e-07, + "loss": 0.0014, + "reward": 4.440625190734863, + "reward_std": 0.08476538956165314, + "rewards/accuracy_reward": 2.953125, + "rewards/format_reward": 1.0, + "step": 187, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 420.28125, + "epoch": 0.3213675213675214, + "grad_norm": 0.780631126170187, + "kl": 0.03973388671875, + "learning_rate": 3.8418333377879503e-07, + "loss": 0.0016, + "reward": 4.362500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 2.875, + "rewards/format_reward": 1.0, + "step": 188, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.3125, + "epoch": 0.3230769230769231, + "grad_norm": 0.7047170411411621, + "kl": 0.041778564453125, + "learning_rate": 3.8304861592709904e-07, + "loss": 0.0017, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 189, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.0625, + "epoch": 0.3247863247863248, + "grad_norm": 0.47632381457177264, + "kl": 0.038116455078125, + "learning_rate": 3.8191006102653317e-07, + "loss": 0.0015, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 190, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 404.34375, + "epoch": 0.3264957264957265, + "grad_norm": 0.4836863827599763, + "kl": 0.040771484375, + "learning_rate": 3.8076770191239436e-07, + "loss": 0.0016, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 191, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.3125, + "epoch": 0.3282051282051282, + "grad_norm": 1.0078286777321448, + "kl": 0.03765869140625, + "learning_rate": 3.796215715296909e-07, + "loss": 0.0015, + "reward": 4.480208396911621, + "reward_std": 0.03958336263895035, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 192, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 410.3125, + "epoch": 0.3299145299145299, + "grad_norm": 0.5610391713071925, + "kl": 0.038177490234375, + "learning_rate": 3.7847170293219216e-07, + "loss": 0.0015, + "reward": 4.462500095367432, + "reward_std": 0.059511907398700714, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 193, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 402.03125, + "epoch": 0.3316239316239316, + "grad_norm": 1.1299949442190973, + "kl": 0.043731689453125, + "learning_rate": 3.7731812928147586e-07, + "loss": 0.0017, + "reward": 4.447916507720947, + "reward_std": 0.10416672378778458, + "rewards/accuracy_reward": 2.9479167461395264, + "rewards/format_reward": 1.0, + "step": 194, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.90625, + "epoch": 0.3333333333333333, + "grad_norm": 0.10446066137589627, + "kl": 0.036376953125, + "learning_rate": 3.761608838459713e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 195, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 387.28125, + "epoch": 0.335042735042735, + "grad_norm": 0.7300977881818299, + "kl": 0.038238525390625, + "learning_rate": 3.75e-07, + "loss": 0.0015, + "reward": 4.258333206176758, + "reward_std": 0.308286190032959, + "rewards/accuracy_reward": 2.8020832538604736, + "rewards/format_reward": 1.0, + "step": 196, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.15625, + "epoch": 0.3367521367521368, + "grad_norm": 0.12871250708865678, + "kl": 0.036651611328125, + "learning_rate": 3.7383551122281333e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 197, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 386.75, + "epoch": 0.3384615384615385, + "grad_norm": 0.9165095170318228, + "kl": 0.04473876953125, + "learning_rate": 3.7266745109762667e-07, + "loss": 0.0018, + "reward": 4.324999809265137, + "reward_std": 0.2749999761581421, + "rewards/accuracy_reward": 2.875, + "rewards/format_reward": 1.0, + "step": 198, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.375, + "epoch": 0.3401709401709402, + "grad_norm": 0.789501213084791, + "kl": 0.04150390625, + "learning_rate": 3.7149585331065145e-07, + "loss": 0.0017, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 199, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.03125, + "epoch": 0.3418803418803419, + "grad_norm": 1.11306921071518, + "kl": 0.037689208984375, + "learning_rate": 3.7032075165012317e-07, + "loss": 0.0015, + "reward": 4.446874618530273, + "reward_std": 0.06790292263031006, + "rewards/accuracy_reward": 2.9593749046325684, + "rewards/format_reward": 1.0, + "step": 200, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.875, + "epoch": 0.3435897435897436, + "grad_norm": 0.7761111972471315, + "kl": 0.04083251953125, + "learning_rate": 3.6914218000532694e-07, + "loss": 0.0016, + "reward": 4.472916603088379, + "reward_std": 0.05416666343808174, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 201, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.21875, + "epoch": 0.3452991452991453, + "grad_norm": 0.10020362663647286, + "kl": 0.036346435546875, + "learning_rate": 3.679601723656205e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 202, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.75, + "epoch": 0.347008547008547, + "grad_norm": 0.7893461647051647, + "kl": 0.035858154296875, + "learning_rate": 3.6677476281945383e-07, + "loss": 0.0014, + "reward": 4.4739580154418945, + "reward_std": 0.05208335071802139, + "rewards/accuracy_reward": 2.9739582538604736, + "rewards/format_reward": 1.0, + "step": 203, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.40625, + "epoch": 0.3487179487179487, + "grad_norm": 1.9248208875026689, + "kl": 0.0372314453125, + "learning_rate": 3.655859855533858e-07, + "loss": 0.0015, + "reward": 4.461458206176758, + "reward_std": 0.06416241079568863, + "rewards/accuracy_reward": 2.9739582538604736, + "rewards/format_reward": 1.0, + "step": 204, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.21875, + "epoch": 0.3504273504273504, + "grad_norm": 1.1205344384928364, + "kl": 0.041900634765625, + "learning_rate": 3.6439387485109883e-07, + "loss": 0.0017, + "reward": 4.352083206176758, + "reward_std": 0.1530117392539978, + "rewards/accuracy_reward": 2.8958334922790527, + "rewards/format_reward": 1.0, + "step": 205, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.5625, + "epoch": 0.35213675213675216, + "grad_norm": 0.14604512652457577, + "kl": 0.03472900390625, + "learning_rate": 3.6319846509240935e-07, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 206, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.375, + "epoch": 0.35384615384615387, + "grad_norm": 0.9121169092040915, + "kl": 0.037811279296875, + "learning_rate": 3.6199979075227703e-07, + "loss": 0.0015, + "reward": 4.4583330154418945, + "reward_std": 0.08333337306976318, + "rewards/accuracy_reward": 2.9583334922790527, + "rewards/format_reward": 1.0, + "step": 207, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.6875, + "epoch": 0.35555555555555557, + "grad_norm": 1.1060318359964751, + "kl": 0.04547119140625, + "learning_rate": 3.6079788639981036e-07, + "loss": 0.0018, + "reward": 4.464583396911621, + "reward_std": 0.060267072170972824, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 208, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.28125, + "epoch": 0.3572649572649573, + "grad_norm": 0.5438742329843329, + "kl": 0.035552978515625, + "learning_rate": 3.595927866972693e-07, + "loss": 0.0014, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 209, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.21875, + "epoch": 0.358974358974359, + "grad_norm": 0.7804890857933848, + "kl": 0.03814697265625, + "learning_rate": 3.5838452639906636e-07, + "loss": 0.0015, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 210, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.28125, + "epoch": 0.3606837606837607, + "grad_norm": 1.0411411893808047, + "kl": 0.0460205078125, + "learning_rate": 3.571731403507635e-07, + "loss": 0.0018, + "reward": 4.451041221618652, + "reward_std": 0.0706593245267868, + "rewards/accuracy_reward": 2.9635417461395264, + "rewards/format_reward": 1.0, + "step": 211, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.53125, + "epoch": 0.3623931623931624, + "grad_norm": 1.3492992049831198, + "kl": 0.041290283203125, + "learning_rate": 3.5595866348806784e-07, + "loss": 0.0017, + "reward": 4.449999809265137, + "reward_std": 0.06233753263950348, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 212, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 371.375, + "epoch": 0.3641025641025641, + "grad_norm": 0.9541327328256634, + "kl": 0.038970947265625, + "learning_rate": 3.547411308358238e-07, + "loss": 0.0016, + "reward": 4.477083206176758, + "reward_std": 0.04583332687616348, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 213, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.625, + "epoch": 0.3658119658119658, + "grad_norm": 0.9048739923450602, + "kl": 0.041656494140625, + "learning_rate": 3.53520577507003e-07, + "loss": 0.0017, + "reward": 4.481249809265137, + "reward_std": 0.03750000521540642, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 214, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.1875, + "epoch": 0.36752136752136755, + "grad_norm": 0.7146024062835775, + "kl": 0.03594970703125, + "learning_rate": 3.522970387016919e-07, + "loss": 0.0014, + "reward": 4.4739580154418945, + "reward_std": 0.05208335071802139, + "rewards/accuracy_reward": 2.9739584922790527, + "rewards/format_reward": 1.0, + "step": 215, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.3125, + "epoch": 0.36923076923076925, + "grad_norm": 0.7588941964472184, + "kl": 0.0311737060546875, + "learning_rate": 3.510705497060762e-07, + "loss": 0.0012, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 216, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.0625, + "epoch": 0.37094017094017095, + "grad_norm": 1.0066377974377885, + "kl": 0.040618896484375, + "learning_rate": 3.498411458914238e-07, + "loss": 0.0016, + "reward": 4.4822916984558105, + "reward_std": 0.03541666269302368, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 217, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 372.46875, + "epoch": 0.37264957264957266, + "grad_norm": 1.0197795186320238, + "kl": 0.040863037109375, + "learning_rate": 3.4860886271306425e-07, + "loss": 0.0016, + "reward": 4.327083587646484, + "reward_std": 0.310569703578949, + "rewards/accuracy_reward": 2.8645832538604736, + "rewards/format_reward": 0.96875, + "step": 218, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.0, + "epoch": 0.37435897435897436, + "grad_norm": 0.9536918082343873, + "kl": 0.037261962890625, + "learning_rate": 3.473737357093662e-07, + "loss": 0.0015, + "reward": 4.4739580154418945, + "reward_std": 0.03916243463754654, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 219, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.5, + "epoch": 0.37606837606837606, + "grad_norm": 1.357829764444905, + "kl": 0.040740966796875, + "learning_rate": 3.4613580050071274e-07, + "loss": 0.0016, + "reward": 4.414583206176758, + "reward_std": 0.1612565666437149, + "rewards/accuracy_reward": 2.9208333492279053, + "rewards/format_reward": 1.0, + "step": 220, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.5625, + "epoch": 0.37777777777777777, + "grad_norm": 0.7225751289221025, + "kl": 0.04107666015625, + "learning_rate": 3.4489509278847413e-07, + "loss": 0.0016, + "reward": 4.478124618530273, + "reward_std": 0.04375000670552254, + "rewards/accuracy_reward": 2.984375238418579, + "rewards/format_reward": 1.0, + "step": 221, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.90625, + "epoch": 0.37948717948717947, + "grad_norm": 0.5205268133629182, + "kl": 0.0357666015625, + "learning_rate": 3.43651648353978e-07, + "loss": 0.0014, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 222, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.9375, + "epoch": 0.3811965811965812, + "grad_norm": 0.11832902053787207, + "kl": 0.03363037109375, + "learning_rate": 3.4240550305747775e-07, + "loss": 0.0013, + "reward": 4.375, + "reward_std": 0.0, + "rewards/accuracy_reward": 2.875, + "rewards/format_reward": 1.0, + "step": 223, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.5625, + "epoch": 0.38290598290598293, + "grad_norm": 1.2289186408610433, + "kl": 0.03338623046875, + "learning_rate": 3.411566928371179e-07, + "loss": 0.0013, + "reward": 4.476041793823242, + "reward_std": 0.0479167103767395, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 224, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.46875, + "epoch": 0.38461538461538464, + "grad_norm": 0.833083778353936, + "kl": 0.035369873046875, + "learning_rate": 3.399052537078979e-07, + "loss": 0.0014, + "reward": 4.471875190734863, + "reward_std": 0.044946372509002686, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 225, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.46875, + "epoch": 0.38632478632478634, + "grad_norm": 0.8212684084277662, + "kl": 0.04425048828125, + "learning_rate": 3.3865122176063385e-07, + "loss": 0.0018, + "reward": 4.4822916984558105, + "reward_std": 0.025839870795607567, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 226, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.5625, + "epoch": 0.38803418803418804, + "grad_norm": 2.281771763704506, + "kl": 0.039581298828125, + "learning_rate": 3.3739463316091694e-07, + "loss": 0.0016, + "reward": 4.445833206176758, + "reward_std": 0.10833341628313065, + "rewards/accuracy_reward": 2.952083110809326, + "rewards/format_reward": 1.0, + "step": 227, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 384.1875, + "epoch": 0.38974358974358975, + "grad_norm": 1.1279986015302106, + "kl": 0.039031982421875, + "learning_rate": 3.361355241480709e-07, + "loss": 0.0016, + "reward": 4.329166412353516, + "reward_std": 0.25882017612457275, + "rewards/accuracy_reward": 2.879166603088379, + "rewards/format_reward": 1.0, + "step": 228, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.59375, + "epoch": 0.39145299145299145, + "grad_norm": 1.2133644943426865, + "kl": 0.04425048828125, + "learning_rate": 3.348739310341068e-07, + "loss": 0.0018, + "reward": 4.440625190734863, + "reward_std": 0.11132083833217621, + "rewards/accuracy_reward": 2.9468748569488525, + "rewards/format_reward": 1.0, + "step": 229, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.53125, + "epoch": 0.39316239316239315, + "grad_norm": 0.9498275646544599, + "kl": 0.03912353515625, + "learning_rate": 3.3360989020267577e-07, + "loss": 0.0016, + "reward": 4.480208396911621, + "reward_std": 0.03958336263895035, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 230, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.4375, + "epoch": 0.39487179487179486, + "grad_norm": 0.5639041842584143, + "kl": 0.04180908203125, + "learning_rate": 3.323434381080199e-07, + "loss": 0.0017, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 231, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.4375, + "epoch": 0.39658119658119656, + "grad_norm": 1.0578332982670637, + "kl": 0.038970947265625, + "learning_rate": 3.3107461127392067e-07, + "loss": 0.0016, + "reward": 4.419791221618652, + "reward_std": 0.12802158296108246, + "rewards/accuracy_reward": 2.926041603088379, + "rewards/format_reward": 1.0, + "step": 232, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.9375, + "epoch": 0.39829059829059826, + "grad_norm": 1.1144973425665088, + "kl": 0.04278564453125, + "learning_rate": 3.2980344629264576e-07, + "loss": 0.0017, + "reward": 4.452083110809326, + "reward_std": 0.06614777445793152, + "rewards/accuracy_reward": 2.9520833492279053, + "rewards/format_reward": 1.0, + "step": 233, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.78125, + "epoch": 0.4, + "grad_norm": 0.6621292037188405, + "kl": 0.04400634765625, + "learning_rate": 3.285299798238937e-07, + "loss": 0.0018, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 234, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.78125, + "epoch": 0.4017094017094017, + "grad_norm": 1.0939550473590092, + "kl": 0.0455322265625, + "learning_rate": 3.272542485937368e-07, + "loss": 0.0018, + "reward": 4.474999904632568, + "reward_std": 0.05000005289912224, + "rewards/accuracy_reward": 2.9750001430511475, + "rewards/format_reward": 1.0, + "step": 235, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.0625, + "epoch": 0.40341880341880343, + "grad_norm": 1.116721303420348, + "kl": 0.04193115234375, + "learning_rate": 3.259762893935617e-07, + "loss": 0.0017, + "reward": 4.453125, + "reward_std": 0.05560123175382614, + "rewards/accuracy_reward": 2.9593749046325684, + "rewards/format_reward": 1.0, + "step": 236, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.40625, + "epoch": 0.40512820512820513, + "grad_norm": 0.7336037527089339, + "kl": 0.04364013671875, + "learning_rate": 3.2469613907900844e-07, + "loss": 0.0017, + "reward": 4.484375, + "reward_std": 0.031250037252902985, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 237, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.40625, + "epoch": 0.40683760683760684, + "grad_norm": 1.0249556634834451, + "kl": 0.041595458984375, + "learning_rate": 3.234138345689077e-07, + "loss": 0.0017, + "reward": 4.479166507720947, + "reward_std": 0.04166674613952637, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 238, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.96875, + "epoch": 0.40854700854700854, + "grad_norm": 1.0297579044404537, + "kl": 0.038970947265625, + "learning_rate": 3.221294128442159e-07, + "loss": 0.0016, + "reward": 4.4864583015441895, + "reward_std": 0.027083396911621094, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 239, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.65625, + "epoch": 0.41025641025641024, + "grad_norm": 1.1366054973046644, + "kl": 0.037109375, + "learning_rate": 3.2084291094694877e-07, + "loss": 0.0015, + "reward": 4.471874713897705, + "reward_std": 0.05625009536743164, + "rewards/accuracy_reward": 2.971874952316284, + "rewards/format_reward": 1.0, + "step": 240, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.03125, + "epoch": 0.41196581196581195, + "grad_norm": 0.816127496687785, + "kl": 0.035430908203125, + "learning_rate": 3.1955436597911315e-07, + "loss": 0.0014, + "reward": 4.490624904632568, + "reward_std": 0.018750011920928955, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 241, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.65625, + "epoch": 0.41367521367521365, + "grad_norm": 0.9177988741903882, + "kl": 0.0390625, + "learning_rate": 3.182638151016369e-07, + "loss": 0.0016, + "reward": 4.46875, + "reward_std": 0.042391255497932434, + "rewards/accuracy_reward": 2.968749761581421, + "rewards/format_reward": 1.0, + "step": 242, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.0625, + "epoch": 0.4153846153846154, + "grad_norm": 0.09569801333724084, + "kl": 0.034759521484375, + "learning_rate": 3.16971295533297e-07, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 243, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.875, + "epoch": 0.4170940170940171, + "grad_norm": 1.056398886307418, + "kl": 0.03924560546875, + "learning_rate": 3.1567684454964674e-07, + "loss": 0.0016, + "reward": 4.472916603088379, + "reward_std": 0.045361533761024475, + "rewards/accuracy_reward": 2.972916841506958, + "rewards/format_reward": 1.0, + "step": 244, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.15625, + "epoch": 0.4188034188034188, + "grad_norm": 0.9602785771876751, + "kl": 0.045501708984375, + "learning_rate": 3.1438049948194e-07, + "loss": 0.0018, + "reward": 4.476041793823242, + "reward_std": 0.03833986073732376, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 245, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.71875, + "epoch": 0.4205128205128205, + "grad_norm": 0.9869505927258133, + "kl": 0.04656982421875, + "learning_rate": 3.130822977160554e-07, + "loss": 0.0019, + "reward": 4.452083110809326, + "reward_std": 0.0803452804684639, + "rewards/accuracy_reward": 2.9583332538604736, + "rewards/format_reward": 1.0, + "step": 246, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.5625, + "epoch": 0.4222222222222222, + "grad_norm": 0.5694726965482299, + "kl": 0.037200927734375, + "learning_rate": 3.117822766914174e-07, + "loss": 0.0015, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 247, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.5, + "epoch": 0.4239316239316239, + "grad_norm": 0.9716575381117358, + "kl": 0.038238525390625, + "learning_rate": 3.104804738999169e-07, + "loss": 0.0015, + "reward": 4.351041793823242, + "reward_std": 0.0479167103767395, + "rewards/accuracy_reward": 2.851041793823242, + "rewards/format_reward": 1.0, + "step": 248, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.1875, + "epoch": 0.4256410256410256, + "grad_norm": 0.9842518166866518, + "kl": 0.0413818359375, + "learning_rate": 3.091769268848302e-07, + "loss": 0.0017, + "reward": 4.316666603088379, + "reward_std": 0.10375991463661194, + "rewards/accuracy_reward": 2.8229165077209473, + "rewards/format_reward": 1.0, + "step": 249, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.65625, + "epoch": 0.42735042735042733, + "grad_norm": 1.278001712618916, + "kl": 0.042877197265625, + "learning_rate": 3.078716732397358e-07, + "loss": 0.0017, + "reward": 4.3885416984558105, + "reward_std": 0.20983925461769104, + "rewards/accuracy_reward": 2.894791603088379, + "rewards/format_reward": 1.0, + "step": 250, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.0, + "epoch": 0.42905982905982903, + "grad_norm": 1.1688274691177465, + "kl": 0.050018310546875, + "learning_rate": 3.065647506074306e-07, + "loss": 0.002, + "reward": 4.448958396911621, + "reward_std": 0.08484545350074768, + "rewards/accuracy_reward": 2.9552083015441895, + "rewards/format_reward": 1.0, + "step": 251, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.65625, + "epoch": 0.4307692307692308, + "grad_norm": 0.9872122914899957, + "kl": 0.04180908203125, + "learning_rate": 3.0525619667884406e-07, + "loss": 0.0017, + "reward": 4.469791412353516, + "reward_std": 0.048212792724370956, + "rewards/accuracy_reward": 2.976041555404663, + "rewards/format_reward": 1.0, + "step": 252, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.5625, + "epoch": 0.4324786324786325, + "grad_norm": 0.7672081800546268, + "kl": 0.046142578125, + "learning_rate": 3.0394604919195157e-07, + "loss": 0.0018, + "reward": 4.317708492279053, + "reward_std": 0.08067251741886139, + "rewards/accuracy_reward": 2.8177082538604736, + "rewards/format_reward": 1.0, + "step": 253, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.875, + "epoch": 0.4341880341880342, + "grad_norm": 0.9698339884690548, + "kl": 0.0462646484375, + "learning_rate": 3.026343459306856e-07, + "loss": 0.0019, + "reward": 4.462500095367432, + "reward_std": 0.07499998807907104, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 254, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.8125, + "epoch": 0.4358974358974359, + "grad_norm": 0.7482162088172302, + "kl": 0.040283203125, + "learning_rate": 3.0132112472384646e-07, + "loss": 0.0016, + "reward": 4.484375, + "reward_std": 0.024467820301651955, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 255, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.21875, + "epoch": 0.4376068376068376, + "grad_norm": 0.5797544385226397, + "kl": 0.03729248046875, + "learning_rate": 3.000064234440111e-07, + "loss": 0.0015, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 256, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.59375, + "epoch": 0.4393162393162393, + "grad_norm": 0.5365371682569317, + "kl": 0.039337158203125, + "learning_rate": 2.98690280006441e-07, + "loss": 0.0016, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 257, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.90625, + "epoch": 0.441025641025641, + "grad_norm": 1.2492723527722578, + "kl": 0.037811279296875, + "learning_rate": 2.973727323679887e-07, + "loss": 0.0015, + "reward": 4.382291793823242, + "reward_std": 0.22722449898719788, + "rewards/accuracy_reward": 2.894791841506958, + "rewards/format_reward": 1.0, + "step": 258, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.5, + "epoch": 0.4427350427350427, + "grad_norm": 1.26066913404296, + "kl": 0.04364013671875, + "learning_rate": 2.9605381852600284e-07, + "loss": 0.0017, + "reward": 4.379166603088379, + "reward_std": 0.2093537300825119, + "rewards/accuracy_reward": 2.8916664123535156, + "rewards/format_reward": 1.0, + "step": 259, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.03125, + "epoch": 0.4444444444444444, + "grad_norm": 0.7554523113224114, + "kl": 0.03900146484375, + "learning_rate": 2.947335765172332e-07, + "loss": 0.0016, + "reward": 4.483333587646484, + "reward_std": 0.03333333879709244, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 260, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.875, + "epoch": 0.4461538461538462, + "grad_norm": 0.9144306336964317, + "kl": 0.0413818359375, + "learning_rate": 2.934120444167326e-07, + "loss": 0.0017, + "reward": 4.447916507720947, + "reward_std": 0.08289644867181778, + "rewards/accuracy_reward": 2.9541666507720947, + "rewards/format_reward": 1.0, + "step": 261, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.15625, + "epoch": 0.4478632478632479, + "grad_norm": 1.2006861780363822, + "kl": 0.0380859375, + "learning_rate": 2.920892603367596e-07, + "loss": 0.0015, + "reward": 4.485416412353516, + "reward_std": 0.029166698455810547, + "rewards/accuracy_reward": 2.991666793823242, + "rewards/format_reward": 1.0, + "step": 262, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.375, + "epoch": 0.4495726495726496, + "grad_norm": 0.6264774336128276, + "kl": 0.038055419921875, + "learning_rate": 2.9076526242567934e-07, + "loss": 0.0015, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 263, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.03125, + "epoch": 0.4512820512820513, + "grad_norm": 0.9285367833144407, + "kl": 0.03717041015625, + "learning_rate": 2.894400888668628e-07, + "loss": 0.0015, + "reward": 4.460416793823242, + "reward_std": 0.0791667103767395, + "rewards/accuracy_reward": 2.991666555404663, + "rewards/format_reward": 0.96875, + "step": 264, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.875, + "epoch": 0.452991452991453, + "grad_norm": 0.7830763693612015, + "kl": 0.044525146484375, + "learning_rate": 2.881137778775863e-07, + "loss": 0.0018, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 265, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.40625, + "epoch": 0.4547008547008547, + "grad_norm": 0.9316521232276594, + "kl": 0.037506103515625, + "learning_rate": 2.86786367707929e-07, + "loss": 0.0015, + "reward": 4.480208396911621, + "reward_std": 0.03000655397772789, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 266, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.96875, + "epoch": 0.4564102564102564, + "grad_norm": 0.9675437072980656, + "kl": 0.036773681640625, + "learning_rate": 2.854578966396697e-07, + "loss": 0.0015, + "reward": 4.464583396911621, + "reward_std": 0.057912446558475494, + "rewards/accuracy_reward": 2.9708333015441895, + "rewards/format_reward": 1.0, + "step": 267, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.59375, + "epoch": 0.4581196581196581, + "grad_norm": 1.1220859220665398, + "kl": 0.03741455078125, + "learning_rate": 2.841284029851829e-07, + "loss": 0.0015, + "reward": 4.4666666984558105, + "reward_std": 0.057089872658252716, + "rewards/accuracy_reward": 2.972916603088379, + "rewards/format_reward": 1.0, + "step": 268, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.875, + "epoch": 0.4598290598290598, + "grad_norm": 0.5689642749646727, + "kl": 0.03631591796875, + "learning_rate": 2.827979250863341e-07, + "loss": 0.0015, + "reward": 4.493750095367432, + "reward_std": 0.012500028125941753, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 269, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.1875, + "epoch": 0.46153846153846156, + "grad_norm": 0.9826672817363714, + "kl": 0.0389404296875, + "learning_rate": 2.814665013133737e-07, + "loss": 0.0016, + "reward": 4.476041793823242, + "reward_std": 0.03833986073732376, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 270, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.8125, + "epoch": 0.46324786324786327, + "grad_norm": 0.8299425945656934, + "kl": 0.039581298828125, + "learning_rate": 2.801341700638307e-07, + "loss": 0.0016, + "reward": 4.465624809265137, + "reward_std": 0.06875002384185791, + "rewards/accuracy_reward": 2.965625047683716, + "rewards/format_reward": 1.0, + "step": 271, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.875, + "epoch": 0.46495726495726497, + "grad_norm": 0.8021355694486639, + "kl": 0.03424072265625, + "learning_rate": 2.788009697614053e-07, + "loss": 0.0014, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 272, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.96875, + "epoch": 0.4666666666666667, + "grad_norm": 0.5179000391744707, + "kl": 0.038330078125, + "learning_rate": 2.774669388548604e-07, + "loss": 0.0015, + "reward": 4.487500190734863, + "reward_std": 0.014433743432164192, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 273, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 362.53125, + "epoch": 0.4683760683760684, + "grad_norm": 0.15075527427477844, + "kl": 0.035400390625, + "learning_rate": 2.761321158169134e-07, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 274, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.625, + "epoch": 0.4700854700854701, + "grad_norm": 0.6023840972420389, + "kl": 0.04217529296875, + "learning_rate": 2.7479653914312606e-07, + "loss": 0.0017, + "reward": 4.487500190734863, + "reward_std": 0.014433743432164192, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 275, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 365.34375, + "epoch": 0.4717948717948718, + "grad_norm": 1.0588371671437944, + "kl": 0.034088134765625, + "learning_rate": 2.7346024735079484e-07, + "loss": 0.0014, + "reward": 4.4666666984558105, + "reward_std": 0.06666667759418488, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 276, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.875, + "epoch": 0.4735042735042735, + "grad_norm": 0.9311754494662231, + "kl": 0.0364990234375, + "learning_rate": 2.721232789778396e-07, + "loss": 0.0015, + "reward": 4.46875, + "reward_std": 0.06250005960464478, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 277, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.6875, + "epoch": 0.4752136752136752, + "grad_norm": 0.7622217392129294, + "kl": 0.036468505859375, + "learning_rate": 2.707856725816926e-07, + "loss": 0.0015, + "reward": 4.4822916984558105, + "reward_std": 0.024850429967045784, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 278, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.21875, + "epoch": 0.47692307692307695, + "grad_norm": 0.07605318199841635, + "kl": 0.036376953125, + "learning_rate": 2.694474667381862e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 279, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.40625, + "epoch": 0.47863247863247865, + "grad_norm": 0.08281321697663974, + "kl": 0.031402587890625, + "learning_rate": 2.681087000404406e-07, + "loss": 0.0013, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 280, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.9375, + "epoch": 0.48034188034188036, + "grad_norm": 0.9158024249719325, + "kl": 0.035797119140625, + "learning_rate": 2.667694110977506e-07, + "loss": 0.0014, + "reward": 4.428124904632568, + "reward_std": 0.14375001192092896, + "rewards/accuracy_reward": 2.9281249046325684, + "rewards/format_reward": 1.0, + "step": 281, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.875, + "epoch": 0.48205128205128206, + "grad_norm": 1.2380224918264358, + "kl": 0.035797119140625, + "learning_rate": 2.6542963853447236e-07, + "loss": 0.0014, + "reward": 4.409374713897705, + "reward_std": 0.18125000596046448, + "rewards/accuracy_reward": 2.921875238418579, + "rewards/format_reward": 1.0, + "step": 282, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.0, + "epoch": 0.48376068376068376, + "grad_norm": 0.8805302625019681, + "kl": 0.039886474609375, + "learning_rate": 2.6408942098890937e-07, + "loss": 0.0016, + "reward": 4.478125095367432, + "reward_std": 0.04375004768371582, + "rewards/accuracy_reward": 2.9781250953674316, + "rewards/format_reward": 1.0, + "step": 283, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.84375, + "epoch": 0.48547008547008547, + "grad_norm": 0.7527809954073914, + "kl": 0.03570556640625, + "learning_rate": 2.627487971121981e-07, + "loss": 0.0014, + "reward": 4.4739580154418945, + "reward_std": 0.03987946733832359, + "rewards/accuracy_reward": 2.9739582538604736, + "rewards/format_reward": 1.0, + "step": 284, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.5625, + "epoch": 0.48717948717948717, + "grad_norm": 0.09612785925134994, + "kl": 0.037109375, + "learning_rate": 2.6140780556719347e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 285, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 371.9375, + "epoch": 0.4888888888888889, + "grad_norm": 1.0426281181628902, + "kl": 0.037139892578125, + "learning_rate": 2.600664850273538e-07, + "loss": 0.0015, + "reward": 4.431250095367432, + "reward_std": 0.12285532057285309, + "rewards/accuracy_reward": 2.981250047683716, + "rewards/format_reward": 0.96875, + "step": 286, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.6875, + "epoch": 0.4905982905982906, + "grad_norm": 0.6014376687211012, + "kl": 0.031768798828125, + "learning_rate": 2.5872487417562527e-07, + "loss": 0.0013, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 287, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.0625, + "epoch": 0.49230769230769234, + "grad_norm": 0.7733776094186604, + "kl": 0.03582763671875, + "learning_rate": 2.573830117033266e-07, + "loss": 0.0014, + "reward": 4.478125095367432, + "reward_std": 0.043749988079071045, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 288, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.71875, + "epoch": 0.49401709401709404, + "grad_norm": 0.7138068342578241, + "kl": 0.035797119140625, + "learning_rate": 2.5604093630903305e-07, + "loss": 0.0014, + "reward": 4.479166507720947, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 289, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.5625, + "epoch": 0.49572649572649574, + "grad_norm": 0.7761774380810611, + "kl": 0.031585693359375, + "learning_rate": 2.546986866974606e-07, + "loss": 0.0013, + "reward": 4.363541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.8697915077209473, + "rewards/format_reward": 1.0, + "step": 290, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.78125, + "epoch": 0.49743589743589745, + "grad_norm": 0.8551880159617881, + "kl": 0.0345458984375, + "learning_rate": 2.5335630157834935e-07, + "loss": 0.0014, + "reward": 4.463541507720947, + "reward_std": 0.05353529751300812, + "rewards/accuracy_reward": 2.9697914123535156, + "rewards/format_reward": 1.0, + "step": 291, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.09375, + "epoch": 0.49914529914529915, + "grad_norm": 0.5529263640378508, + "kl": 0.03204345703125, + "learning_rate": 2.520138196653475e-07, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 292, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.53125, + "epoch": 0.5008547008547009, + "grad_norm": 0.9791061277473894, + "kl": 0.032501220703125, + "learning_rate": 2.506712796748946e-07, + "loss": 0.0013, + "reward": 4.474999904632568, + "reward_std": 0.03943371772766113, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 293, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.0, + "epoch": 0.5025641025641026, + "grad_norm": 0.8579863225514387, + "kl": 0.044525146484375, + "learning_rate": 2.4932872032510537e-07, + "loss": 0.0018, + "reward": 4.453125, + "reward_std": 0.09375003725290298, + "rewards/accuracy_reward": 2.953125, + "rewards/format_reward": 1.0, + "step": 294, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.8125, + "epoch": 0.5042735042735043, + "grad_norm": 0.8092357571514325, + "kl": 0.0313720703125, + "learning_rate": 2.4798618033465255e-07, + "loss": 0.0013, + "reward": 4.478125095367432, + "reward_std": 0.03244640305638313, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 295, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 365.84375, + "epoch": 0.505982905982906, + "grad_norm": 0.47095499450394307, + "kl": 0.03460693359375, + "learning_rate": 2.466436984216507e-07, + "loss": 0.0014, + "reward": 4.478125095367432, + "reward_std": 0.029536345973610878, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 296, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.0, + "epoch": 0.5076923076923077, + "grad_norm": 0.531091931616097, + "kl": 0.02972412109375, + "learning_rate": 2.453013133025394e-07, + "loss": 0.0012, + "reward": 4.484375, + "reward_std": 0.019946394488215446, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 297, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 365.8125, + "epoch": 0.5094017094017094, + "grad_norm": 1.2659627324757727, + "kl": 0.0391845703125, + "learning_rate": 2.43959063690967e-07, + "loss": 0.0016, + "reward": 4.417708396911621, + "reward_std": 0.15726101398468018, + "rewards/accuracy_reward": 2.930208206176758, + "rewards/format_reward": 1.0, + "step": 298, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.9375, + "epoch": 0.5111111111111111, + "grad_norm": 0.7314429482901976, + "kl": 0.042449951171875, + "learning_rate": 2.4261698829667347e-07, + "loss": 0.0017, + "reward": 4.479166507720947, + "reward_std": 0.041666705161333084, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 299, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.125, + "epoch": 0.5128205128205128, + "grad_norm": 0.09120795369226402, + "kl": 0.029754638671875, + "learning_rate": 2.412751258243748e-07, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 300, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.90625, + "epoch": 0.5145299145299145, + "grad_norm": 0.7868332165703579, + "kl": 0.0433349609375, + "learning_rate": 2.399335149726463e-07, + "loss": 0.0017, + "reward": 4.4822916984558105, + "reward_std": 0.025839831680059433, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 301, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.0, + "epoch": 0.5162393162393163, + "grad_norm": 0.763574402727439, + "kl": 0.03216552734375, + "learning_rate": 2.3859219443280656e-07, + "loss": 0.0013, + "reward": 4.490624904632568, + "reward_std": 0.018750011920928955, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 302, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.1875, + "epoch": 0.517948717948718, + "grad_norm": 1.0058943562387195, + "kl": 0.03338623046875, + "learning_rate": 2.3725120288780184e-07, + "loss": 0.0013, + "reward": 4.425000190734863, + "reward_std": 0.1499999761581421, + "rewards/accuracy_reward": 2.9375, + "rewards/format_reward": 1.0, + "step": 303, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 428.5625, + "epoch": 0.5196581196581197, + "grad_norm": 0.42116405277150404, + "kl": 0.03240966796875, + "learning_rate": 2.359105790110906e-07, + "loss": 0.0013, + "reward": 4.488541603088379, + "reward_std": 0.013339842669665813, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 304, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.34375, + "epoch": 0.5213675213675214, + "grad_norm": 1.0950683758577704, + "kl": 0.0372314453125, + "learning_rate": 2.3457036146552762e-07, + "loss": 0.0015, + "reward": 4.431250095367432, + "reward_std": 0.11542317271232605, + "rewards/accuracy_reward": 2.9437499046325684, + "rewards/format_reward": 1.0, + "step": 305, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.09375, + "epoch": 0.5230769230769231, + "grad_norm": 0.7371614253995173, + "kl": 0.03411865234375, + "learning_rate": 2.3323058890224938e-07, + "loss": 0.0014, + "reward": 4.432291507720947, + "reward_std": 0.1354166865348816, + "rewards/accuracy_reward": 2.9322915077209473, + "rewards/format_reward": 1.0, + "step": 306, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.71875, + "epoch": 0.5247863247863248, + "grad_norm": 0.5134694091227199, + "kl": 0.032379150390625, + "learning_rate": 2.3189129995955942e-07, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 307, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.28125, + "epoch": 0.5264957264957265, + "grad_norm": 0.8781524864900261, + "kl": 0.038330078125, + "learning_rate": 2.305525332618138e-07, + "loss": 0.0015, + "reward": 4.482291221618652, + "reward_std": 0.03541666269302368, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 308, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.0, + "epoch": 0.5282051282051282, + "grad_norm": 1.0415017947017724, + "kl": 0.033416748046875, + "learning_rate": 2.292143274183074e-07, + "loss": 0.0013, + "reward": 4.460416793823242, + "reward_std": 0.0791667103767395, + "rewards/accuracy_reward": 2.960416793823242, + "rewards/format_reward": 1.0, + "step": 309, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.875, + "epoch": 0.5299145299145299, + "grad_norm": 0.6008838011788349, + "kl": 0.03973388671875, + "learning_rate": 2.278767210221604e-07, + "loss": 0.0016, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 310, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.71875, + "epoch": 0.5316239316239316, + "grad_norm": 0.732823233771545, + "kl": 0.04248046875, + "learning_rate": 2.265397526492052e-07, + "loss": 0.0017, + "reward": 4.4583330154418945, + "reward_std": 0.08333335071802139, + "rewards/accuracy_reward": 2.9583334922790527, + "rewards/format_reward": 1.0, + "step": 311, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 403.90625, + "epoch": 0.5333333333333333, + "grad_norm": 0.8066497268246617, + "kl": 0.037872314453125, + "learning_rate": 2.2520346085687397e-07, + "loss": 0.0015, + "reward": 4.351041793823242, + "reward_std": 0.22291666269302368, + "rewards/accuracy_reward": 2.9010415077209473, + "rewards/format_reward": 1.0, + "step": 312, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 405.625, + "epoch": 0.535042735042735, + "grad_norm": 0.5736301292855377, + "kl": 0.033050537109375, + "learning_rate": 2.2386788418308665e-07, + "loss": 0.0013, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 313, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 402.34375, + "epoch": 0.5367521367521367, + "grad_norm": 0.92886366537301, + "kl": 0.04010009765625, + "learning_rate": 2.225330611451396e-07, + "loss": 0.0016, + "reward": 4.475000381469727, + "reward_std": 0.050000011920928955, + "rewards/accuracy_reward": 2.981250047683716, + "rewards/format_reward": 1.0, + "step": 314, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 424.21875, + "epoch": 0.5384615384615384, + "grad_norm": 0.8095666212677034, + "kl": 0.035491943359375, + "learning_rate": 2.2119903023859473e-07, + "loss": 0.0014, + "reward": 4.462500095367432, + "reward_std": 0.07500001788139343, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 315, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 396.75, + "epoch": 0.5401709401709401, + "grad_norm": 0.9069226760525908, + "kl": 0.033599853515625, + "learning_rate": 2.1986582993616925e-07, + "loss": 0.0013, + "reward": 4.3458333015441895, + "reward_std": 0.23333333432674408, + "rewards/accuracy_reward": 2.8958334922790527, + "rewards/format_reward": 1.0, + "step": 316, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.4375, + "epoch": 0.5418803418803418, + "grad_norm": 0.08823704142856094, + "kl": 0.03680419921875, + "learning_rate": 2.1853349868662632e-07, + "loss": 0.0015, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 317, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.65625, + "epoch": 0.5435897435897435, + "grad_norm": 0.731770212008259, + "kl": 0.039154052734375, + "learning_rate": 2.1720207491366595e-07, + "loss": 0.0016, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 318, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 406.25, + "epoch": 0.5452991452991452, + "grad_norm": 0.12136352666188718, + "kl": 0.04296875, + "learning_rate": 2.1587159701481713e-07, + "loss": 0.0017, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 319, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.125, + "epoch": 0.5470085470085471, + "grad_norm": 0.5122046662863808, + "kl": 0.039886474609375, + "learning_rate": 2.1454210336033038e-07, + "loss": 0.0016, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 320, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 473.65625, + "epoch": 0.5487179487179488, + "grad_norm": 0.7720865025563447, + "kl": 0.037506103515625, + "learning_rate": 2.1321363229207094e-07, + "loss": 0.0015, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.987499952316284, + "rewards/format_reward": 1.0, + "step": 321, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.46875, + "epoch": 0.5504273504273505, + "grad_norm": 0.7221011874275695, + "kl": 0.039031982421875, + "learning_rate": 2.1188622212241363e-07, + "loss": 0.0016, + "reward": 4.4375, + "reward_std": 0.125, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 0.96875, + "step": 322, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 406.4375, + "epoch": 0.5521367521367522, + "grad_norm": 0.540236345777918, + "kl": 0.03485107421875, + "learning_rate": 2.1055991113313716e-07, + "loss": 0.0014, + "reward": 4.365624904632568, + "reward_std": 0.018750011920928955, + "rewards/accuracy_reward": 2.8656249046325684, + "rewards/format_reward": 1.0, + "step": 323, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.84375, + "epoch": 0.5538461538461539, + "grad_norm": 0.11607538632372809, + "kl": 0.03997802734375, + "learning_rate": 2.092347375743207e-07, + "loss": 0.0016, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 324, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.03125, + "epoch": 0.5555555555555556, + "grad_norm": 0.5056705769715667, + "kl": 0.039154052734375, + "learning_rate": 2.0791073966324034e-07, + "loss": 0.0016, + "reward": 4.307291507720947, + "reward_std": 0.1354166716337204, + "rewards/accuracy_reward": 2.8072915077209473, + "rewards/format_reward": 1.0, + "step": 325, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.375, + "epoch": 0.5572649572649573, + "grad_norm": 0.9995739010662272, + "kl": 0.03607177734375, + "learning_rate": 2.065879555832674e-07, + "loss": 0.0014, + "reward": 4.459374904632568, + "reward_std": 0.07446783035993576, + "rewards/accuracy_reward": 2.965625047683716, + "rewards/format_reward": 1.0, + "step": 326, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.03125, + "epoch": 0.558974358974359, + "grad_norm": 0.5593797794480546, + "kl": 0.037933349609375, + "learning_rate": 2.052664234827668e-07, + "loss": 0.0015, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 327, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.40625, + "epoch": 0.5606837606837607, + "grad_norm": 0.7643758219950748, + "kl": 0.0341796875, + "learning_rate": 2.039461814739971e-07, + "loss": 0.0014, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.9906249046325684, + "rewards/format_reward": 1.0, + "step": 328, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.84375, + "epoch": 0.5623931623931624, + "grad_norm": 0.5355887210870947, + "kl": 0.0362548828125, + "learning_rate": 2.0262726763201138e-07, + "loss": 0.0015, + "reward": 4.371874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.871875047683716, + "rewards/format_reward": 1.0, + "step": 329, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.9375, + "epoch": 0.5641025641025641, + "grad_norm": 0.08255482149118182, + "kl": 0.030914306640625, + "learning_rate": 2.0130971999355901e-07, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 330, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.5, + "epoch": 0.5658119658119658, + "grad_norm": 0.4549853870360058, + "kl": 0.03265380859375, + "learning_rate": 1.9999357655598891e-07, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 331, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.1875, + "epoch": 0.5675213675213675, + "grad_norm": 1.0004535826942826, + "kl": 0.04522705078125, + "learning_rate": 1.9867887527615357e-07, + "loss": 0.0018, + "reward": 4.446874618530273, + "reward_std": 0.10625002533197403, + "rewards/accuracy_reward": 2.953125, + "rewards/format_reward": 1.0, + "step": 332, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.46875, + "epoch": 0.5692307692307692, + "grad_norm": 0.9598847700617524, + "kl": 0.03875732421875, + "learning_rate": 1.9736565406931444e-07, + "loss": 0.0015, + "reward": 4.418749809265137, + "reward_std": 0.16250000894069672, + "rewards/accuracy_reward": 2.9312500953674316, + "rewards/format_reward": 1.0, + "step": 333, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 390.09375, + "epoch": 0.5709401709401709, + "grad_norm": 0.5355616265414169, + "kl": 0.03277587890625, + "learning_rate": 1.960539508080485e-07, + "loss": 0.0013, + "reward": 4.356249809265137, + "reward_std": 0.19618761539459229, + "rewards/accuracy_reward": 2.90625, + "rewards/format_reward": 1.0, + "step": 334, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.625, + "epoch": 0.5726495726495726, + "grad_norm": 0.9278958288143269, + "kl": 0.033538818359375, + "learning_rate": 1.9474380332115597e-07, + "loss": 0.0013, + "reward": 4.477083206176758, + "reward_std": 0.045833367854356766, + "rewards/accuracy_reward": 2.9833333492279053, + "rewards/format_reward": 1.0, + "step": 335, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.78125, + "epoch": 0.5743589743589743, + "grad_norm": 0.7140882017996247, + "kl": 0.03717041015625, + "learning_rate": 1.934352493925695e-07, + "loss": 0.0015, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 336, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.28125, + "epoch": 0.576068376068376, + "grad_norm": 0.7140882017996247, + "kl": 0.03448486328125, + "learning_rate": 1.934352493925695e-07, + "loss": 0.0014, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 337, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.5, + "epoch": 0.5777777777777777, + "grad_norm": 0.8126061578453346, + "kl": 0.03192138671875, + "learning_rate": 1.9212832676026427e-07, + "loss": 0.0013, + "reward": 4.483333587646484, + "reward_std": 0.03333333879709244, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 338, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.96875, + "epoch": 0.5794871794871795, + "grad_norm": 0.7335705389742384, + "kl": 0.035125732421875, + "learning_rate": 1.9082307311516984e-07, + "loss": 0.0014, + "reward": 4.456250190734863, + "reward_std": 0.07693374156951904, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 339, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.4375, + "epoch": 0.5811965811965812, + "grad_norm": 0.7657935514777842, + "kl": 0.034881591796875, + "learning_rate": 1.895195261000831e-07, + "loss": 0.0014, + "reward": 4.484375, + "reward_std": 0.031250037252902985, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 340, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.71875, + "epoch": 0.582905982905983, + "grad_norm": 0.09547936317636307, + "kl": 0.034759521484375, + "learning_rate": 1.8821772330858257e-07, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 341, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.0625, + "epoch": 0.5846153846153846, + "grad_norm": 0.9453045319770538, + "kl": 0.03839111328125, + "learning_rate": 1.8691770228394454e-07, + "loss": 0.0015, + "reward": 4.477083206176758, + "reward_std": 0.04583340883255005, + "rewards/accuracy_reward": 2.977083206176758, + "rewards/format_reward": 1.0, + "step": 342, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.78125, + "epoch": 0.5863247863247864, + "grad_norm": 0.7927846448695134, + "kl": 0.03741455078125, + "learning_rate": 1.856195005180599e-07, + "loss": 0.0015, + "reward": 4.479166507720947, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 343, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.5625, + "epoch": 0.588034188034188, + "grad_norm": 0.523415665615348, + "kl": 0.0307769775390625, + "learning_rate": 1.8432315545035327e-07, + "loss": 0.0012, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 344, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 408.28125, + "epoch": 0.5897435897435898, + "grad_norm": 0.07967194840458601, + "kl": 0.030059814453125, + "learning_rate": 1.8302870446670298e-07, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 345, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.34375, + "epoch": 0.5914529914529915, + "grad_norm": 0.6739955777442904, + "kl": 0.032501220703125, + "learning_rate": 1.8173618489836313e-07, + "loss": 0.0013, + "reward": 4.452083110809326, + "reward_std": 0.04834301769733429, + "rewards/accuracy_reward": 2.9583334922790527, + "rewards/format_reward": 1.0, + "step": 346, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.21875, + "epoch": 0.5931623931623932, + "grad_norm": 0.9588346689039634, + "kl": 0.03369140625, + "learning_rate": 1.8044563402088682e-07, + "loss": 0.0013, + "reward": 4.463541507720947, + "reward_std": 0.04878613352775574, + "rewards/accuracy_reward": 2.9697914123535156, + "rewards/format_reward": 1.0, + "step": 347, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.78125, + "epoch": 0.5948717948717949, + "grad_norm": 0.91793025589706, + "kl": 0.0355224609375, + "learning_rate": 1.791570890530512e-07, + "loss": 0.0014, + "reward": 4.40625, + "reward_std": 0.1875000298023224, + "rewards/accuracy_reward": 2.90625, + "rewards/format_reward": 1.0, + "step": 348, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.9375, + "epoch": 0.5965811965811966, + "grad_norm": 0.5238665028970252, + "kl": 0.0275115966796875, + "learning_rate": 1.7787058715578413e-07, + "loss": 0.0011, + "reward": 4.479166507720947, + "reward_std": 0.029462780803442, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 349, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.625, + "epoch": 0.5982905982905983, + "grad_norm": 0.8802153456160818, + "kl": 0.0352783203125, + "learning_rate": 1.7658616543109234e-07, + "loss": 0.0014, + "reward": 4.485416889190674, + "reward_std": 0.029166698455810547, + "rewards/accuracy_reward": 2.991666793823242, + "rewards/format_reward": 1.0, + "step": 350, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.5, + "epoch": 0.6, + "grad_norm": 0.6080026072792618, + "kl": 0.033416748046875, + "learning_rate": 1.7530386092099156e-07, + "loss": 0.0013, + "reward": 4.478125095367432, + "reward_std": 0.025769436731934547, + "rewards/accuracy_reward": 2.9781248569488525, + "rewards/format_reward": 1.0, + "step": 351, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.84375, + "epoch": 0.6017094017094017, + "grad_norm": 0.6052317335576483, + "kl": 0.031951904296875, + "learning_rate": 1.7402371060643827e-07, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 352, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.46875, + "epoch": 0.6034188034188034, + "grad_norm": 0.5873898518946016, + "kl": 0.031982421875, + "learning_rate": 1.7274575140626315e-07, + "loss": 0.0013, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 353, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.5625, + "epoch": 0.6051282051282051, + "grad_norm": 0.534965438696097, + "kl": 0.03204345703125, + "learning_rate": 1.7147002017610623e-07, + "loss": 0.0013, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 354, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.125, + "epoch": 0.6068376068376068, + "grad_norm": 0.19326846885011464, + "kl": 0.034912109375, + "learning_rate": 1.7019655370735424e-07, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 355, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.78125, + "epoch": 0.6085470085470085, + "grad_norm": 1.2061684015115641, + "kl": 0.0396728515625, + "learning_rate": 1.6892538872607933e-07, + "loss": 0.0016, + "reward": 4.342708587646484, + "reward_std": 0.2912980318069458, + "rewards/accuracy_reward": 2.8489584922790527, + "rewards/format_reward": 1.0, + "step": 356, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.3125, + "epoch": 0.6102564102564103, + "grad_norm": 1.1222016670512132, + "kl": 0.03387451171875, + "learning_rate": 1.6765656189198011e-07, + "loss": 0.0014, + "reward": 4.426041603088379, + "reward_std": 0.14791671931743622, + "rewards/accuracy_reward": 2.926041603088379, + "rewards/format_reward": 1.0, + "step": 357, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.34375, + "epoch": 0.611965811965812, + "grad_norm": 0.9056732169837467, + "kl": 0.031890869140625, + "learning_rate": 1.6639010979732428e-07, + "loss": 0.0013, + "reward": 4.484375, + "reward_std": 0.031250059604644775, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 358, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.3125, + "epoch": 0.6136752136752137, + "grad_norm": 0.10457460621666292, + "kl": 0.0273895263671875, + "learning_rate": 1.6512606896589322e-07, + "loss": 0.0011, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 359, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.34375, + "epoch": 0.6153846153846154, + "grad_norm": 0.8589108015834329, + "kl": 0.03399658203125, + "learning_rate": 1.6386447585192908e-07, + "loss": 0.0014, + "reward": 4.476041793823242, + "reward_std": 0.03735045716166496, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 360, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.875, + "epoch": 0.6170940170940171, + "grad_norm": 0.5237910767837538, + "kl": 0.03424072265625, + "learning_rate": 1.6260536683908298e-07, + "loss": 0.0014, + "reward": 4.485416412353516, + "reward_std": 0.018162094056606293, + "rewards/accuracy_reward": 2.9854166507720947, + "rewards/format_reward": 1.0, + "step": 361, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.6875, + "epoch": 0.6188034188034188, + "grad_norm": 0.7206265941802336, + "kl": 0.0287017822265625, + "learning_rate": 1.6134877823936607e-07, + "loss": 0.0011, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 362, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 372.96875, + "epoch": 0.6205128205128205, + "grad_norm": 1.383167067518889, + "kl": 0.034393310546875, + "learning_rate": 1.6009474629210202e-07, + "loss": 0.0014, + "reward": 4.219791412353516, + "reward_std": 0.28802648186683655, + "rewards/accuracy_reward": 2.788541555404663, + "rewards/format_reward": 1.0, + "step": 363, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.59375, + "epoch": 0.6222222222222222, + "grad_norm": 1.1165705187064898, + "kl": 0.030120849609375, + "learning_rate": 1.5884330716288212e-07, + "loss": 0.0012, + "reward": 4.449999809265137, + "reward_std": 0.07795004546642303, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 364, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.1875, + "epoch": 0.6239316239316239, + "grad_norm": 0.7203681625804031, + "kl": 0.0330810546875, + "learning_rate": 1.5759449694252225e-07, + "loss": 0.0013, + "reward": 4.478125095367432, + "reward_std": 0.032446444034576416, + "rewards/accuracy_reward": 2.9781250953674316, + "rewards/format_reward": 1.0, + "step": 365, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.9375, + "epoch": 0.6256410256410256, + "grad_norm": 0.7514346669369655, + "kl": 0.035736083984375, + "learning_rate": 1.5634835164602196e-07, + "loss": 0.0014, + "reward": 4.457291603088379, + "reward_std": 0.08541668206453323, + "rewards/accuracy_reward": 2.9635415077209473, + "rewards/format_reward": 1.0, + "step": 366, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 364.0, + "epoch": 0.6273504273504273, + "grad_norm": 0.5135545227710894, + "kl": 0.0283050537109375, + "learning_rate": 1.551049072115259e-07, + "loss": 0.0011, + "reward": 4.481249809265137, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 367, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.1875, + "epoch": 0.629059829059829, + "grad_norm": 0.7832820562893558, + "kl": 0.035369873046875, + "learning_rate": 1.5386419949928732e-07, + "loss": 0.0014, + "reward": 4.464583396911621, + "reward_std": 0.07083334028720856, + "rewards/accuracy_reward": 2.964583396911621, + "rewards/format_reward": 1.0, + "step": 368, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.1875, + "epoch": 0.6307692307692307, + "grad_norm": 1.3554980343088359, + "kl": 0.03521728515625, + "learning_rate": 1.5262626429063382e-07, + "loss": 0.0014, + "reward": 4.3854169845581055, + "reward_std": 0.16212184727191925, + "rewards/accuracy_reward": 2.8916666507720947, + "rewards/format_reward": 1.0, + "step": 369, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.03125, + "epoch": 0.6324786324786325, + "grad_norm": 1.024651879366642, + "kl": 0.032989501953125, + "learning_rate": 1.5139113728693572e-07, + "loss": 0.0013, + "reward": 4.415624618530273, + "reward_std": 0.16875003278255463, + "rewards/accuracy_reward": 2.921875, + "rewards/format_reward": 1.0, + "step": 370, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.40625, + "epoch": 0.6341880341880342, + "grad_norm": 1.0957802284772251, + "kl": 0.0343017578125, + "learning_rate": 1.5015885410857614e-07, + "loss": 0.0014, + "reward": 4.446875095367432, + "reward_std": 0.10624998807907104, + "rewards/accuracy_reward": 2.953125, + "rewards/format_reward": 1.0, + "step": 371, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 366.65625, + "epoch": 0.6358974358974359, + "grad_norm": 0.8228003354495702, + "kl": 0.03204345703125, + "learning_rate": 1.4892945029392378e-07, + "loss": 0.0013, + "reward": 4.465624809265137, + "reward_std": 0.048403505235910416, + "rewards/accuracy_reward": 2.971874952316284, + "rewards/format_reward": 1.0, + "step": 372, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.96875, + "epoch": 0.6376068376068376, + "grad_norm": 0.7670469397108076, + "kl": 0.036468505859375, + "learning_rate": 1.4770296129830817e-07, + "loss": 0.0015, + "reward": 4.481249809265137, + "reward_std": 0.03017766959965229, + "rewards/accuracy_reward": 2.987499952316284, + "rewards/format_reward": 1.0, + "step": 373, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 371.875, + "epoch": 0.6393162393162393, + "grad_norm": 0.5412882759369413, + "kl": 0.028717041015625, + "learning_rate": 1.4647942249299704e-07, + "loss": 0.0011, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 374, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.09375, + "epoch": 0.6410256410256411, + "grad_norm": 0.7394286144083122, + "kl": 0.0311279296875, + "learning_rate": 1.4525886916417629e-07, + "loss": 0.0012, + "reward": 4.356249809265137, + "reward_std": 0.03750002384185791, + "rewards/accuracy_reward": 2.856250047683716, + "rewards/format_reward": 1.0, + "step": 375, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.65625, + "epoch": 0.6427350427350428, + "grad_norm": 0.8768404382198639, + "kl": 0.033660888671875, + "learning_rate": 1.4404133651193212e-07, + "loss": 0.0013, + "reward": 4.469791412353516, + "reward_std": 0.04280628263950348, + "rewards/accuracy_reward": 2.976041793823242, + "rewards/format_reward": 1.0, + "step": 376, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.25, + "epoch": 0.6444444444444445, + "grad_norm": 0.7423460450093495, + "kl": 0.031463623046875, + "learning_rate": 1.428268596492364e-07, + "loss": 0.0013, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 377, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.96875, + "epoch": 0.6461538461538462, + "grad_norm": 0.6563307928419717, + "kl": 0.028289794921875, + "learning_rate": 1.4161547360093362e-07, + "loss": 0.0011, + "reward": 4.478125095367432, + "reward_std": 0.043749988079071045, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 378, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.6875, + "epoch": 0.6478632478632479, + "grad_norm": 0.7929408963957806, + "kl": 0.0272674560546875, + "learning_rate": 1.404072133027306e-07, + "loss": 0.0011, + "reward": 4.478125095367432, + "reward_std": 0.043749988079071045, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 379, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.71875, + "epoch": 0.6495726495726496, + "grad_norm": 0.7163262109272821, + "kl": 0.033660888671875, + "learning_rate": 1.392021136001897e-07, + "loss": 0.0013, + "reward": 4.491666793823242, + "reward_std": 0.016666710376739502, + "rewards/accuracy_reward": 2.991666793823242, + "rewards/format_reward": 1.0, + "step": 380, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.84375, + "epoch": 0.6512820512820513, + "grad_norm": 0.7711914301051815, + "kl": 0.033843994140625, + "learning_rate": 1.3800020924772292e-07, + "loss": 0.0014, + "reward": 4.478125095367432, + "reward_std": 0.030829090625047684, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 381, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.0625, + "epoch": 0.652991452991453, + "grad_norm": 0.7855800057850417, + "kl": 0.03350830078125, + "learning_rate": 1.3680153490759073e-07, + "loss": 0.0013, + "reward": 4.478125095367432, + "reward_std": 0.043749988079071045, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 382, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.4375, + "epoch": 0.6547008547008547, + "grad_norm": 0.791258474349177, + "kl": 0.034576416015625, + "learning_rate": 1.3560612514890117e-07, + "loss": 0.0014, + "reward": 4.483333110809326, + "reward_std": 0.02375653013586998, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 383, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.59375, + "epoch": 0.6564102564102564, + "grad_norm": 0.549626026484795, + "kl": 0.03564453125, + "learning_rate": 1.3441401444661416e-07, + "loss": 0.0014, + "reward": 4.428124904632568, + "reward_std": 0.14375001192092896, + "rewards/accuracy_reward": 2.9281249046325684, + "rewards/format_reward": 1.0, + "step": 384, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.28125, + "epoch": 0.6581196581196581, + "grad_norm": 0.9408993865860127, + "kl": 0.0305633544921875, + "learning_rate": 1.3322523718054612e-07, + "loss": 0.0012, + "reward": 4.482291221618652, + "reward_std": 0.03541666269302368, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 385, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.75, + "epoch": 0.6598290598290598, + "grad_norm": 0.9335523737841157, + "kl": 0.032958984375, + "learning_rate": 1.320398276343795e-07, + "loss": 0.0013, + "reward": 4.462500095367432, + "reward_std": 0.07500000298023224, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 386, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.5, + "epoch": 0.6615384615384615, + "grad_norm": 0.49404137427048744, + "kl": 0.0316162109375, + "learning_rate": 1.30857819994673e-07, + "loss": 0.0013, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 387, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.5, + "epoch": 0.6632478632478632, + "grad_norm": 0.09086186442266743, + "kl": 0.0306854248046875, + "learning_rate": 1.2967924834987686e-07, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 388, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.21875, + "epoch": 0.6649572649572649, + "grad_norm": 0.9109818811980164, + "kl": 0.03759765625, + "learning_rate": 1.2850414668934847e-07, + "loss": 0.0015, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 389, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.59375, + "epoch": 0.6666666666666666, + "grad_norm": 0.7334463778326866, + "kl": 0.03118896484375, + "learning_rate": 1.2733254890237334e-07, + "loss": 0.0012, + "reward": 4.478124618530273, + "reward_std": 0.030829111114144325, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 390, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.65625, + "epoch": 0.6683760683760683, + "grad_norm": 0.7840492710187797, + "kl": 0.0286407470703125, + "learning_rate": 1.2616448877718673e-07, + "loss": 0.0011, + "reward": 4.487500190734863, + "reward_std": 0.025000017136335373, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 391, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.75, + "epoch": 0.67008547008547, + "grad_norm": 0.9495331007616914, + "kl": 0.0318603515625, + "learning_rate": 1.2500000000000005e-07, + "loss": 0.0013, + "reward": 4.441666603088379, + "reward_std": 0.10923752188682556, + "rewards/accuracy_reward": 2.941666603088379, + "rewards/format_reward": 1.0, + "step": 392, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.625, + "epoch": 0.6717948717948717, + "grad_norm": 0.8886420165233839, + "kl": 0.031982421875, + "learning_rate": 1.238391161540287e-07, + "loss": 0.0013, + "reward": 4.346875190734863, + "reward_std": 0.04866214841604233, + "rewards/accuracy_reward": 2.846874952316284, + "rewards/format_reward": 1.0, + "step": 393, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.03125, + "epoch": 0.6735042735042736, + "grad_norm": 0.49575490041548353, + "kl": 0.03472900390625, + "learning_rate": 1.2268187071852417e-07, + "loss": 0.0014, + "reward": 4.496874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 394, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.75, + "epoch": 0.6752136752136753, + "grad_norm": 0.5842336321782885, + "kl": 0.031982421875, + "learning_rate": 1.2152829706780784e-07, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 395, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.59375, + "epoch": 0.676923076923077, + "grad_norm": 0.8273180840153671, + "kl": 0.032012939453125, + "learning_rate": 1.203784284703091e-07, + "loss": 0.0013, + "reward": 4.469791889190674, + "reward_std": 0.06041668727993965, + "rewards/accuracy_reward": 2.976041793823242, + "rewards/format_reward": 1.0, + "step": 396, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.65625, + "epoch": 0.6786324786324787, + "grad_norm": 0.09560569515302797, + "kl": 0.033966064453125, + "learning_rate": 1.1923229808760561e-07, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 397, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.59375, + "epoch": 0.6803418803418804, + "grad_norm": 1.1765257678004042, + "kl": 0.034149169921875, + "learning_rate": 1.1808993897346678e-07, + "loss": 0.0014, + "reward": 4.453125, + "reward_std": 0.07613962143659592, + "rewards/accuracy_reward": 2.9593749046325684, + "rewards/format_reward": 1.0, + "step": 398, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.53125, + "epoch": 0.6820512820512821, + "grad_norm": 0.5454270246051581, + "kl": 0.03558349609375, + "learning_rate": 1.16951384072901e-07, + "loss": 0.0014, + "reward": 4.493750095367432, + "reward_std": 0.0072169057093560696, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 399, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.28125, + "epoch": 0.6837606837606838, + "grad_norm": 1.093814427642604, + "kl": 0.035003662109375, + "learning_rate": 1.1581666622120492e-07, + "loss": 0.0014, + "reward": 4.474999904632568, + "reward_std": 0.05000007152557373, + "rewards/accuracy_reward": 2.9750001430511475, + "rewards/format_reward": 1.0, + "step": 400, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.21875, + "epoch": 0.6854700854700855, + "grad_norm": 0.7507411819451001, + "kl": 0.032745361328125, + "learning_rate": 1.1468581814301717e-07, + "loss": 0.0013, + "reward": 4.464583396911621, + "reward_std": 0.07083334028720856, + "rewards/accuracy_reward": 2.964583396911621, + "rewards/format_reward": 1.0, + "step": 401, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.5, + "epoch": 0.6871794871794872, + "grad_norm": 0.7522102069840583, + "kl": 0.032196044921875, + "learning_rate": 1.1355887245137383e-07, + "loss": 0.0013, + "reward": 4.472916603088379, + "reward_std": 0.05416666343808174, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 402, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.53125, + "epoch": 0.6888888888888889, + "grad_norm": 0.7302779915955037, + "kl": 0.033203125, + "learning_rate": 1.1243586164676871e-07, + "loss": 0.0013, + "reward": 4.477083206176758, + "reward_std": 0.03407880663871765, + "rewards/accuracy_reward": 2.977083206176758, + "rewards/format_reward": 1.0, + "step": 403, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.875, + "epoch": 0.6905982905982906, + "grad_norm": 1.1002198477594594, + "kl": 0.032806396484375, + "learning_rate": 1.1131681811621527e-07, + "loss": 0.0013, + "reward": 4.471874713897705, + "reward_std": 0.03420006483793259, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 404, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.3125, + "epoch": 0.6923076923076923, + "grad_norm": 0.7461741999370974, + "kl": 0.0283203125, + "learning_rate": 1.1020177413231332e-07, + "loss": 0.0011, + "reward": 4.483333587646484, + "reward_std": 0.03333333879709244, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 405, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.0625, + "epoch": 0.694017094017094, + "grad_norm": 0.10058816234853812, + "kl": 0.032928466796875, + "learning_rate": 1.0909076185231761e-07, + "loss": 0.0013, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 406, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.4375, + "epoch": 0.6957264957264957, + "grad_norm": 0.4593278233452189, + "kl": 0.03375244140625, + "learning_rate": 1.0798381331721107e-07, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 407, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.90625, + "epoch": 0.6974358974358974, + "grad_norm": 0.9300004852827248, + "kl": 0.029571533203125, + "learning_rate": 1.0688096045078022e-07, + "loss": 0.0012, + "reward": 4.3458333015441895, + "reward_std": 0.0583333745598793, + "rewards/accuracy_reward": 2.8458333015441895, + "rewards/format_reward": 1.0, + "step": 408, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.21875, + "epoch": 0.6991452991452991, + "grad_norm": 0.9315789428494882, + "kl": 0.030853271484375, + "learning_rate": 1.0578223505869493e-07, + "loss": 0.0012, + "reward": 4.474999904632568, + "reward_std": 0.05000004917383194, + "rewards/accuracy_reward": 2.9749999046325684, + "rewards/format_reward": 1.0, + "step": 409, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.03125, + "epoch": 0.7008547008547008, + "grad_norm": 0.750030869242921, + "kl": 0.03131103515625, + "learning_rate": 1.0468766882759092e-07, + "loss": 0.0013, + "reward": 4.483333587646484, + "reward_std": 0.03333337977528572, + "rewards/accuracy_reward": 2.9833333492279053, + "rewards/format_reward": 1.0, + "step": 410, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 401.375, + "epoch": 0.7025641025641025, + "grad_norm": 0.9122749493870264, + "kl": 0.033782958984375, + "learning_rate": 1.0359729332415582e-07, + "loss": 0.0014, + "reward": 4.4239583015441895, + "reward_std": 0.1520833820104599, + "rewards/accuracy_reward": 2.9239583015441895, + "rewards/format_reward": 1.0, + "step": 411, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.8125, + "epoch": 0.7042735042735043, + "grad_norm": 0.08071277485359081, + "kl": 0.031341552734375, + "learning_rate": 1.0251113999421935e-07, + "loss": 0.0013, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 412, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.71875, + "epoch": 0.705982905982906, + "grad_norm": 0.6055413986472074, + "kl": 0.03631591796875, + "learning_rate": 1.0142924016184568e-07, + "loss": 0.0015, + "reward": 4.469791412353516, + "reward_std": 0.026434533298015594, + "rewards/accuracy_reward": 2.9697916507720947, + "rewards/format_reward": 1.0, + "step": 413, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.40625, + "epoch": 0.7076923076923077, + "grad_norm": 0.8151039715648754, + "kl": 0.033416748046875, + "learning_rate": 1.0035162502843073e-07, + "loss": 0.0013, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 414, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 406.40625, + "epoch": 0.7094017094017094, + "grad_norm": 0.7739588314582155, + "kl": 0.032196044921875, + "learning_rate": 9.927832567180192e-08, + "loss": 0.0013, + "reward": 4.471875190734863, + "reward_std": 0.05624999850988388, + "rewards/accuracy_reward": 2.971874952316284, + "rewards/format_reward": 1.0, + "step": 415, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.03125, + "epoch": 0.7111111111111111, + "grad_norm": 1.1556123751806124, + "kl": 0.0338134765625, + "learning_rate": 9.82093730453222e-08, + "loss": 0.0014, + "reward": 4.472916603088379, + "reward_std": 0.05416671186685562, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 416, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.75, + "epoch": 0.7128205128205128, + "grad_norm": 0.9375743065286576, + "kl": 0.0308837890625, + "learning_rate": 9.714479797699692e-08, + "loss": 0.0012, + "reward": 4.471875190734863, + "reward_std": 0.045683782547712326, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 417, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.71875, + "epoch": 0.7145299145299145, + "grad_norm": 0.5137791697239468, + "kl": 0.031402587890625, + "learning_rate": 9.608463116858542e-08, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012500028125941753, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 418, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.21875, + "epoch": 0.7162393162393162, + "grad_norm": 0.9293546876839371, + "kl": 0.033905029296875, + "learning_rate": 9.50289031947149e-08, + "loss": 0.0014, + "reward": 4.465624809265137, + "reward_std": 0.0565461590886116, + "rewards/accuracy_reward": 2.9656248092651367, + "rewards/format_reward": 1.0, + "step": 419, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.34375, + "epoch": 0.717948717948718, + "grad_norm": 0.5212991604802728, + "kl": 0.035247802734375, + "learning_rate": 9.397764450199936e-08, + "loss": 0.0014, + "reward": 4.490624904632568, + "reward_std": 0.018750011920928955, + "rewards/accuracy_reward": 2.9906249046325684, + "rewards/format_reward": 1.0, + "step": 420, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.59375, + "epoch": 0.7196581196581197, + "grad_norm": 0.5197704693590894, + "kl": 0.033599853515625, + "learning_rate": 9.293088540816079e-08, + "loss": 0.0013, + "reward": 4.3645830154418945, + "reward_std": 0.012028153985738754, + "rewards/accuracy_reward": 2.8645832538604736, + "rewards/format_reward": 1.0, + "step": 421, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 403.71875, + "epoch": 0.7213675213675214, + "grad_norm": 0.7154972281077171, + "kl": 0.0341796875, + "learning_rate": 9.18886561011557e-08, + "loss": 0.0014, + "reward": 4.4864583015441895, + "reward_std": 0.027083376422524452, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 422, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.03125, + "epoch": 0.7230769230769231, + "grad_norm": 0.7894497576328201, + "kl": 0.034088134765625, + "learning_rate": 9.085098663830365e-08, + "loss": 0.0014, + "reward": 4.479166507720947, + "reward_std": 0.041666705161333084, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 423, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.28125, + "epoch": 0.7247863247863248, + "grad_norm": 0.790896159678788, + "kl": 0.03564453125, + "learning_rate": 8.981790694542087e-08, + "loss": 0.0014, + "reward": 4.478125095367432, + "reward_std": 0.04375002905726433, + "rewards/accuracy_reward": 2.9781250953674316, + "rewards/format_reward": 1.0, + "step": 424, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 399.6875, + "epoch": 0.7264957264957265, + "grad_norm": 0.50330884911545, + "kl": 0.033538818359375, + "learning_rate": 8.87894468159574e-08, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 425, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.15625, + "epoch": 0.7282051282051282, + "grad_norm": 0.8186288594902728, + "kl": 0.034576416015625, + "learning_rate": 8.776563591013728e-08, + "loss": 0.0014, + "reward": 4.340624809265137, + "reward_std": 0.06875002384185791, + "rewards/accuracy_reward": 2.840625047683716, + "rewards/format_reward": 1.0, + "step": 426, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.1875, + "epoch": 0.7299145299145299, + "grad_norm": 0.6945302885094401, + "kl": 0.0308685302734375, + "learning_rate": 8.674650375410378e-08, + "loss": 0.0012, + "reward": 4.479166507720947, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 427, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.6875, + "epoch": 0.7316239316239316, + "grad_norm": 0.44261007122030527, + "kl": 0.0377197265625, + "learning_rate": 8.573207973906735e-08, + "loss": 0.0015, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 428, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.6875, + "epoch": 0.7333333333333333, + "grad_norm": 0.9243327957772149, + "kl": 0.0330810546875, + "learning_rate": 8.47223931204585e-08, + "loss": 0.0013, + "reward": 4.46875, + "reward_std": 0.06250005960464478, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 429, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 393.875, + "epoch": 0.7350427350427351, + "grad_norm": 0.5341812028772001, + "kl": 0.03497314453125, + "learning_rate": 8.371747301708357e-08, + "loss": 0.0014, + "reward": 4.478125095367432, + "reward_std": 0.03590351715683937, + "rewards/accuracy_reward": 2.9781250953674316, + "rewards/format_reward": 1.0, + "step": 430, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 406.25, + "epoch": 0.7367521367521368, + "grad_norm": 0.7251723433495721, + "kl": 0.037750244140625, + "learning_rate": 8.271734841028552e-08, + "loss": 0.0015, + "reward": 4.306249618530273, + "reward_std": 0.19895486533641815, + "rewards/accuracy_reward": 2.84375, + "rewards/format_reward": 1.0, + "step": 431, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.875, + "epoch": 0.7384615384615385, + "grad_norm": 0.7511545208510357, + "kl": 0.03411865234375, + "learning_rate": 8.17220481431074e-08, + "loss": 0.0014, + "reward": 4.493750095367432, + "reward_std": 0.01250004768371582, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 432, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 405.34375, + "epoch": 0.7401709401709402, + "grad_norm": 0.6079327745864885, + "kl": 0.03179931640625, + "learning_rate": 8.073160091946155e-08, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012500028125941753, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 433, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 401.1875, + "epoch": 0.7418803418803419, + "grad_norm": 0.5700935104266779, + "kl": 0.0306243896484375, + "learning_rate": 7.974603530330067e-08, + "loss": 0.0012, + "reward": 4.487500190734863, + "reward_std": 0.014433743432164192, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 434, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.53125, + "epoch": 0.7435897435897436, + "grad_norm": 0.753083570670292, + "kl": 0.038360595703125, + "learning_rate": 7.876537971779493e-08, + "loss": 0.0015, + "reward": 4.431250095367432, + "reward_std": 0.13749998807907104, + "rewards/accuracy_reward": 2.9375, + "rewards/format_reward": 1.0, + "step": 435, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.53125, + "epoch": 0.7452991452991453, + "grad_norm": 0.8013844938082755, + "kl": 0.0286407470703125, + "learning_rate": 7.778966244451168e-08, + "loss": 0.0011, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895830154418945, + "rewards/format_reward": 1.0, + "step": 436, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.875, + "epoch": 0.747008547008547, + "grad_norm": 0.5320691070134214, + "kl": 0.03106689453125, + "learning_rate": 7.681891162260015e-08, + "loss": 0.0012, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 437, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.75, + "epoch": 0.7487179487179487, + "grad_norm": 0.7052346428951947, + "kl": 0.03802490234375, + "learning_rate": 7.585315524797998e-08, + "loss": 0.0015, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 438, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 406.59375, + "epoch": 0.7504273504273504, + "grad_norm": 0.8602799183038976, + "kl": 0.035736083984375, + "learning_rate": 7.489242117253341e-08, + "loss": 0.0014, + "reward": 4.4114580154418945, + "reward_std": 0.17708338797092438, + "rewards/accuracy_reward": 2.9114582538604736, + "rewards/format_reward": 1.0, + "step": 439, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 406.15625, + "epoch": 0.7521367521367521, + "grad_norm": 1.3190919507909775, + "kl": 0.0305633544921875, + "learning_rate": 7.39367371033027e-08, + "loss": 0.0012, + "reward": 4.4322919845581055, + "reward_std": 0.13541670143604279, + "rewards/accuracy_reward": 2.944791793823242, + "rewards/format_reward": 1.0, + "step": 440, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.78125, + "epoch": 0.7538461538461538, + "grad_norm": 0.47304749095078713, + "kl": 0.032012939453125, + "learning_rate": 7.298613060169034e-08, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 441, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.4375, + "epoch": 0.7555555555555555, + "grad_norm": 0.5246643007021763, + "kl": 0.034881591796875, + "learning_rate": 7.204062908266489e-08, + "loss": 0.0014, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 442, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.21875, + "epoch": 0.7572649572649572, + "grad_norm": 0.09360830328304175, + "kl": 0.032196044921875, + "learning_rate": 7.110025981396975e-08, + "loss": 0.0013, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 443, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.875, + "epoch": 0.7589743589743589, + "grad_norm": 0.9678920676984522, + "kl": 0.031280517578125, + "learning_rate": 7.016504991533726e-08, + "loss": 0.0013, + "reward": 4.476041793823242, + "reward_std": 0.0366131067276001, + "rewards/accuracy_reward": 2.976041316986084, + "rewards/format_reward": 1.0, + "step": 444, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 397.3125, + "epoch": 0.7606837606837606, + "grad_norm": 0.8190169960494074, + "kl": 0.036529541015625, + "learning_rate": 6.923502635770617e-08, + "loss": 0.0015, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 445, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.84375, + "epoch": 0.7623931623931623, + "grad_norm": 0.5344964134250557, + "kl": 0.030120849609375, + "learning_rate": 6.831021596244424e-08, + "loss": 0.0012, + "reward": 4.490624904632568, + "reward_std": 0.018750011920928955, + "rewards/accuracy_reward": 2.9906249046325684, + "rewards/format_reward": 1.0, + "step": 446, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.125, + "epoch": 0.764102564102564, + "grad_norm": 0.5104896778148305, + "kl": 0.0340576171875, + "learning_rate": 6.739064540057424e-08, + "loss": 0.0014, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 447, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 394.90625, + "epoch": 0.7658119658119659, + "grad_norm": 0.923990603996032, + "kl": 0.036163330078125, + "learning_rate": 6.64763411920053e-08, + "loss": 0.0014, + "reward": 4.346874713897705, + "reward_std": 0.23125003278255463, + "rewards/accuracy_reward": 2.890625, + "rewards/format_reward": 1.0, + "step": 448, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.28125, + "epoch": 0.7675213675213676, + "grad_norm": 0.7213594899516708, + "kl": 0.027252197265625, + "learning_rate": 6.556732970476747e-08, + "loss": 0.0011, + "reward": 4.478125095367432, + "reward_std": 0.043749988079071045, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 449, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.84375, + "epoch": 0.7692307692307693, + "grad_norm": 0.1570549185979674, + "kl": 0.0416259765625, + "learning_rate": 6.466363715425199e-08, + "loss": 0.0017, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 450, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.53125, + "epoch": 0.770940170940171, + "grad_norm": 0.076118115415132, + "kl": 0.0299072265625, + "learning_rate": 6.376528960245476e-08, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 451, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.375, + "epoch": 0.7726495726495727, + "grad_norm": 0.9166985950940988, + "kl": 0.03118896484375, + "learning_rate": 6.28723129572247e-08, + "loss": 0.0012, + "reward": 4.457291603088379, + "reward_std": 0.08541667461395264, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 0.96875, + "step": 452, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.1875, + "epoch": 0.7743589743589744, + "grad_norm": 0.7765296821892308, + "kl": 0.03106689453125, + "learning_rate": 6.198473297151704e-08, + "loss": 0.0012, + "reward": 4.463541507720947, + "reward_std": 0.07291668653488159, + "rewards/accuracy_reward": 2.9635415077209473, + "rewards/format_reward": 1.0, + "step": 453, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 390.3125, + "epoch": 0.7760683760683761, + "grad_norm": 0.08247118622297261, + "kl": 0.035369873046875, + "learning_rate": 6.110257524264997e-08, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 454, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.1875, + "epoch": 0.7777777777777778, + "grad_norm": 0.1026029910501858, + "kl": 0.0347900390625, + "learning_rate": 6.022586521156714e-08, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 455, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.21875, + "epoch": 0.7794871794871795, + "grad_norm": 0.11223910005504889, + "kl": 0.0296478271484375, + "learning_rate": 5.935462816210324e-08, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 456, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 396.46875, + "epoch": 0.7811965811965812, + "grad_norm": 0.09582940298335704, + "kl": 0.03265380859375, + "learning_rate": 5.848888922025552e-08, + "loss": 0.0013, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 457, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.46875, + "epoch": 0.7829059829059829, + "grad_norm": 1.0500760370031257, + "kl": 0.035552978515625, + "learning_rate": 5.76286733534585e-08, + "loss": 0.0014, + "reward": 4.467708587646484, + "reward_std": 0.05401712283492088, + "rewards/accuracy_reward": 2.980208396911621, + "rewards/format_reward": 1.0, + "step": 458, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.4375, + "epoch": 0.7846153846153846, + "grad_norm": 0.7419885297063772, + "kl": 0.02899169921875, + "learning_rate": 5.67740053698646e-08, + "loss": 0.0012, + "reward": 4.452083587646484, + "reward_std": 0.05791240185499191, + "rewards/accuracy_reward": 2.9583332538604736, + "rewards/format_reward": 1.0, + "step": 459, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 394.5, + "epoch": 0.7863247863247863, + "grad_norm": 0.6994476739758024, + "kl": 0.0301666259765625, + "learning_rate": 5.5924909917627995e-08, + "loss": 0.0012, + "reward": 4.453125, + "reward_std": 0.09375, + "rewards/accuracy_reward": 2.953125, + "rewards/format_reward": 1.0, + "step": 460, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 381.625, + "epoch": 0.788034188034188, + "grad_norm": 1.1649572465381595, + "kl": 0.034759521484375, + "learning_rate": 5.508141148419443e-08, + "loss": 0.0014, + "reward": 4.337499618530273, + "reward_std": 0.24119488894939423, + "rewards/accuracy_reward": 2.8812499046325684, + "rewards/format_reward": 1.0, + "step": 461, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.5, + "epoch": 0.7897435897435897, + "grad_norm": 0.6974141792088103, + "kl": 0.028564453125, + "learning_rate": 5.424353439559445e-08, + "loss": 0.0011, + "reward": 4.487500190734863, + "reward_std": 0.019716894254088402, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 462, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 375.03125, + "epoch": 0.7914529914529914, + "grad_norm": 0.8862170174981424, + "kl": 0.031463623046875, + "learning_rate": 5.3411302815742324e-08, + "loss": 0.0013, + "reward": 4.3729166984558105, + "reward_std": 0.17916667461395264, + "rewards/accuracy_reward": 2.9166665077209473, + "rewards/format_reward": 1.0, + "step": 463, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.46875, + "epoch": 0.7931623931623931, + "grad_norm": 0.7808535507743032, + "kl": 0.033905029296875, + "learning_rate": 5.2584740745738766e-08, + "loss": 0.0014, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 464, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.625, + "epoch": 0.7948717948717948, + "grad_norm": 0.9890091754072906, + "kl": 0.0303192138671875, + "learning_rate": 5.176387202317914e-08, + "loss": 0.0012, + "reward": 4.483333110809326, + "reward_std": 0.03333336114883423, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 465, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.90625, + "epoch": 0.7965811965811965, + "grad_norm": 0.8810236829977253, + "kl": 0.03179931640625, + "learning_rate": 5.0948720321465605e-08, + "loss": 0.0013, + "reward": 4.484375, + "reward_std": 0.031250059604644775, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 466, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 375.875, + "epoch": 0.7982905982905983, + "grad_norm": 0.7607492562548063, + "kl": 0.031219482421875, + "learning_rate": 5.013930914912476e-08, + "loss": 0.0012, + "reward": 4.368750095367432, + "reward_std": 0.1875, + "rewards/accuracy_reward": 2.90625, + "rewards/format_reward": 1.0, + "step": 467, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.5625, + "epoch": 0.8, + "grad_norm": 0.6990381492224721, + "kl": 0.0308837890625, + "learning_rate": 4.9335661849129295e-08, + "loss": 0.0012, + "reward": 4.429166793823242, + "reward_std": 0.1416666954755783, + "rewards/accuracy_reward": 2.929166793823242, + "rewards/format_reward": 1.0, + "step": 468, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.59375, + "epoch": 0.8017094017094017, + "grad_norm": 0.7391548171546465, + "kl": 0.0304718017578125, + "learning_rate": 4.853780159822521e-08, + "loss": 0.0012, + "reward": 4.485416412353516, + "reward_std": 0.029166698455810547, + "rewards/accuracy_reward": 2.991666555404663, + "rewards/format_reward": 1.0, + "step": 469, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.25, + "epoch": 0.8034188034188035, + "grad_norm": 1.0602483865953845, + "kl": 0.0290985107421875, + "learning_rate": 4.774575140626316e-08, + "loss": 0.0012, + "reward": 4.453125, + "reward_std": 0.06269193440675735, + "rewards/accuracy_reward": 2.9593753814697266, + "rewards/format_reward": 1.0, + "step": 470, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.1875, + "epoch": 0.8051282051282052, + "grad_norm": 0.7802955828215046, + "kl": 0.028533935546875, + "learning_rate": 4.695953411553466e-08, + "loss": 0.0011, + "reward": 4.484375, + "reward_std": 0.020683767274022102, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 471, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.71875, + "epoch": 0.8068376068376069, + "grad_norm": 0.7183676034820473, + "kl": 0.033111572265625, + "learning_rate": 4.617917240011393e-08, + "loss": 0.0013, + "reward": 4.474999904632568, + "reward_std": 0.032993994653224945, + "rewards/accuracy_reward": 2.9812498092651367, + "rewards/format_reward": 1.0, + "step": 472, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.375, + "epoch": 0.8085470085470086, + "grad_norm": 0.8113707858305483, + "kl": 0.0306854248046875, + "learning_rate": 4.5404688765203233e-08, + "loss": 0.0012, + "reward": 4.487500190734863, + "reward_std": 0.025000035762786865, + "rewards/accuracy_reward": 2.9875001907348633, + "rewards/format_reward": 1.0, + "step": 473, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 366.96875, + "epoch": 0.8102564102564103, + "grad_norm": 1.5001336107988816, + "kl": 0.0313720703125, + "learning_rate": 4.463610554648459e-08, + "loss": 0.0013, + "reward": 4.327083587646484, + "reward_std": 0.25816476345062256, + "rewards/accuracy_reward": 2.8489584922790527, + "rewards/format_reward": 1.0, + "step": 474, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 375.75, + "epoch": 0.811965811965812, + "grad_norm": 0.9830876497961911, + "kl": 0.0305328369140625, + "learning_rate": 4.387344490947498e-08, + "loss": 0.0012, + "reward": 4.349999904632568, + "reward_std": 0.22499996423721313, + "rewards/accuracy_reward": 2.90625, + "rewards/format_reward": 1.0, + "step": 475, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.34375, + "epoch": 0.8136752136752137, + "grad_norm": 0.8463083045702553, + "kl": 0.034942626953125, + "learning_rate": 4.311672884888756e-08, + "loss": 0.0014, + "reward": 4.462499618530273, + "reward_std": 0.03685124218463898, + "rewards/accuracy_reward": 2.968749761581421, + "rewards/format_reward": 1.0, + "step": 476, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.96875, + "epoch": 0.8153846153846154, + "grad_norm": 0.8916852982526418, + "kl": 0.03509521484375, + "learning_rate": 4.2365979187997084e-08, + "loss": 0.0014, + "reward": 4.457291603088379, + "reward_std": 0.08541667461395264, + "rewards/accuracy_reward": 2.9635415077209473, + "rewards/format_reward": 1.0, + "step": 477, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.15625, + "epoch": 0.8170940170940171, + "grad_norm": 2.5252893617820487, + "kl": 0.036285400390625, + "learning_rate": 4.162121757801068e-08, + "loss": 0.0015, + "reward": 4.465624809265137, + "reward_std": 0.06132083013653755, + "rewards/accuracy_reward": 2.9656248092651367, + "rewards/format_reward": 1.0, + "step": 478, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.59375, + "epoch": 0.8188034188034188, + "grad_norm": 0.10243465928061427, + "kl": 0.038970947265625, + "learning_rate": 4.0882465497443313e-08, + "loss": 0.0016, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 479, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.5625, + "epoch": 0.8205128205128205, + "grad_norm": 0.455371320569445, + "kl": 0.0281524658203125, + "learning_rate": 4.014974425149853e-08, + "loss": 0.0011, + "reward": 4.487500190734863, + "reward_std": 0.014433743432164192, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 480, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.65625, + "epoch": 0.8222222222222222, + "grad_norm": 0.8196123722085543, + "kl": 0.03131103515625, + "learning_rate": 3.942307497145378e-08, + "loss": 0.0013, + "reward": 4.3302083015441895, + "reward_std": 0.07463379204273224, + "rewards/accuracy_reward": 2.8489582538604736, + "rewards/format_reward": 1.0, + "step": 481, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 367.78125, + "epoch": 0.8239316239316239, + "grad_norm": 0.7022762843056454, + "kl": 0.032073974609375, + "learning_rate": 3.8702478614051345e-08, + "loss": 0.0013, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 482, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.125, + "epoch": 0.8256410256410256, + "grad_norm": 0.597726990167221, + "kl": 0.02642822265625, + "learning_rate": 3.798797596089351e-08, + "loss": 0.0011, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 483, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.40625, + "epoch": 0.8273504273504273, + "grad_norm": 0.9212880375324389, + "kl": 0.0294189453125, + "learning_rate": 3.727958761784375e-08, + "loss": 0.0012, + "reward": 4.481249809265137, + "reward_std": 0.03750000521540642, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 484, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.25, + "epoch": 0.8290598290598291, + "grad_norm": 0.7980192569673967, + "kl": 0.0251007080078125, + "learning_rate": 3.6577334014431997e-08, + "loss": 0.001, + "reward": 4.460416793823242, + "reward_std": 0.07112002372741699, + "rewards/accuracy_reward": 2.960416555404663, + "rewards/format_reward": 1.0, + "step": 485, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.21875, + "epoch": 0.8307692307692308, + "grad_norm": 0.08452297569542742, + "kl": 0.031005859375, + "learning_rate": 3.588123540326571e-08, + "loss": 0.0012, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 486, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.4375, + "epoch": 0.8324786324786325, + "grad_norm": 0.9056999881313452, + "kl": 0.027587890625, + "learning_rate": 3.5191311859445795e-08, + "loss": 0.0011, + "reward": 4.459374904632568, + "reward_std": 0.08125001192092896, + "rewards/accuracy_reward": 2.965625047683716, + "rewards/format_reward": 1.0, + "step": 487, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.25, + "epoch": 0.8341880341880342, + "grad_norm": 0.4452409717412741, + "kl": 0.03271484375, + "learning_rate": 3.450758327998768e-08, + "loss": 0.0013, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 488, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.46875, + "epoch": 0.8358974358974359, + "grad_norm": 0.06945328996157386, + "kl": 0.032135009765625, + "learning_rate": 3.383006938324734e-08, + "loss": 0.0013, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 489, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.875, + "epoch": 0.8376068376068376, + "grad_norm": 1.0104190261642534, + "kl": 0.033935546875, + "learning_rate": 3.315878970835267e-08, + "loss": 0.0014, + "reward": 4.445833206176758, + "reward_std": 0.09952813386917114, + "rewards/accuracy_reward": 2.9583330154418945, + "rewards/format_reward": 1.0, + "step": 490, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.84375, + "epoch": 0.8393162393162393, + "grad_norm": 0.5840091013811081, + "kl": 0.03118896484375, + "learning_rate": 3.249376361464021e-08, + "loss": 0.0012, + "reward": 4.371874809265137, + "reward_std": 0.00625002384185791, + "rewards/accuracy_reward": 2.871875047683716, + "rewards/format_reward": 1.0, + "step": 491, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 367.53125, + "epoch": 0.841025641025641, + "grad_norm": 0.9683824952769233, + "kl": 0.029205322265625, + "learning_rate": 3.183501028109642e-08, + "loss": 0.0012, + "reward": 4.474999904632568, + "reward_std": 0.03943371772766113, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 492, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.28125, + "epoch": 0.8427350427350427, + "grad_norm": 0.9037075245637992, + "kl": 0.028076171875, + "learning_rate": 3.1182548705805056e-08, + "loss": 0.0011, + "reward": 4.471875190734863, + "reward_std": 0.05624999478459358, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 493, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.6875, + "epoch": 0.8444444444444444, + "grad_norm": 0.7836588351126467, + "kl": 0.026031494140625, + "learning_rate": 3.053639770539884e-08, + "loss": 0.001, + "reward": 4.487500190734863, + "reward_std": 0.025000017136335373, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 494, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.59375, + "epoch": 0.8461538461538461, + "grad_norm": 0.8963328689269091, + "kl": 0.0316162109375, + "learning_rate": 2.989657591451716e-08, + "loss": 0.0013, + "reward": 4.456249713897705, + "reward_std": 0.07264164090156555, + "rewards/accuracy_reward": 2.9625000953674316, + "rewards/format_reward": 1.0, + "step": 495, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.96875, + "epoch": 0.8478632478632478, + "grad_norm": 0.4743907243744337, + "kl": 0.0283660888671875, + "learning_rate": 2.9263101785268252e-08, + "loss": 0.0011, + "reward": 4.359375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.859375, + "rewards/format_reward": 1.0, + "step": 496, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.5, + "epoch": 0.8495726495726496, + "grad_norm": 0.49878949373559706, + "kl": 0.028564453125, + "learning_rate": 2.863599358669755e-08, + "loss": 0.0011, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 497, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.125, + "epoch": 0.8512820512820513, + "grad_norm": 0.9923991962632441, + "kl": 0.0313720703125, + "learning_rate": 2.8015269404260327e-08, + "loss": 0.0013, + "reward": 4.391666412353516, + "reward_std": 0.20358917117118835, + "rewards/accuracy_reward": 2.897916793823242, + "rewards/format_reward": 1.0, + "step": 498, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.40625, + "epoch": 0.852991452991453, + "grad_norm": 0.760643580547958, + "kl": 0.03564453125, + "learning_rate": 2.740094713930044e-08, + "loss": 0.0014, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 499, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 367.4375, + "epoch": 0.8547008547008547, + "grad_norm": 0.7783159068664107, + "kl": 0.0299224853515625, + "learning_rate": 2.679304450853401e-08, + "loss": 0.0012, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 500, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.03125, + "epoch": 0.8564102564102564, + "grad_norm": 0.8319195898451789, + "kl": 0.0276336669921875, + "learning_rate": 2.6191579043538333e-08, + "loss": 0.0011, + "reward": 4.4270830154418945, + "reward_std": 0.14583337306976318, + "rewards/accuracy_reward": 2.9270832538604736, + "rewards/format_reward": 1.0, + "step": 501, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.90625, + "epoch": 0.8581196581196581, + "grad_norm": 0.5761278935527648, + "kl": 0.03009033203125, + "learning_rate": 2.5596568090246545e-08, + "loss": 0.0012, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 502, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.34375, + "epoch": 0.8598290598290599, + "grad_norm": 0.5211704779844807, + "kl": 0.0301666259765625, + "learning_rate": 2.500802880844699e-08, + "loss": 0.0012, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 503, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.09375, + "epoch": 0.8615384615384616, + "grad_norm": 1.2959231439342969, + "kl": 0.029937744140625, + "learning_rate": 2.4425978171288802e-08, + "loss": 0.0012, + "reward": 4.444791793823242, + "reward_std": 0.07988535612821579, + "rewards/accuracy_reward": 2.9635417461395264, + "rewards/format_reward": 1.0, + "step": 504, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.59375, + "epoch": 0.8632478632478633, + "grad_norm": 0.4892530739596482, + "kl": 0.03399658203125, + "learning_rate": 2.3850432964791945e-08, + "loss": 0.0014, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 505, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.78125, + "epoch": 0.864957264957265, + "grad_norm": 0.770760898444276, + "kl": 0.03436279296875, + "learning_rate": 2.3281409787363648e-08, + "loss": 0.0014, + "reward": 4.481249809265137, + "reward_std": 0.03750002384185791, + "rewards/accuracy_reward": 2.981250047683716, + "rewards/format_reward": 1.0, + "step": 506, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 364.90625, + "epoch": 0.8666666666666667, + "grad_norm": 1.184501188781967, + "kl": 0.028167724609375, + "learning_rate": 2.2718925049319048e-08, + "loss": 0.0011, + "reward": 4.462500095367432, + "reward_std": 0.062079109251499176, + "rewards/accuracy_reward": 2.9749999046325684, + "rewards/format_reward": 1.0, + "step": 507, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 398.5625, + "epoch": 0.8683760683760684, + "grad_norm": 0.5037798400384479, + "kl": 0.0311126708984375, + "learning_rate": 2.2162994972408643e-08, + "loss": 0.0012, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 508, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.15625, + "epoch": 0.8700854700854701, + "grad_norm": 0.7114656942445862, + "kl": 0.029266357421875, + "learning_rate": 2.1613635589349756e-08, + "loss": 0.0012, + "reward": 4.403124809265137, + "reward_std": 0.18555781245231628, + "rewards/accuracy_reward": 2.903125047683716, + "rewards/format_reward": 1.0, + "step": 509, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.53125, + "epoch": 0.8717948717948718, + "grad_norm": 0.9284269082444429, + "kl": 0.0289459228515625, + "learning_rate": 2.1070862743364836e-08, + "loss": 0.0012, + "reward": 4.4583330154418945, + "reward_std": 0.08333337306976318, + "rewards/accuracy_reward": 2.9583332538604736, + "rewards/format_reward": 1.0, + "step": 510, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.8125, + "epoch": 0.8735042735042735, + "grad_norm": 1.0818595327849414, + "kl": 0.02764892578125, + "learning_rate": 2.0534692087724015e-08, + "loss": 0.0011, + "reward": 4.462499618530273, + "reward_std": 0.07500006258487701, + "rewards/accuracy_reward": 2.9625000953674316, + "rewards/format_reward": 1.0, + "step": 511, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 371.59375, + "epoch": 0.8752136752136752, + "grad_norm": 0.8015844416518322, + "kl": 0.032684326171875, + "learning_rate": 2.0005139085293942e-08, + "loss": 0.0013, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 512, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.4375, + "epoch": 0.8769230769230769, + "grad_norm": 0.5608605027967921, + "kl": 0.0270233154296875, + "learning_rate": 1.9482219008091883e-08, + "loss": 0.0011, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 513, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.4375, + "epoch": 0.8786324786324786, + "grad_norm": 1.0197644879391652, + "kl": 0.02972412109375, + "learning_rate": 1.8965946936845027e-08, + "loss": 0.0012, + "reward": 4.482291221618652, + "reward_std": 0.035416703671216965, + "rewards/accuracy_reward": 2.988541603088379, + "rewards/format_reward": 1.0, + "step": 514, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 489.0625, + "epoch": 0.8803418803418803, + "grad_norm": 0.29736027233547324, + "kl": 0.0288543701171875, + "learning_rate": 1.845633776055591e-08, + "loss": 0.0012, + "reward": 4.331250190734863, + "reward_std": 0.26249998807907104, + "rewards/accuracy_reward": 2.90625, + "rewards/format_reward": 0.96875, + "step": 515, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.84375, + "epoch": 0.882051282051282, + "grad_norm": 1.2676126849679452, + "kl": 0.02862548828125, + "learning_rate": 1.7953406176072632e-08, + "loss": 0.0011, + "reward": 4.431250095367432, + "reward_std": 0.1286948174238205, + "rewards/accuracy_reward": 2.9375, + "rewards/format_reward": 1.0, + "step": 516, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.71875, + "epoch": 0.8837606837606837, + "grad_norm": 0.7440933489373224, + "kl": 0.031646728515625, + "learning_rate": 1.7457166687665447e-08, + "loss": 0.0013, + "reward": 4.353125095367432, + "reward_std": 0.043749988079071045, + "rewards/accuracy_reward": 2.859375, + "rewards/format_reward": 1.0, + "step": 517, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.3125, + "epoch": 0.8854700854700854, + "grad_norm": 0.7393577005284849, + "kl": 0.027435302734375, + "learning_rate": 1.6967633606608077e-08, + "loss": 0.0011, + "reward": 4.484375, + "reward_std": 0.031250037252902985, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 518, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.375, + "epoch": 0.8871794871794871, + "grad_norm": 0.9007448687343432, + "kl": 0.033538818359375, + "learning_rate": 1.6484821050765207e-08, + "loss": 0.0013, + "reward": 4.457291603088379, + "reward_std": 0.08541667461395264, + "rewards/accuracy_reward": 2.9635415077209473, + "rewards/format_reward": 1.0, + "step": 519, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.4375, + "epoch": 0.8888888888888888, + "grad_norm": 0.8896325906031658, + "kl": 0.027740478515625, + "learning_rate": 1.600874294418528e-08, + "loss": 0.0011, + "reward": 4.481249809265137, + "reward_std": 0.02693377062678337, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 520, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 395.34375, + "epoch": 0.8905982905982905, + "grad_norm": 0.4679849808583376, + "kl": 0.029876708984375, + "learning_rate": 1.553941301669892e-08, + "loss": 0.0012, + "reward": 4.4864583015441895, + "reward_std": 0.01965414360165596, + "rewards/accuracy_reward": 2.9864583015441895, + "rewards/format_reward": 1.0, + "step": 521, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.25, + "epoch": 0.8923076923076924, + "grad_norm": 0.5524700061127931, + "kl": 0.0288848876953125, + "learning_rate": 1.507684480352292e-08, + "loss": 0.0012, + "reward": 4.46875, + "reward_std": 0.0625, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 522, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 367.5, + "epoch": 0.8940170940170941, + "grad_norm": 0.7202553581598627, + "kl": 0.027191162109375, + "learning_rate": 1.4621051644870097e-08, + "loss": 0.0011, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 523, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.125, + "epoch": 0.8957264957264958, + "grad_norm": 0.7887618526962882, + "kl": 0.0266876220703125, + "learning_rate": 1.4172046685564209e-08, + "loss": 0.0011, + "reward": 4.481249809265137, + "reward_std": 0.02693377062678337, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 524, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.6875, + "epoch": 0.8974358974358975, + "grad_norm": 0.5656571630137281, + "kl": 0.0274505615234375, + "learning_rate": 1.3729842874661362e-08, + "loss": 0.0011, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 525, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.40625, + "epoch": 0.8991452991452992, + "grad_norm": 0.7540298311550256, + "kl": 0.030029296875, + "learning_rate": 1.3294452965076031e-08, + "loss": 0.0012, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 526, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.46875, + "epoch": 0.9008547008547009, + "grad_norm": 0.8815509637917851, + "kl": 0.0297698974609375, + "learning_rate": 1.2865889513213628e-08, + "loss": 0.0012, + "reward": 4.4395833015441895, + "reward_std": 0.12083335220813751, + "rewards/accuracy_reward": 2.9395835399627686, + "rewards/format_reward": 1.0, + "step": 527, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.34375, + "epoch": 0.9025641025641026, + "grad_norm": 1.063031546967266, + "kl": 0.0306854248046875, + "learning_rate": 1.2444164878608304e-08, + "loss": 0.0012, + "reward": 4.477083206176758, + "reward_std": 0.04583334922790527, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 528, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.4375, + "epoch": 0.9042735042735043, + "grad_norm": 0.8469698276822571, + "kl": 0.028076171875, + "learning_rate": 1.2029291223566412e-08, + "loss": 0.0011, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 529, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.40625, + "epoch": 0.905982905982906, + "grad_norm": 0.9273190591489977, + "kl": 0.0251617431640625, + "learning_rate": 1.162128051281594e-08, + "loss": 0.001, + "reward": 4.479166507720947, + "reward_std": 0.041666723787784576, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 530, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 387.125, + "epoch": 0.9076923076923077, + "grad_norm": 0.7771106350182588, + "kl": 0.0276641845703125, + "learning_rate": 1.1220144513161195e-08, + "loss": 0.0011, + "reward": 4.4510416984558105, + "reward_std": 0.08833983540534973, + "rewards/accuracy_reward": 2.9635417461395264, + "rewards/format_reward": 1.0, + "step": 531, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.375, + "epoch": 0.9094017094017094, + "grad_norm": 0.6664581026137811, + "kl": 0.02801513671875, + "learning_rate": 1.082589479314372e-08, + "loss": 0.0011, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 532, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 383.78125, + "epoch": 0.9111111111111111, + "grad_norm": 0.5498490702554147, + "kl": 0.0247955322265625, + "learning_rate": 1.0438542722708444e-08, + "loss": 0.001, + "reward": 4.327083587646484, + "reward_std": 0.15728820860385895, + "rewards/accuracy_reward": 2.8645832538604736, + "rewards/format_reward": 1.0, + "step": 533, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 366.21875, + "epoch": 0.9128205128205128, + "grad_norm": 0.9348854652779771, + "kl": 0.03265380859375, + "learning_rate": 1.0058099472876003e-08, + "loss": 0.0013, + "reward": 4.4197916984558105, + "reward_std": 0.14692401885986328, + "rewards/accuracy_reward": 2.9510416984558105, + "rewards/format_reward": 0.96875, + "step": 534, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 372.71875, + "epoch": 0.9145299145299145, + "grad_norm": 0.730567908020696, + "kl": 0.026275634765625, + "learning_rate": 9.684576015420275e-09, + "loss": 0.0011, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 535, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 384.375, + "epoch": 0.9162393162393162, + "grad_norm": 0.5545120278161326, + "kl": 0.02777099609375, + "learning_rate": 9.31798312255233e-09, + "loss": 0.0011, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 536, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.75, + "epoch": 0.9179487179487179, + "grad_norm": 0.7880922490785651, + "kl": 0.028564453125, + "learning_rate": 8.958331366609423e-09, + "loss": 0.0011, + "reward": 4.431250095367432, + "reward_std": 0.13749998807907104, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 0.96875, + "step": 537, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.75, + "epoch": 0.9196581196581196, + "grad_norm": 0.5125805217517629, + "kl": 0.0287017822265625, + "learning_rate": 8.605631119750295e-09, + "loss": 0.0011, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 538, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 480.46875, + "epoch": 0.9213675213675213, + "grad_norm": 1.0440304610202615, + "kl": 0.03271484375, + "learning_rate": 8.259892553655945e-09, + "loss": 0.0013, + "reward": 4.306249618530273, + "reward_std": 0.29610198736190796, + "rewards/accuracy_reward": 2.9000000953674316, + "rewards/format_reward": 0.96875, + "step": 539, + "temporal_rewards": 0.875 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 495.625, + "epoch": 0.9230769230769231, + "grad_norm": 0.805580154456117, + "kl": 0.0282135009765625, + "learning_rate": 7.921125639236415e-09, + "loss": 0.0011, + "reward": 4.376041412353516, + "reward_std": 0.2258828580379486, + "rewards/accuracy_reward": 2.913541793823242, + "rewards/format_reward": 0.96875, + "step": 540, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 392.46875, + "epoch": 0.9247863247863248, + "grad_norm": 0.9493367853110416, + "kl": 0.031402587890625, + "learning_rate": 7.589340146343077e-09, + "loss": 0.0013, + "reward": 4.4791669845581055, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.9854166507720947, + "rewards/format_reward": 1.0, + "step": 541, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.84375, + "epoch": 0.9264957264957265, + "grad_norm": 0.7381117295532499, + "kl": 0.031951904296875, + "learning_rate": 7.2645456434869965e-09, + "loss": 0.0013, + "reward": 4.483333110809326, + "reward_std": 0.03333333879709244, + "rewards/accuracy_reward": 2.9895834922790527, + "rewards/format_reward": 1.0, + "step": 542, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.0625, + "epoch": 0.9282051282051282, + "grad_norm": 1.1740784609197572, + "kl": 0.03228759765625, + "learning_rate": 6.946751497562908e-09, + "loss": 0.0013, + "reward": 4.4552083015441895, + "reward_std": 0.08958329260349274, + "rewards/accuracy_reward": 2.9739584922790527, + "rewards/format_reward": 1.0, + "step": 543, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.375, + "epoch": 0.9299145299145299, + "grad_norm": 0.5328501688828602, + "kl": 0.037506103515625, + "learning_rate": 6.635966873579063e-09, + "loss": 0.0015, + "reward": 4.488541603088379, + "reward_std": 0.013339842669665813, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 544, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.78125, + "epoch": 0.9316239316239316, + "grad_norm": 0.5149268240763294, + "kl": 0.029327392578125, + "learning_rate": 6.332200734393056e-09, + "loss": 0.0012, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 545, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.3125, + "epoch": 0.9333333333333333, + "grad_norm": 1.281928028608375, + "kl": 0.035430908203125, + "learning_rate": 6.0354618404531156e-09, + "loss": 0.0014, + "reward": 4.435416221618652, + "reward_std": 0.11367866396903992, + "rewards/accuracy_reward": 2.941666841506958, + "rewards/format_reward": 1.0, + "step": 546, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.3125, + "epoch": 0.935042735042735, + "grad_norm": 0.9215949590119717, + "kl": 0.03155517578125, + "learning_rate": 5.745758749545749e-09, + "loss": 0.0013, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 547, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 369.0625, + "epoch": 0.9367521367521368, + "grad_norm": 0.7928067990222769, + "kl": 0.033538818359375, + "learning_rate": 5.463099816548577e-09, + "loss": 0.0013, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 548, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.0, + "epoch": 0.9384615384615385, + "grad_norm": 0.4648407109039686, + "kl": 0.031524658203125, + "learning_rate": 5.187493193189784e-09, + "loss": 0.0013, + "reward": 4.432291507720947, + "reward_std": 0.12192395329475403, + "rewards/accuracy_reward": 2.9322915077209473, + "rewards/format_reward": 1.0, + "step": 549, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 389.03125, + "epoch": 0.9401709401709402, + "grad_norm": 0.5271865784036786, + "kl": 0.0309295654296875, + "learning_rate": 4.918946827812659e-09, + "loss": 0.0012, + "reward": 4.493750095367432, + "reward_std": 0.012500028125941753, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 550, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 386.0, + "epoch": 0.9418803418803419, + "grad_norm": 0.5330250509324654, + "kl": 0.028472900390625, + "learning_rate": 4.657468465146641e-09, + "loss": 0.0011, + "reward": 4.4895830154418945, + "reward_std": 0.020833352580666542, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 551, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.21875, + "epoch": 0.9435897435897436, + "grad_norm": 0.07650932815833315, + "kl": 0.0266265869140625, + "learning_rate": 4.4030656460838086e-09, + "loss": 0.0011, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 552, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.6875, + "epoch": 0.9452991452991453, + "grad_norm": 0.6953906482391564, + "kl": 0.029205322265625, + "learning_rate": 4.155745707461466e-09, + "loss": 0.0012, + "reward": 4.4895830154418945, + "reward_std": 0.020833373069763184, + "rewards/accuracy_reward": 2.9895832538604736, + "rewards/format_reward": 1.0, + "step": 553, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.6875, + "epoch": 0.947008547008547, + "grad_norm": 0.9205409521267338, + "kl": 0.0343017578125, + "learning_rate": 3.915515781850564e-09, + "loss": 0.0014, + "reward": 4.454166412353516, + "reward_std": 0.08520621806383133, + "rewards/accuracy_reward": 2.960416793823242, + "rewards/format_reward": 1.0, + "step": 554, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 382.53125, + "epoch": 0.9487179487179487, + "grad_norm": 0.4991360720043758, + "kl": 0.031890869140625, + "learning_rate": 3.6823827973499754e-09, + "loss": 0.0013, + "reward": 4.375, + "reward_std": 0.14433756470680237, + "rewards/accuracy_reward": 2.9375, + "rewards/format_reward": 0.9375, + "step": 555, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 391.25, + "epoch": 0.9504273504273504, + "grad_norm": 1.159726036834094, + "kl": 0.032623291015625, + "learning_rate": 3.4563534773866256e-09, + "loss": 0.0013, + "reward": 4.4791669845581055, + "reward_std": 0.04166668653488159, + "rewards/accuracy_reward": 2.991666793823242, + "rewards/format_reward": 1.0, + "step": 556, + "temporal_rewards": 1.0 + }, + { + "all_correct": 0.875, + "all_wrong": 0.0, + "completion_length": 385.53125, + "epoch": 0.9521367521367521, + "grad_norm": 0.8136589121786045, + "kl": 0.0294342041015625, + "learning_rate": 3.2374343405217884e-09, + "loss": 0.0012, + "reward": 4.253125190734863, + "reward_std": 0.24969010055065155, + "rewards/accuracy_reward": 2.809375047683716, + "rewards/format_reward": 1.0, + "step": 557, + "temporal_rewards": 0.875 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 378.71875, + "epoch": 0.9538461538461539, + "grad_norm": 0.782839713763375, + "kl": 0.0260162353515625, + "learning_rate": 3.025631700262876e-09, + "loss": 0.001, + "reward": 4.488541603088379, + "reward_std": 0.022916674613952637, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 558, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.4375, + "epoch": 0.9555555555555556, + "grad_norm": 0.9440948087160835, + "kl": 0.0279541015625, + "learning_rate": 2.820951664881499e-09, + "loss": 0.0011, + "reward": 4.471875190734863, + "reward_std": 0.05625001713633537, + "rewards/accuracy_reward": 2.9781250953674316, + "rewards/format_reward": 1.0, + "step": 559, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 388.21875, + "epoch": 0.9572649572649573, + "grad_norm": 0.7473828250827926, + "kl": 0.0294342041015625, + "learning_rate": 2.6234001372372193e-09, + "loss": 0.0012, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.9906249046325684, + "rewards/format_reward": 1.0, + "step": 560, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 374.09375, + "epoch": 0.958974358974359, + "grad_norm": 0.8309998398272985, + "kl": 0.028839111328125, + "learning_rate": 2.4329828146074096e-09, + "loss": 0.0012, + "reward": 4.462500095367432, + "reward_std": 0.07499998807907104, + "rewards/accuracy_reward": 2.96875, + "rewards/format_reward": 1.0, + "step": 561, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 368.4375, + "epoch": 0.9606837606837607, + "grad_norm": 0.7583845881032686, + "kl": 0.034393310546875, + "learning_rate": 2.2497051885228824e-09, + "loss": 0.0014, + "reward": 4.472916603088379, + "reward_std": 0.05416667461395264, + "rewards/accuracy_reward": 2.9791665077209473, + "rewards/format_reward": 1.0, + "step": 562, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.46875, + "epoch": 0.9623931623931624, + "grad_norm": 1.0016959232014684, + "kl": 0.029144287109375, + "learning_rate": 2.073572544609492e-09, + "loss": 0.0012, + "reward": 4.481249809265137, + "reward_std": 0.03750000521540642, + "rewards/accuracy_reward": 2.9937500953674316, + "rewards/format_reward": 1.0, + "step": 563, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.75, + "epoch": 0.9641025641025641, + "grad_norm": 0.7923931211654085, + "kl": 0.0256805419921875, + "learning_rate": 1.904589962435782e-09, + "loss": 0.001, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 564, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.875, + "epoch": 0.9658119658119658, + "grad_norm": 0.9030914923749467, + "kl": 0.037689208984375, + "learning_rate": 1.7427623153664362e-09, + "loss": 0.0015, + "reward": 4.474999904632568, + "reward_std": 0.050000011920928955, + "rewards/accuracy_reward": 2.981250047683716, + "rewards/format_reward": 1.0, + "step": 565, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.3125, + "epoch": 0.9675213675213675, + "grad_norm": 0.7288360265115156, + "kl": 0.028076171875, + "learning_rate": 1.5880942704217526e-09, + "loss": 0.0011, + "reward": 4.483333587646484, + "reward_std": 0.03333337977528572, + "rewards/accuracy_reward": 2.9833333492279053, + "rewards/format_reward": 1.0, + "step": 566, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 385.34375, + "epoch": 0.9692307692307692, + "grad_norm": 0.7433585233182205, + "kl": 0.03173828125, + "learning_rate": 1.4405902881430287e-09, + "loss": 0.0013, + "reward": 4.360416412353516, + "reward_std": 0.0223845187574625, + "rewards/accuracy_reward": 2.866666793823242, + "rewards/format_reward": 1.0, + "step": 567, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 375.78125, + "epoch": 0.9709401709401709, + "grad_norm": 0.08909391886324619, + "kl": 0.0293426513671875, + "learning_rate": 1.3002546224639145e-09, + "loss": 0.0012, + "reward": 4.375, + "reward_std": 0.0, + "rewards/accuracy_reward": 2.875, + "rewards/format_reward": 1.0, + "step": 568, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.09375, + "epoch": 0.9726495726495726, + "grad_norm": 0.10649006652034008, + "kl": 0.0347900390625, + "learning_rate": 1.167091320587843e-09, + "loss": 0.0014, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 569, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.1875, + "epoch": 0.9743589743589743, + "grad_norm": 0.8990118809823487, + "kl": 0.02984619140625, + "learning_rate": 1.0411042228711253e-09, + "loss": 0.0012, + "reward": 4.463541507720947, + "reward_std": 0.07291668653488159, + "rewards/accuracy_reward": 2.9635417461395264, + "rewards/format_reward": 1.0, + "step": 570, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 400.4375, + "epoch": 0.976068376068376, + "grad_norm": 0.8780869416335514, + "kl": 0.025970458984375, + "learning_rate": 9.222969627123433e-10, + "loss": 0.001, + "reward": 4.452083587646484, + "reward_std": 0.06897321343421936, + "rewards/accuracy_reward": 2.9520833492279053, + "rewards/format_reward": 1.0, + "step": 571, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.03125, + "epoch": 0.9777777777777777, + "grad_norm": 0.9906545174936237, + "kl": 0.0267333984375, + "learning_rate": 8.106729664475176e-10, + "loss": 0.0011, + "reward": 4.466666221618652, + "reward_std": 0.05708986893296242, + "rewards/accuracy_reward": 2.9791667461395264, + "rewards/format_reward": 1.0, + "step": 572, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.375, + "epoch": 0.9794871794871794, + "grad_norm": 1.1320410965972476, + "kl": 0.034027099609375, + "learning_rate": 7.062354532512416e-10, + "loss": 0.0014, + "reward": 4.445833206176758, + "reward_std": 0.10833339393138885, + "rewards/accuracy_reward": 2.952083110809326, + "rewards/format_reward": 1.0, + "step": 573, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 383.65625, + "epoch": 0.9811965811965812, + "grad_norm": 0.7180503459998184, + "kl": 0.028411865234375, + "learning_rate": 6.089874350439505e-10, + "loss": 0.0011, + "reward": 4.487500190734863, + "reward_std": 0.02499997615814209, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 574, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.34375, + "epoch": 0.9829059829059829, + "grad_norm": 0.558477385748197, + "kl": 0.0304718017578125, + "learning_rate": 5.189317164049633e-10, + "loss": 0.0012, + "reward": 4.243750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 2.75, + "rewards/format_reward": 1.0, + "step": 575, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 379.0625, + "epoch": 0.9846153846153847, + "grad_norm": 0.5290803575151364, + "kl": 0.02960205078125, + "learning_rate": 4.36070894491658e-10, + "loss": 0.0012, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947915077209473, + "rewards/format_reward": 1.0, + "step": 576, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 371.6875, + "epoch": 0.9863247863247864, + "grad_norm": 1.0791845979808021, + "kl": 0.027618408203125, + "learning_rate": 3.6040735896455953e-10, + "loss": 0.0011, + "reward": 4.471874713897705, + "reward_std": 0.04568374156951904, + "rewards/accuracy_reward": 2.996875047683716, + "rewards/format_reward": 1.0, + "step": 577, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 370.75, + "epoch": 0.9880341880341881, + "grad_norm": 0.884976221370628, + "kl": 0.031036376953125, + "learning_rate": 2.9194329191833953e-10, + "loss": 0.0012, + "reward": 4.467708110809326, + "reward_std": 0.06458337604999542, + "rewards/accuracy_reward": 2.9739582538604736, + "rewards/format_reward": 1.0, + "step": 578, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 380.03125, + "epoch": 0.9897435897435898, + "grad_norm": 0.5538748255975208, + "kl": 0.03741455078125, + "learning_rate": 2.3068066781908867e-10, + "loss": 0.0015, + "reward": 4.494791507720947, + "reward_std": 0.010416686534881592, + "rewards/accuracy_reward": 2.9947917461395264, + "rewards/format_reward": 1.0, + "step": 579, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 376.15625, + "epoch": 0.9914529914529915, + "grad_norm": 0.5204349719783644, + "kl": 0.03594970703125, + "learning_rate": 1.7662125344714008e-10, + "loss": 0.0014, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 580, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 377.59375, + "epoch": 0.9931623931623932, + "grad_norm": 0.07787086970335863, + "kl": 0.0251312255859375, + "learning_rate": 1.297666078462767e-10, + "loss": 0.001, + "reward": 4.5, + "reward_std": 0.0, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 581, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 373.71875, + "epoch": 0.9948717948717949, + "grad_norm": 0.9414974976233311, + "kl": 0.0290679931640625, + "learning_rate": 9.011808227865625e-11, + "loss": 0.0012, + "reward": 4.452083110809326, + "reward_std": 0.08161969482898712, + "rewards/accuracy_reward": 2.9583332538604736, + "rewards/format_reward": 1.0, + "step": 582, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.9375, + "epoch": 0.9965811965811966, + "grad_norm": 1.0306901148938268, + "kl": 0.032440185546875, + "learning_rate": 5.7676820185953434e-11, + "loss": 0.0013, + "reward": 4.484375, + "reward_std": 0.03125, + "rewards/accuracy_reward": 2.984375, + "rewards/format_reward": 1.0, + "step": 583, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 381.8125, + "epoch": 0.9982905982905983, + "grad_norm": 0.5694885078067841, + "kl": 0.02459716796875, + "learning_rate": 3.244375715633074e-11, + "loss": 0.001, + "reward": 4.493750095367432, + "reward_std": 0.012499988079071045, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 584, + "temporal_rewards": 1.0 + }, + { + "all_correct": 1.0, + "all_wrong": 0.0, + "completion_length": 344.0, + "epoch": 1.0, + "grad_norm": 0.7766650645462043, + "kl": 0.0360107421875, + "learning_rate": 1.4419620897432316e-11, + "loss": 0.0014, + "reward": 4.449999809265137, + "reward_std": 0.09999990463256836, + "rewards/accuracy_reward": 3.0, + "rewards/format_reward": 1.0, + "step": 585, + "temporal_rewards": 1.0 + } + ], + "logging_steps": 1.0, + "max_steps": 585, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 290, + "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": 1, + "trial_name": null, + "trial_params": null +}