| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08, | |
| "eval_steps": 500, | |
| "global_step": 450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 497.109375, | |
| "epoch": 0.00035555555555555557, | |
| "grad_norm": 1.5540773635891456, | |
| "kl": 0.0, | |
| "learning_rate": 7.142857142857142e-08, | |
| "loss": 0.1037, | |
| "reward": 0.328125, | |
| "reward_std": 0.4159187823534012, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.265625, | |
| "step": 2 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 532.40625, | |
| "epoch": 0.0007111111111111111, | |
| "grad_norm": 1.2357520393415342, | |
| "kl": 0.0007323026657104492, | |
| "learning_rate": 1.4285714285714285e-07, | |
| "loss": 0.1034, | |
| "reward": 0.25, | |
| "reward_std": 0.3582531735301018, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.1875, | |
| "step": 4 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 501.78125, | |
| "epoch": 0.0010666666666666667, | |
| "grad_norm": 1.207958912839064, | |
| "kl": 0.0007942914962768555, | |
| "learning_rate": 2.1428571428571426e-07, | |
| "loss": 0.1091, | |
| "reward": 0.234375, | |
| "reward_std": 0.3534187823534012, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.203125, | |
| "step": 6 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 472.109375, | |
| "epoch": 0.0014222222222222223, | |
| "grad_norm": 1.6667490438178998, | |
| "kl": 0.000766754150390625, | |
| "learning_rate": 2.857142857142857e-07, | |
| "loss": 0.1158, | |
| "reward": 0.390625, | |
| "reward_std": 0.4387340322136879, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.25, | |
| "step": 8 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 508.5, | |
| "epoch": 0.0017777777777777779, | |
| "grad_norm": 1.7217421147986447, | |
| "kl": 0.00079345703125, | |
| "learning_rate": 3.5714285714285716e-07, | |
| "loss": 0.0744, | |
| "reward": 0.375, | |
| "reward_std": 0.48542676120996475, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.34375, | |
| "step": 10 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 563.03125, | |
| "epoch": 0.0021333333333333334, | |
| "grad_norm": 1.4870770624622829, | |
| "kl": 0.0007028579711914062, | |
| "learning_rate": 4.285714285714285e-07, | |
| "loss": 0.16, | |
| "reward": 0.28125, | |
| "reward_std": 0.3846687823534012, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.25, | |
| "step": 12 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 463.609375, | |
| "epoch": 0.002488888888888889, | |
| "grad_norm": 1.7635576195780378, | |
| "kl": 0.0008461475372314453, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0834, | |
| "reward": 0.328125, | |
| "reward_std": 0.46167195588350296, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.296875, | |
| "step": 14 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 485.484375, | |
| "epoch": 0.0028444444444444446, | |
| "grad_norm": 1.5351814757263196, | |
| "kl": 0.0009121894836425781, | |
| "learning_rate": 4.999740409224932e-07, | |
| "loss": 0.1649, | |
| "reward": 0.4375, | |
| "reward_std": 0.45151595771312714, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.328125, | |
| "step": 16 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 473.578125, | |
| "epoch": 0.0032, | |
| "grad_norm": 1.790949387851571, | |
| "kl": 0.0012161731719970703, | |
| "learning_rate": 4.998961690809627e-07, | |
| "loss": 0.1101, | |
| "reward": 0.40625, | |
| "reward_std": 0.5236847400665283, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.359375, | |
| "step": 18 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 437.8125, | |
| "epoch": 0.0035555555555555557, | |
| "grad_norm": 1.8537703194309358, | |
| "kl": 0.0023336410522460938, | |
| "learning_rate": 4.997664006472578e-07, | |
| "loss": 0.1244, | |
| "reward": 0.546875, | |
| "reward_std": 0.42558756470680237, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.5, | |
| "step": 20 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 406.515625, | |
| "epoch": 0.003911111111111111, | |
| "grad_norm": 1.7394706416500154, | |
| "kl": 0.0038557052612304688, | |
| "learning_rate": 4.995847625707292e-07, | |
| "loss": 0.0125, | |
| "reward": 0.59375, | |
| "reward_std": 0.5879059880971909, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.53125, | |
| "step": 22 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 459.703125, | |
| "epoch": 0.004266666666666667, | |
| "grad_norm": 1.454676166117581, | |
| "kl": 0.004199504852294922, | |
| "learning_rate": 4.993512925726318e-07, | |
| "loss": 0.0336, | |
| "reward": 0.625, | |
| "reward_std": 0.40400634706020355, | |
| "rewards/equation_reward_func": 0.015625, | |
| "rewards/format_reward_func": 0.609375, | |
| "step": 24 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 481.953125, | |
| "epoch": 0.004622222222222222, | |
| "grad_norm": 1.7838057007741426, | |
| "kl": 0.0038619041442871094, | |
| "learning_rate": 4.990660391382923e-07, | |
| "loss": 0.0952, | |
| "reward": 0.765625, | |
| "reward_std": 0.34194982051849365, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.734375, | |
| "step": 26 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 443.234375, | |
| "epoch": 0.004977777777777778, | |
| "grad_norm": 2.0657041070771247, | |
| "kl": 0.00693511962890625, | |
| "learning_rate": 4.987290615070384e-07, | |
| "loss": 0.0515, | |
| "reward": 0.796875, | |
| "reward_std": 0.5276496410369873, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.71875, | |
| "step": 28 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 524.359375, | |
| "epoch": 0.005333333333333333, | |
| "grad_norm": 1.5368658141890732, | |
| "kl": 0.007770538330078125, | |
| "learning_rate": 4.983404296598978e-07, | |
| "loss": 0.1967, | |
| "reward": 0.75, | |
| "reward_std": 0.5364577993750572, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.6875, | |
| "step": 30 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 455.6875, | |
| "epoch": 0.005688888888888889, | |
| "grad_norm": 1.7349608259853453, | |
| "kl": 0.009317398071289062, | |
| "learning_rate": 4.979002243050646e-07, | |
| "loss": 0.1176, | |
| "reward": 0.875, | |
| "reward_std": 0.47960907220840454, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.765625, | |
| "step": 32 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 477.53125, | |
| "epoch": 0.006044444444444444, | |
| "grad_norm": 1.2174715114943861, | |
| "kl": 0.011419296264648438, | |
| "learning_rate": 4.974085368611381e-07, | |
| "loss": 0.0599, | |
| "reward": 0.96875, | |
| "reward_std": 0.3978152498602867, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.828125, | |
| "step": 34 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 544.34375, | |
| "epoch": 0.0064, | |
| "grad_norm": 0.9373070959302399, | |
| "kl": 0.009111404418945312, | |
| "learning_rate": 4.968654694381379e-07, | |
| "loss": 0.0615, | |
| "reward": 0.8125, | |
| "reward_std": 0.35645299404859543, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.78125, | |
| "step": 36 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 522.078125, | |
| "epoch": 0.0067555555555555554, | |
| "grad_norm": 1.69272651339043, | |
| "kl": 0.01029205322265625, | |
| "learning_rate": 4.962711348162987e-07, | |
| "loss": 0.1866, | |
| "reward": 0.84375, | |
| "reward_std": 0.5025907382369041, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.71875, | |
| "step": 38 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 507.984375, | |
| "epoch": 0.0071111111111111115, | |
| "grad_norm": 1.2015156610010047, | |
| "kl": 0.012607574462890625, | |
| "learning_rate": 4.956256564226487e-07, | |
| "loss": 0.0963, | |
| "reward": 1.015625, | |
| "reward_std": 0.407927505671978, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.875, | |
| "step": 40 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 444.296875, | |
| "epoch": 0.007466666666666667, | |
| "grad_norm": 1.1020450998312568, | |
| "kl": 0.015117645263671875, | |
| "learning_rate": 4.949291683053768e-07, | |
| "loss": 0.1049, | |
| "reward": 0.953125, | |
| "reward_std": 0.38331207633018494, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.875, | |
| "step": 42 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 467.03125, | |
| "epoch": 0.007822222222222222, | |
| "grad_norm": 1.4356691747538686, | |
| "kl": 0.015163421630859375, | |
| "learning_rate": 4.941818151059955e-07, | |
| "loss": 0.0563, | |
| "reward": 1.0, | |
| "reward_std": 0.31069982051849365, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.859375, | |
| "step": 44 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 466.1875, | |
| "epoch": 0.008177777777777779, | |
| "grad_norm": 1.1138441770229877, | |
| "kl": 0.014316558837890625, | |
| "learning_rate": 4.933837520293017e-07, | |
| "loss": 0.0513, | |
| "reward": 0.953125, | |
| "reward_std": 0.20728103816509247, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.921875, | |
| "step": 46 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 483.21875, | |
| "epoch": 0.008533333333333334, | |
| "grad_norm": 1.5073854647894624, | |
| "kl": 0.0155029296875, | |
| "learning_rate": 4.925351448111454e-07, | |
| "loss": 0.0312, | |
| "reward": 0.984375, | |
| "reward_std": 0.37053901702165604, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.875, | |
| "step": 48 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 441.59375, | |
| "epoch": 0.008888888888888889, | |
| "grad_norm": 1.0379492454910968, | |
| "kl": 0.015254974365234375, | |
| "learning_rate": 4.91636169684011e-07, | |
| "loss": 0.0637, | |
| "reward": 1.015625, | |
| "reward_std": 0.24336542934179306, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.921875, | |
| "step": 50 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 491.46875, | |
| "epoch": 0.009244444444444444, | |
| "grad_norm": 1.3198501259898483, | |
| "kl": 0.016269683837890625, | |
| "learning_rate": 4.906870133404186e-07, | |
| "loss": 0.062, | |
| "reward": 1.0, | |
| "reward_std": 0.31069982051849365, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.90625, | |
| "step": 52 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 450.234375, | |
| "epoch": 0.0096, | |
| "grad_norm": 1.1484527235656556, | |
| "kl": 0.021270751953125, | |
| "learning_rate": 4.896878728941531e-07, | |
| "loss": 0.0533, | |
| "reward": 1.015625, | |
| "reward_std": 0.408423587679863, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.875, | |
| "step": 54 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 408.03125, | |
| "epoch": 0.009955555555555556, | |
| "grad_norm": 1.0788691812319753, | |
| "kl": 0.0235595703125, | |
| "learning_rate": 4.886389558393284e-07, | |
| "loss": -0.0358, | |
| "reward": 1.015625, | |
| "reward_std": 0.20728103816509247, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 56 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 487.09375, | |
| "epoch": 0.010311111111111111, | |
| "grad_norm": 1.4656646934435877, | |
| "kl": 0.02265167236328125, | |
| "learning_rate": 4.875404800072976e-07, | |
| "loss": 0.0995, | |
| "reward": 0.953125, | |
| "reward_std": 0.26978103816509247, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.90625, | |
| "step": 58 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 435.6875, | |
| "epoch": 0.010666666666666666, | |
| "grad_norm": 1.1291328060832448, | |
| "kl": 0.02149200439453125, | |
| "learning_rate": 4.86392673521415e-07, | |
| "loss": 0.0992, | |
| "reward": 1.046875, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 60 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 453.59375, | |
| "epoch": 0.011022222222222221, | |
| "grad_norm": 0.7889399494362057, | |
| "kl": 0.025543212890625, | |
| "learning_rate": 4.851957747496606e-07, | |
| "loss": 0.0177, | |
| "reward": 1.03125, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 62 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 413.859375, | |
| "epoch": 0.011377777777777778, | |
| "grad_norm": 0.8220739984789721, | |
| "kl": 0.02587890625, | |
| "learning_rate": 4.839500322551386e-07, | |
| "loss": 0.0939, | |
| "reward": 1.0625, | |
| "reward_std": 0.22706207633018494, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 64 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 394.328125, | |
| "epoch": 0.011733333333333333, | |
| "grad_norm": 0.3782612004884283, | |
| "kl": 0.02734375, | |
| "learning_rate": 4.826557047444563e-07, | |
| "loss": 0.0026, | |
| "reward": 1.015625, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 66 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 437.765625, | |
| "epoch": 0.012088888888888889, | |
| "grad_norm": 1.0169159547009763, | |
| "kl": 0.0303802490234375, | |
| "learning_rate": 4.813130610139993e-07, | |
| "loss": 0.1134, | |
| "reward": 1.09375, | |
| "reward_std": 0.34678421169519424, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 0.9375, | |
| "step": 68 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 395.046875, | |
| "epoch": 0.012444444444444444, | |
| "grad_norm": 1.1708246723062659, | |
| "kl": 0.0349884033203125, | |
| "learning_rate": 4.799223798941089e-07, | |
| "loss": 0.0274, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 70 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 409.109375, | |
| "epoch": 0.0128, | |
| "grad_norm": 0.9619605643432636, | |
| "kl": 0.041015625, | |
| "learning_rate": 4.78483950191177e-07, | |
| "loss": 0.0364, | |
| "reward": 1.03125, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 72 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 406.109375, | |
| "epoch": 0.013155555555555556, | |
| "grad_norm": 1.5109017153953268, | |
| "kl": 0.036956787109375, | |
| "learning_rate": 4.769980706276687e-07, | |
| "loss": 0.1356, | |
| "reward": 1.15625, | |
| "reward_std": 0.3713996410369873, | |
| "rewards/equation_reward_func": 0.203125, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 74 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 401.671875, | |
| "epoch": 0.013511111111111111, | |
| "grad_norm": 1.4832479712654314, | |
| "kl": 0.0571441650390625, | |
| "learning_rate": 4.7546504978008595e-07, | |
| "loss": 0.0329, | |
| "reward": 1.078125, | |
| "reward_std": 0.15625, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 76 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 383.390625, | |
| "epoch": 0.013866666666666666, | |
| "grad_norm": 1.6506361354745738, | |
| "kl": 0.04904937744140625, | |
| "learning_rate": 4.738852060148848e-07, | |
| "loss": 0.0359, | |
| "reward": 1.109375, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 78 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 428.203125, | |
| "epoch": 0.014222222222222223, | |
| "grad_norm": 1.2805072360716652, | |
| "kl": 0.0862884521484375, | |
| "learning_rate": 4.722588674223593e-07, | |
| "loss": 0.0975, | |
| "reward": 1.0625, | |
| "reward_std": 0.2596687823534012, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 80 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 334.765625, | |
| "epoch": 0.014577777777777778, | |
| "grad_norm": 1.5003659203947817, | |
| "kl": 0.083343505859375, | |
| "learning_rate": 4.70586371748506e-07, | |
| "loss": 0.0372, | |
| "reward": 1.140625, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 82 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 375.625, | |
| "epoch": 0.014933333333333333, | |
| "grad_norm": 1.4069013627610942, | |
| "kl": 0.0662841796875, | |
| "learning_rate": 4.6886806632488363e-07, | |
| "loss": 0.0729, | |
| "reward": 1.140625, | |
| "reward_std": 0.21694982051849365, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 84 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 374.171875, | |
| "epoch": 0.015288888888888888, | |
| "grad_norm": 1.533899888374318, | |
| "kl": 0.11968994140625, | |
| "learning_rate": 4.6710430799648143e-07, | |
| "loss": 0.0581, | |
| "reward": 1.15625, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 86 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 338.90625, | |
| "epoch": 0.015644444444444443, | |
| "grad_norm": 1.213080736189347, | |
| "kl": 0.24658203125, | |
| "learning_rate": 4.652954630476127e-07, | |
| "loss": 0.0379, | |
| "reward": 1.078125, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 88 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 371.796875, | |
| "epoch": 0.016, | |
| "grad_norm": 1.709623394747104, | |
| "kl": 1.02197265625, | |
| "learning_rate": 4.6344190712584713e-07, | |
| "loss": -0.0009, | |
| "reward": 1.15625, | |
| "reward_std": 0.23325317353010178, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 90 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 372.4375, | |
| "epoch": 0.016355555555555557, | |
| "grad_norm": 1.1039741000193948, | |
| "kl": 0.51763916015625, | |
| "learning_rate": 4.615440251639995e-07, | |
| "loss": 0.0138, | |
| "reward": 1.03125, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 92 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 383.03125, | |
| "epoch": 0.01671111111111111, | |
| "grad_norm": 1.0615946335422568, | |
| "kl": 0.32763671875, | |
| "learning_rate": 4.596022113001894e-07, | |
| "loss": 0.0168, | |
| "reward": 1.0625, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 94 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 387.046875, | |
| "epoch": 0.017066666666666667, | |
| "grad_norm": 1.0567721975237463, | |
| "kl": 0.186248779296875, | |
| "learning_rate": 4.576168687959895e-07, | |
| "loss": 0.014, | |
| "reward": 1.125, | |
| "reward_std": 0.17075317353010178, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 96 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 357.375, | |
| "epoch": 0.01742222222222222, | |
| "grad_norm": 117.02412491992783, | |
| "kl": 8.14306640625, | |
| "learning_rate": 4.555884099526793e-07, | |
| "loss": 0.0505, | |
| "reward": 1.09375, | |
| "reward_std": 0.13466878235340118, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 98 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 361.265625, | |
| "epoch": 0.017777777777777778, | |
| "grad_norm": 1.2109959204361576, | |
| "kl": 0.1463623046875, | |
| "learning_rate": 4.5351725602562174e-07, | |
| "loss": -0.0073, | |
| "reward": 1.125, | |
| "reward_std": 0.24819982051849365, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 100 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 374.34375, | |
| "epoch": 0.018133333333333335, | |
| "grad_norm": 1.3629062892707553, | |
| "kl": 0.22503662109375, | |
| "learning_rate": 4.514038371367791e-07, | |
| "loss": 0.0304, | |
| "reward": 1.109375, | |
| "reward_std": 0.16591878235340118, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 102 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 355.671875, | |
| "epoch": 0.018488888888888888, | |
| "grad_norm": 1.066354910378942, | |
| "kl": 0.237884521484375, | |
| "learning_rate": 4.4924859218538936e-07, | |
| "loss": 0.0266, | |
| "reward": 1.0625, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 104 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 346.40625, | |
| "epoch": 0.018844444444444445, | |
| "grad_norm": 1.244076703781482, | |
| "kl": 0.2713623046875, | |
| "learning_rate": 4.470519687568185e-07, | |
| "loss": 0.0573, | |
| "reward": 1.15625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 106 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 343.875, | |
| "epoch": 0.0192, | |
| "grad_norm": 1.5317514782871564, | |
| "kl": 0.193603515625, | |
| "learning_rate": 4.4481442302960923e-07, | |
| "loss": 0.0681, | |
| "reward": 1.125, | |
| "reward_std": 0.22178421169519424, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 108 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 357.453125, | |
| "epoch": 0.019555555555555555, | |
| "grad_norm": 1.3615194783712425, | |
| "kl": 0.257049560546875, | |
| "learning_rate": 4.4253641968074505e-07, | |
| "loss": 0.0324, | |
| "reward": 1.046875, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 110 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 330.875, | |
| "epoch": 0.019911111111111112, | |
| "grad_norm": 1.6404324826318923, | |
| "kl": 0.42608642578125, | |
| "learning_rate": 4.402184317891501e-07, | |
| "loss": 0.0444, | |
| "reward": 1.15625, | |
| "reward_std": 0.26933756470680237, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 112 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 327.734375, | |
| "epoch": 0.020266666666666665, | |
| "grad_norm": 1.1970449285988152, | |
| "kl": 0.30352783203125, | |
| "learning_rate": 4.37860940737443e-07, | |
| "loss": 0.0289, | |
| "reward": 1.03125, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 114 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 345.984375, | |
| "epoch": 0.020622222222222222, | |
| "grad_norm": 3.6830472868585975, | |
| "kl": 0.693603515625, | |
| "learning_rate": 4.354644361119671e-07, | |
| "loss": 0.0444, | |
| "reward": 1.03125, | |
| "reward_std": 0.15358919650316238, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 116 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 317.34375, | |
| "epoch": 0.02097777777777778, | |
| "grad_norm": 0.8803960644083081, | |
| "kl": 0.40740966796875, | |
| "learning_rate": 4.3302941560111716e-07, | |
| "loss": 0.0669, | |
| "reward": 1.09375, | |
| "reward_std": 0.1875, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 118 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 325.359375, | |
| "epoch": 0.021333333333333333, | |
| "grad_norm": 1.7222894595317144, | |
| "kl": 0.9599609375, | |
| "learning_rate": 4.3055638489198236e-07, | |
| "loss": 0.021, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 120 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 343.03125, | |
| "epoch": 0.02168888888888889, | |
| "grad_norm": 3.1092809656603206, | |
| "kl": 1.03076171875, | |
| "learning_rate": 4.280458575653296e-07, | |
| "loss": 0.0242, | |
| "reward": 1.09375, | |
| "reward_std": 0.1875, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 122 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 311.359375, | |
| "epoch": 0.022044444444444443, | |
| "grad_norm": 2.3189720464471533, | |
| "kl": 0.631591796875, | |
| "learning_rate": 4.2549835498894665e-07, | |
| "loss": 0.0467, | |
| "reward": 1.09375, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 124 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 307.046875, | |
| "epoch": 0.0224, | |
| "grad_norm": 0.7441216613827443, | |
| "kl": 0.5086669921875, | |
| "learning_rate": 4.229144062093679e-07, | |
| "loss": -0.0085, | |
| "reward": 1.140625, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 126 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 291.640625, | |
| "epoch": 0.022755555555555557, | |
| "grad_norm": 1.7398896558997667, | |
| "kl": 0.682373046875, | |
| "learning_rate": 4.2029454784200675e-07, | |
| "loss": 0.0178, | |
| "reward": 1.03125, | |
| "reward_std": 0.0625, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 128 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 286.921875, | |
| "epoch": 0.02311111111111111, | |
| "grad_norm": 1.4445137095871459, | |
| "kl": 0.564208984375, | |
| "learning_rate": 4.1763932395971433e-07, | |
| "loss": 0.0149, | |
| "reward": 1.078125, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 130 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 299.28125, | |
| "epoch": 0.023466666666666667, | |
| "grad_norm": 1.082942893318675, | |
| "kl": 0.3787841796875, | |
| "learning_rate": 4.1494928597979117e-07, | |
| "loss": 0.0099, | |
| "reward": 1.140625, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 132 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 298.84375, | |
| "epoch": 0.023822222222222224, | |
| "grad_norm": 2.813667402859087, | |
| "kl": 1.5950927734375, | |
| "learning_rate": 4.122249925494726e-07, | |
| "loss": 0.0468, | |
| "reward": 1.21875, | |
| "reward_std": 0.2957531735301018, | |
| "rewards/equation_reward_func": 0.21875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 134 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 304.90625, | |
| "epoch": 0.024177777777777777, | |
| "grad_norm": 1.6133284823417406, | |
| "kl": 0.628173828125, | |
| "learning_rate": 4.094670094299131e-07, | |
| "loss": 0.0012, | |
| "reward": 1.0625, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 136 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 330.609375, | |
| "epoch": 0.024533333333333334, | |
| "grad_norm": 1.5618980968427136, | |
| "kl": 0.620361328125, | |
| "learning_rate": 4.066759093786931e-07, | |
| "loss": 0.0299, | |
| "reward": 1.046875, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 138 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 315.84375, | |
| "epoch": 0.024888888888888887, | |
| "grad_norm": 0.7496675007517801, | |
| "kl": 0.5653076171875, | |
| "learning_rate": 4.038522720308732e-07, | |
| "loss": 0.0383, | |
| "reward": 1.046875, | |
| "reward_std": 0.06733439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 140 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 279.46875, | |
| "epoch": 0.025244444444444444, | |
| "grad_norm": 1.623499251970735, | |
| "kl": 0.6290283203125, | |
| "learning_rate": 4.009966837786194e-07, | |
| "loss": 0.0041, | |
| "reward": 1.046875, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 142 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 301.1875, | |
| "epoch": 0.0256, | |
| "grad_norm": 1.5816454578936159, | |
| "kl": 0.6375732421875, | |
| "learning_rate": 3.981097376494259e-07, | |
| "loss": -0.0138, | |
| "reward": 1.0625, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 144 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 286.390625, | |
| "epoch": 0.025955555555555555, | |
| "grad_norm": 0.9976268635346477, | |
| "kl": 0.594970703125, | |
| "learning_rate": 3.951920331829592e-07, | |
| "loss": 0.0283, | |
| "reward": 1.15625, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 146 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 290.21875, | |
| "epoch": 0.02631111111111111, | |
| "grad_norm": 1.453799485818376, | |
| "kl": 1.276611328125, | |
| "learning_rate": 3.922441763065506e-07, | |
| "loss": 0.0691, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 148 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 330.203125, | |
| "epoch": 0.02666666666666667, | |
| "grad_norm": 1.4585609222742653, | |
| "kl": 0.7332763671875, | |
| "learning_rate": 3.8926677920936093e-07, | |
| "loss": 0.0235, | |
| "reward": 1.0625, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 150 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 314.703125, | |
| "epoch": 0.027022222222222222, | |
| "grad_norm": 1.2450868511160544, | |
| "kl": 0.74951171875, | |
| "learning_rate": 3.862604602152464e-07, | |
| "loss": 0.0255, | |
| "reward": 1.03125, | |
| "reward_std": 0.0625, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 152 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 313.203125, | |
| "epoch": 0.02737777777777778, | |
| "grad_norm": 1.3495788554258188, | |
| "kl": 0.805908203125, | |
| "learning_rate": 3.8322584365434934e-07, | |
| "loss": 0.0098, | |
| "reward": 1.03125, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 154 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 324.9375, | |
| "epoch": 0.027733333333333332, | |
| "grad_norm": 2.1466128173407117, | |
| "kl": 1.1510009765625, | |
| "learning_rate": 3.8016355973344173e-07, | |
| "loss": 0.09, | |
| "reward": 1.140625, | |
| "reward_std": 0.23808756470680237, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 156 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 324.859375, | |
| "epoch": 0.02808888888888889, | |
| "grad_norm": 2.204629350726093, | |
| "kl": 1.16015625, | |
| "learning_rate": 3.7707424440504863e-07, | |
| "loss": 0.0803, | |
| "reward": 1.140625, | |
| "reward_std": 0.2548343911767006, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 158 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 302.859375, | |
| "epoch": 0.028444444444444446, | |
| "grad_norm": 1.8398150430903868, | |
| "kl": 1.251708984375, | |
| "learning_rate": 3.739585392353787e-07, | |
| "loss": 0.0197, | |
| "reward": 1.15625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 160 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 292.640625, | |
| "epoch": 0.0288, | |
| "grad_norm": 2.13847808834117, | |
| "kl": 1.3173828125, | |
| "learning_rate": 3.7081709127108767e-07, | |
| "loss": 0.0155, | |
| "reward": 1.1875, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.1875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 162 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 261.234375, | |
| "epoch": 0.029155555555555556, | |
| "grad_norm": 3.506581199733204, | |
| "kl": 1.736328125, | |
| "learning_rate": 3.6765055290490513e-07, | |
| "loss": -0.0109, | |
| "reward": 1.171875, | |
| "reward_std": 0.28125, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 164 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 302.953125, | |
| "epoch": 0.02951111111111111, | |
| "grad_norm": 2.5724923821581025, | |
| "kl": 6.1572265625, | |
| "learning_rate": 3.644595817401501e-07, | |
| "loss": 0.0332, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 166 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 280.03125, | |
| "epoch": 0.029866666666666666, | |
| "grad_norm": 1.5126543710719036, | |
| "kl": 1.48974609375, | |
| "learning_rate": 3.6124484045416483e-07, | |
| "loss": 0.0267, | |
| "reward": 1.140625, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 168 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 302.546875, | |
| "epoch": 0.030222222222222223, | |
| "grad_norm": 3.4982226449385156, | |
| "kl": 3.032470703125, | |
| "learning_rate": 3.580069966606949e-07, | |
| "loss": 0.0559, | |
| "reward": 1.03125, | |
| "reward_std": 0.1875, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 170 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 334.515625, | |
| "epoch": 0.030577777777777777, | |
| "grad_norm": 3.0985982569829638, | |
| "kl": 14.5616455078125, | |
| "learning_rate": 3.547467227712444e-07, | |
| "loss": 0.0302, | |
| "reward": 1.03125, | |
| "reward_std": 0.23853103816509247, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 172 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 300.71875, | |
| "epoch": 0.030933333333333334, | |
| "grad_norm": 10.688658199362251, | |
| "kl": 3.142333984375, | |
| "learning_rate": 3.5146469585543386e-07, | |
| "loss": -0.0049, | |
| "reward": 1.171875, | |
| "reward_std": 0.16591878235340118, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 174 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 292.484375, | |
| "epoch": 0.03128888888888889, | |
| "grad_norm": 6.967303370780864, | |
| "kl": 4.5751953125, | |
| "learning_rate": 3.481615975003922e-07, | |
| "loss": 0.0089, | |
| "reward": 1.125, | |
| "reward_std": 0.13466878235340118, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 176 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 292.1875, | |
| "epoch": 0.03164444444444445, | |
| "grad_norm": 2.379129578155761, | |
| "kl": 1.24462890625, | |
| "learning_rate": 3.448381136692089e-07, | |
| "loss": 0.0412, | |
| "reward": 1.15625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 178 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 336.0625, | |
| "epoch": 0.032, | |
| "grad_norm": 1.8115427356791605, | |
| "kl": 0.7757568359375, | |
| "learning_rate": 3.4149493455847897e-07, | |
| "loss": 0.0274, | |
| "reward": 1.015625, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 180 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 297.34375, | |
| "epoch": 0.032355555555555554, | |
| "grad_norm": 2.279790073120627, | |
| "kl": 0.8720703125, | |
| "learning_rate": 3.3813275445496766e-07, | |
| "loss": 0.0637, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 182 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 311.6875, | |
| "epoch": 0.032711111111111114, | |
| "grad_norm": 1.9111976240917574, | |
| "kl": 1.3289794921875, | |
| "learning_rate": 3.347522715914262e-07, | |
| "loss": -0.0126, | |
| "reward": 1.0625, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 184 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 284.0625, | |
| "epoch": 0.03306666666666667, | |
| "grad_norm": 48.68973900874978, | |
| "kl": 8.955810546875, | |
| "learning_rate": 3.313541880015877e-07, | |
| "loss": 0.0674, | |
| "reward": 1.171875, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 186 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 334.1875, | |
| "epoch": 0.03342222222222222, | |
| "grad_norm": 0.3837227064051435, | |
| "kl": 0.7593994140625, | |
| "learning_rate": 3.279392093743747e-07, | |
| "loss": 0.0074, | |
| "reward": 1.015625, | |
| "reward_std": 0.03125, | |
| "rewards/equation_reward_func": 0.015625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 188 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 276.140625, | |
| "epoch": 0.033777777777777775, | |
| "grad_norm": 2.2048786067003374, | |
| "kl": 0.738037109375, | |
| "learning_rate": 3.245080449073459e-07, | |
| "loss": 0.0079, | |
| "reward": 1.1875, | |
| "reward_std": 0.2596687823534012, | |
| "rewards/equation_reward_func": 0.1875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 190 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 324.796875, | |
| "epoch": 0.034133333333333335, | |
| "grad_norm": 1.4988476602299554, | |
| "kl": 0.73046875, | |
| "learning_rate": 3.210614071594162e-07, | |
| "loss": 0.0788, | |
| "reward": 1.125, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 192 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 260.8125, | |
| "epoch": 0.03448888888888889, | |
| "grad_norm": 3.9875059912502744, | |
| "kl": 1.3214111328125, | |
| "learning_rate": 3.1760001190287695e-07, | |
| "loss": 0.0088, | |
| "reward": 1.140625, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 194 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 278.03125, | |
| "epoch": 0.03484444444444444, | |
| "grad_norm": 2.1591229522344846, | |
| "kl": 0.6678466796875, | |
| "learning_rate": 3.141245779747502e-07, | |
| "loss": 0.0746, | |
| "reward": 1.203125, | |
| "reward_std": 0.30058756470680237, | |
| "rewards/equation_reward_func": 0.203125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 196 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 333.765625, | |
| "epoch": 0.0352, | |
| "grad_norm": 3.410997217745853, | |
| "kl": 0.9820556640625, | |
| "learning_rate": 3.106358271275056e-07, | |
| "loss": 0.0738, | |
| "reward": 1.09375, | |
| "reward_std": 0.2596687823534012, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 198 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 327.84375, | |
| "epoch": 0.035555555555555556, | |
| "grad_norm": 2.2453024821477534, | |
| "kl": 1.880126953125, | |
| "learning_rate": 3.0713448387917227e-07, | |
| "loss": -0.0002, | |
| "reward": 1.046875, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 311.515625, | |
| "epoch": 0.03591111111111111, | |
| "grad_norm": 2.3752697048728186, | |
| "kl": 2.144775390625, | |
| "learning_rate": 3.0362127536287636e-07, | |
| "loss": 0.0192, | |
| "reward": 1.09375, | |
| "reward_std": 0.18569982051849365, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 202 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 284.3125, | |
| "epoch": 0.03626666666666667, | |
| "grad_norm": 135.00953365889976, | |
| "kl": 28.689697265625, | |
| "learning_rate": 3.0009693117583523e-07, | |
| "loss": 0.0403, | |
| "reward": 1.0625, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 204 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 299.71875, | |
| "epoch": 0.03662222222222222, | |
| "grad_norm": 1.8674915516374395, | |
| "kl": 2.016845703125, | |
| "learning_rate": 2.965621832278401e-07, | |
| "loss": 0.049, | |
| "reward": 1.109375, | |
| "reward_std": 0.13950317353010178, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 206 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 334.296875, | |
| "epoch": 0.036977777777777776, | |
| "grad_norm": 1.6829162960798076, | |
| "kl": 2.673828125, | |
| "learning_rate": 2.9301776558925875e-07, | |
| "loss": 0.0532, | |
| "reward": 1.0625, | |
| "reward_std": 0.12717358767986298, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 208 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 342.4375, | |
| "epoch": 0.037333333333333336, | |
| "grad_norm": 0.8086021793065252, | |
| "kl": 1.893310546875, | |
| "learning_rate": 2.894644143385885e-07, | |
| "loss": 0.0337, | |
| "reward": 1.078125, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 210 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 307.765625, | |
| "epoch": 0.03768888888888889, | |
| "grad_norm": 2.4643376957229237, | |
| "kl": 1.91650390625, | |
| "learning_rate": 2.859028674095937e-07, | |
| "loss": 0.0027, | |
| "reward": 1.203125, | |
| "reward_std": 0.30058756470680237, | |
| "rewards/equation_reward_func": 0.203125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 212 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 298.515625, | |
| "epoch": 0.03804444444444444, | |
| "grad_norm": 3.029946510557144, | |
| "kl": 1.754638671875, | |
| "learning_rate": 2.823338644380566e-07, | |
| "loss": 0.022, | |
| "reward": 1.15625, | |
| "reward_std": 0.17075317353010178, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 214 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 315.765625, | |
| "epoch": 0.0384, | |
| "grad_norm": 1.810123541436162, | |
| "kl": 1.518310546875, | |
| "learning_rate": 2.7875814660817504e-07, | |
| "loss": -0.0159, | |
| "reward": 1.078125, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 216 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 269.6875, | |
| "epoch": 0.03875555555555556, | |
| "grad_norm": 2.337741190093351, | |
| "kl": 1.03564453125, | |
| "learning_rate": 2.751764564986396e-07, | |
| "loss": 0.0113, | |
| "reward": 1.140625, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 218 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 329.484375, | |
| "epoch": 0.03911111111111111, | |
| "grad_norm": 0.7489189831580635, | |
| "kl": 1.310302734375, | |
| "learning_rate": 2.715895379284194e-07, | |
| "loss": 0.0084, | |
| "reward": 1.015625, | |
| "reward_std": 0.03125, | |
| "rewards/equation_reward_func": 0.015625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 220 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 306.265625, | |
| "epoch": 0.039466666666666664, | |
| "grad_norm": 0.8093400893734075, | |
| "kl": 1.0770263671875, | |
| "learning_rate": 2.6799813580229174e-07, | |
| "loss": 0.0144, | |
| "reward": 1.015625, | |
| "reward_std": 0.03125, | |
| "rewards/equation_reward_func": 0.015625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 222 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 286.640625, | |
| "epoch": 0.039822222222222224, | |
| "grad_norm": 1.6055403629730776, | |
| "kl": 1.1387939453125, | |
| "learning_rate": 2.6440299595614606e-07, | |
| "loss": 0.0386, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 224 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 304.046875, | |
| "epoch": 0.04017777777777778, | |
| "grad_norm": 2.0302718515892164, | |
| "kl": 1.0672607421875, | |
| "learning_rate": 2.6080486500209347e-07, | |
| "loss": 0.0873, | |
| "reward": 1.1875, | |
| "reward_std": 0.2596687823534012, | |
| "rewards/equation_reward_func": 0.203125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 226 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 332.578125, | |
| "epoch": 0.04053333333333333, | |
| "grad_norm": 109.08080062943176, | |
| "kl": 12.7587890625, | |
| "learning_rate": 2.572044901734166e-07, | |
| "loss": 0.0036, | |
| "reward": 1.09375, | |
| "reward_std": 0.14961542934179306, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 228 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 361.203125, | |
| "epoch": 0.04088888888888889, | |
| "grad_norm": 3.7661219080037074, | |
| "kl": 2.75927734375, | |
| "learning_rate": 2.536026191693893e-07, | |
| "loss": 0.1178, | |
| "reward": 1.140625, | |
| "reward_std": 0.2645031735301018, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 230 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 325.75, | |
| "epoch": 0.041244444444444445, | |
| "grad_norm": 2.0246518816727224, | |
| "kl": 1.3505859375, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.0314, | |
| "reward": 1.15625, | |
| "reward_std": 0.20683756470680237, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 232 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 301.09375, | |
| "epoch": 0.0416, | |
| "grad_norm": 3.4777576883520958, | |
| "kl": 1.7216796875, | |
| "learning_rate": 2.4639738083061073e-07, | |
| "loss": 0.0796, | |
| "reward": 1.140625, | |
| "reward_std": 0.28911860287189484, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 234 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 323.640625, | |
| "epoch": 0.04195555555555556, | |
| "grad_norm": 2.269233789387794, | |
| "kl": 1.789794921875, | |
| "learning_rate": 2.4279550982658345e-07, | |
| "loss": 0.0343, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 236 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 285.75, | |
| "epoch": 0.04231111111111111, | |
| "grad_norm": 3.232143067478558, | |
| "kl": 1.150146484375, | |
| "learning_rate": 2.3919513499790646e-07, | |
| "loss": 0.0351, | |
| "reward": 1.109375, | |
| "reward_std": 0.24733919650316238, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 238 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 301.1875, | |
| "epoch": 0.042666666666666665, | |
| "grad_norm": 3.9490792042976843, | |
| "kl": 1.3486328125, | |
| "learning_rate": 2.3559700404385394e-07, | |
| "loss": 0.0803, | |
| "reward": 1.140625, | |
| "reward_std": 0.2909187823534012, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 240 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 290.25, | |
| "epoch": 0.043022222222222226, | |
| "grad_norm": 3.0823221892184796, | |
| "kl": 2.60009765625, | |
| "learning_rate": 2.3200186419770823e-07, | |
| "loss": 0.019, | |
| "reward": 1.125, | |
| "reward_std": 0.2596687823534012, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 242 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 330.953125, | |
| "epoch": 0.04337777777777778, | |
| "grad_norm": 8.855287463518104, | |
| "kl": 4.67138671875, | |
| "learning_rate": 2.284104620715807e-07, | |
| "loss": -0.002, | |
| "reward": 1.03125, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 244 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 289.71875, | |
| "epoch": 0.04373333333333333, | |
| "grad_norm": 9.922124025430394, | |
| "kl": 7.11962890625, | |
| "learning_rate": 2.2482354350136043e-07, | |
| "loss": 0.0296, | |
| "reward": 1.046875, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 246 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 347.328125, | |
| "epoch": 0.044088888888888886, | |
| "grad_norm": 4.9145352811838805, | |
| "kl": 2.637939453125, | |
| "learning_rate": 2.2124185339182496e-07, | |
| "loss": 0.1017, | |
| "reward": 1.140625, | |
| "reward_std": 0.28911860287189484, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 248 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 308.96875, | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 2.1751878201806485, | |
| "kl": 3.7567138671875, | |
| "learning_rate": 2.1766613556194344e-07, | |
| "loss": 0.0106, | |
| "reward": 1.0625, | |
| "reward_std": 0.1875, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 250 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 304.1875, | |
| "epoch": 0.0448, | |
| "grad_norm": 0.5379516641958847, | |
| "kl": 1.821044921875, | |
| "learning_rate": 2.1409713259040628e-07, | |
| "loss": 0.0195, | |
| "reward": 1.0625, | |
| "reward_std": 0.0625, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 252 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 309.921875, | |
| "epoch": 0.04515555555555555, | |
| "grad_norm": 1.5422659826295908, | |
| "kl": 2.447998046875, | |
| "learning_rate": 2.105355856614115e-07, | |
| "loss": 0.0521, | |
| "reward": 1.1875, | |
| "reward_std": 0.25, | |
| "rewards/equation_reward_func": 0.1875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 254 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 312.9375, | |
| "epoch": 0.04551111111111111, | |
| "grad_norm": 1.881389397761525, | |
| "kl": 4.099609375, | |
| "learning_rate": 2.069822344107413e-07, | |
| "loss": -0.0096, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 256 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 339.296875, | |
| "epoch": 0.04586666666666667, | |
| "grad_norm": 2.7062053922429947, | |
| "kl": 2.461669921875, | |
| "learning_rate": 2.034378167721599e-07, | |
| "loss": 0.0942, | |
| "reward": 1.140625, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 258 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 309.84375, | |
| "epoch": 0.04622222222222222, | |
| "grad_norm": 3.420752165914464, | |
| "kl": 1.4273681640625, | |
| "learning_rate": 1.9990306882416485e-07, | |
| "loss": 0.0293, | |
| "reward": 1.109375, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 260 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 330.09375, | |
| "epoch": 0.04657777777777778, | |
| "grad_norm": 1.4438609155615563, | |
| "kl": 3.030517578125, | |
| "learning_rate": 1.9637872463712362e-07, | |
| "loss": 0.0386, | |
| "reward": 1.078125, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 262 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 293.234375, | |
| "epoch": 0.046933333333333334, | |
| "grad_norm": 3.756782187930511, | |
| "kl": 1.635498046875, | |
| "learning_rate": 1.9286551612082773e-07, | |
| "loss": 0.0679, | |
| "reward": 1.15625, | |
| "reward_std": 0.32036860287189484, | |
| "rewards/equation_reward_func": 0.1875, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 264 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 355.5625, | |
| "epoch": 0.04728888888888889, | |
| "grad_norm": 5.013875423554075, | |
| "kl": 2.47119140625, | |
| "learning_rate": 1.8936417287249446e-07, | |
| "loss": 0.0631, | |
| "reward": 1.0625, | |
| "reward_std": 0.27461542934179306, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 266 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 310.15625, | |
| "epoch": 0.04764444444444445, | |
| "grad_norm": 0.5561076088635254, | |
| "kl": 1.971435546875, | |
| "learning_rate": 1.8587542202524985e-07, | |
| "loss": 0.0363, | |
| "reward": 1.0625, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 268 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 337.90625, | |
| "epoch": 0.048, | |
| "grad_norm": 2.53746279303349, | |
| "kl": 4.04541015625, | |
| "learning_rate": 1.82399988097123e-07, | |
| "loss": 0.0317, | |
| "reward": 1.109375, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 270 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 333.828125, | |
| "epoch": 0.048355555555555554, | |
| "grad_norm": 2.2242108525324134, | |
| "kl": 3.0537109375, | |
| "learning_rate": 1.7893859284058378e-07, | |
| "loss": 0.0213, | |
| "reward": 1.0, | |
| "reward_std": 0.17603103816509247, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 272 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 303.53125, | |
| "epoch": 0.04871111111111111, | |
| "grad_norm": 2.6756650642649373, | |
| "kl": 2.84130859375, | |
| "learning_rate": 1.7549195509265407e-07, | |
| "loss": 0.0452, | |
| "reward": 1.109375, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 274 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 282.046875, | |
| "epoch": 0.04906666666666667, | |
| "grad_norm": 73.95606383267277, | |
| "kl": 28.233154296875, | |
| "learning_rate": 1.7206079062562536e-07, | |
| "loss": 0.0546, | |
| "reward": 1.109375, | |
| "reward_std": 0.15625, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 276 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 325.859375, | |
| "epoch": 0.04942222222222222, | |
| "grad_norm": 1.7788841527513326, | |
| "kl": 29.791015625, | |
| "learning_rate": 1.6864581199841226e-07, | |
| "loss": 0.0807, | |
| "reward": 1.09375, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 278 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 298.21875, | |
| "epoch": 0.049777777777777775, | |
| "grad_norm": 1.5764903296991766, | |
| "kl": 3.01513671875, | |
| "learning_rate": 1.6524772840857388e-07, | |
| "loss": 0.0145, | |
| "reward": 1.140625, | |
| "reward_std": 0.15625, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 280 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 309.125, | |
| "epoch": 0.050133333333333335, | |
| "grad_norm": 7.596181720979034, | |
| "kl": 4.2314453125, | |
| "learning_rate": 1.6186724554503237e-07, | |
| "loss": -0.0037, | |
| "reward": 1.140625, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 282 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 290.640625, | |
| "epoch": 0.05048888888888889, | |
| "grad_norm": 2.848445546750997, | |
| "kl": 3.4404296875, | |
| "learning_rate": 1.5850506544152103e-07, | |
| "loss": 0.0475, | |
| "reward": 1.125, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 284 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 294.75, | |
| "epoch": 0.05084444444444444, | |
| "grad_norm": 1.415636309836905, | |
| "kl": 1.5673828125, | |
| "learning_rate": 1.5516188633079107e-07, | |
| "loss": 0.0296, | |
| "reward": 1.046875, | |
| "reward_std": 0.06733439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 286 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 335.703125, | |
| "epoch": 0.0512, | |
| "grad_norm": 1.949097702749259, | |
| "kl": 2.3369140625, | |
| "learning_rate": 1.5183840249960784e-07, | |
| "loss": 0.0612, | |
| "reward": 1.09375, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 288 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 280.515625, | |
| "epoch": 0.051555555555555556, | |
| "grad_norm": 2.8460028252876772, | |
| "kl": 1.85888671875, | |
| "learning_rate": 1.4853530414456612e-07, | |
| "loss": 0.0418, | |
| "reward": 1.125, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 290 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 271.75, | |
| "epoch": 0.05191111111111111, | |
| "grad_norm": 1.6592349286689605, | |
| "kl": 2.0166015625, | |
| "learning_rate": 1.4525327722875568e-07, | |
| "loss": 0.0283, | |
| "reward": 1.09375, | |
| "reward_std": 0.13466878235340118, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 292 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 254.75, | |
| "epoch": 0.05226666666666667, | |
| "grad_norm": 2.3645749075795472, | |
| "kl": 1.789794921875, | |
| "learning_rate": 1.4199300333930515e-07, | |
| "loss": 0.0835, | |
| "reward": 1.1875, | |
| "reward_std": 0.3221687823534012, | |
| "rewards/equation_reward_func": 0.203125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 294 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 319.671875, | |
| "epoch": 0.05262222222222222, | |
| "grad_norm": 4.343843945416777, | |
| "kl": 2.50732421875, | |
| "learning_rate": 1.3875515954583523e-07, | |
| "loss": 0.0624, | |
| "reward": 1.109375, | |
| "reward_std": 0.2645031735301018, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 296 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 306.578125, | |
| "epoch": 0.052977777777777776, | |
| "grad_norm": 2.377005295350322, | |
| "kl": 1.9296875, | |
| "learning_rate": 1.3554041825985e-07, | |
| "loss": 0.0352, | |
| "reward": 1.09375, | |
| "reward_std": 0.2992308586835861, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 298 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 321.609375, | |
| "epoch": 0.05333333333333334, | |
| "grad_norm": 2.611273886267531, | |
| "kl": 2.335693359375, | |
| "learning_rate": 1.323494470950949e-07, | |
| "loss": 0.0688, | |
| "reward": 1.0625, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 300 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 298.859375, | |
| "epoch": 0.05368888888888889, | |
| "grad_norm": 2.213608739672042, | |
| "kl": 2.317138671875, | |
| "learning_rate": 1.2918290872891236e-07, | |
| "loss": 0.045, | |
| "reward": 1.046875, | |
| "reward_std": 0.15625, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 302 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 319.390625, | |
| "epoch": 0.054044444444444444, | |
| "grad_norm": 1.1361079950653166, | |
| "kl": 2.716796875, | |
| "learning_rate": 1.260414607646213e-07, | |
| "loss": 0.0547, | |
| "reward": 1.015625, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 304 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 315.734375, | |
| "epoch": 0.0544, | |
| "grad_norm": 2.694956280462497, | |
| "kl": 3.01123046875, | |
| "learning_rate": 1.2292575559495143e-07, | |
| "loss": 0.0712, | |
| "reward": 1.078125, | |
| "reward_std": 0.21694982051849365, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 306 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 314.15625, | |
| "epoch": 0.05475555555555556, | |
| "grad_norm": 1.6293408130888805, | |
| "kl": 2.3837890625, | |
| "learning_rate": 1.1983644026655835e-07, | |
| "loss": 0.0147, | |
| "reward": 1.125, | |
| "reward_std": 0.13466878235340118, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 308 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 273.125, | |
| "epoch": 0.05511111111111111, | |
| "grad_norm": 2.1302515062721126, | |
| "kl": 2.05712890625, | |
| "learning_rate": 1.1677415634565066e-07, | |
| "loss": 0.0406, | |
| "reward": 1.15625, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 310 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 337.59375, | |
| "epoch": 0.055466666666666664, | |
| "grad_norm": 15.014740699660255, | |
| "kl": 7.84521484375, | |
| "learning_rate": 1.1373953978475353e-07, | |
| "loss": 0.0368, | |
| "reward": 1.0625, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 312 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 312.1875, | |
| "epoch": 0.055822222222222224, | |
| "grad_norm": 2.0788629281112727, | |
| "kl": 4.335693359375, | |
| "learning_rate": 1.1073322079063913e-07, | |
| "loss": 0.0683, | |
| "reward": 1.15625, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 314 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 276.34375, | |
| "epoch": 0.05617777777777778, | |
| "grad_norm": 11.038251821932738, | |
| "kl": 4.507568359375, | |
| "learning_rate": 1.0775582369344946e-07, | |
| "loss": 0.0259, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 316 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 285.59375, | |
| "epoch": 0.05653333333333333, | |
| "grad_norm": 4.009960857728169, | |
| "kl": 2.4130859375, | |
| "learning_rate": 1.0480796681704077e-07, | |
| "loss": -0.0194, | |
| "reward": 1.125, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 318 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 277.296875, | |
| "epoch": 0.05688888888888889, | |
| "grad_norm": 4.499177733550329, | |
| "kl": 5.369873046875, | |
| "learning_rate": 1.018902623505741e-07, | |
| "loss": 0.0504, | |
| "reward": 1.078125, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 320 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 316.578125, | |
| "epoch": 0.057244444444444445, | |
| "grad_norm": 3.3047156898766272, | |
| "kl": 3.43701171875, | |
| "learning_rate": 9.900331622138063e-08, | |
| "loss": 0.0595, | |
| "reward": 1.078125, | |
| "reward_std": 0.16591878235340118, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 322 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 297.765625, | |
| "epoch": 0.0576, | |
| "grad_norm": 4.938077502561252, | |
| "kl": 2.2734375, | |
| "learning_rate": 9.614772796912681e-08, | |
| "loss": 0.02, | |
| "reward": 1.15625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 324 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 337.640625, | |
| "epoch": 0.05795555555555556, | |
| "grad_norm": 3.2448715644280974, | |
| "kl": 2.689453125, | |
| "learning_rate": 9.332409062130686e-08, | |
| "loss": 0.0284, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 326 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 297.21875, | |
| "epoch": 0.05831111111111111, | |
| "grad_norm": 3.6089469790675452, | |
| "kl": 1.907470703125, | |
| "learning_rate": 9.053299057008699e-08, | |
| "loss": 0.0125, | |
| "reward": 0.96875, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.015625, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 328 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 298.953125, | |
| "epoch": 0.058666666666666666, | |
| "grad_norm": 3.9902002814055497, | |
| "kl": 2.693603515625, | |
| "learning_rate": 8.777500745052743e-08, | |
| "loss": 0.054, | |
| "reward": 1.109375, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 330 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 310.515625, | |
| "epoch": 0.05902222222222222, | |
| "grad_norm": 2.3258942923733033, | |
| "kl": 1.953857421875, | |
| "learning_rate": 8.505071402020892e-08, | |
| "loss": 0.0404, | |
| "reward": 1.078125, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 332 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 306.328125, | |
| "epoch": 0.05937777777777778, | |
| "grad_norm": 2.105282555937082, | |
| "kl": 2.130615234375, | |
| "learning_rate": 8.236067604028562e-08, | |
| "loss": 0.0711, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 334 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 333.03125, | |
| "epoch": 0.05973333333333333, | |
| "grad_norm": 1.6921617524371768, | |
| "kl": 2.76611328125, | |
| "learning_rate": 7.970545215799327e-08, | |
| "loss": -0.0171, | |
| "reward": 1.09375, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 336 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 295.71875, | |
| "epoch": 0.060088888888888886, | |
| "grad_norm": 2.2366407518555658, | |
| "kl": 2.571533203125, | |
| "learning_rate": 7.708559379063204e-08, | |
| "loss": 0.0171, | |
| "reward": 1.078125, | |
| "reward_std": 0.15625, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 338 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 296.125, | |
| "epoch": 0.060444444444444446, | |
| "grad_norm": 1.7453930694713657, | |
| "kl": 1.99072265625, | |
| "learning_rate": 7.45016450110534e-08, | |
| "loss": 0.086, | |
| "reward": 1.078125, | |
| "reward_std": 0.2548343911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 340 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 287.390625, | |
| "epoch": 0.0608, | |
| "grad_norm": 2.7894086215156477, | |
| "kl": 3.01318359375, | |
| "learning_rate": 7.195414243467029e-08, | |
| "loss": 0.0482, | |
| "reward": 1.171875, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 342 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 317.40625, | |
| "epoch": 0.06115555555555555, | |
| "grad_norm": 0.931178773836098, | |
| "kl": 1.2216796875, | |
| "learning_rate": 6.944361510801763e-08, | |
| "loss": -0.0173, | |
| "reward": 1.015625, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 344 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 299.359375, | |
| "epoch": 0.061511111111111114, | |
| "grad_norm": 3.753569436457584, | |
| "kl": 2.353759765625, | |
| "learning_rate": 6.697058439888283e-08, | |
| "loss": 0.0152, | |
| "reward": 1.078125, | |
| "reward_std": 0.10341878235340118, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 346 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 260.09375, | |
| "epoch": 0.06186666666666667, | |
| "grad_norm": 2.9761032955845192, | |
| "kl": 1.854736328125, | |
| "learning_rate": 6.453556388803288e-08, | |
| "loss": 0.025, | |
| "reward": 1.078125, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 348 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 306.0625, | |
| "epoch": 0.06222222222222222, | |
| "grad_norm": 2.583072279031626, | |
| "kl": 1.90185546875, | |
| "learning_rate": 6.213905926255697e-08, | |
| "loss": -0.0007, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 350 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 284.453125, | |
| "epoch": 0.06257777777777777, | |
| "grad_norm": 2.5644083370227495, | |
| "kl": 10.295654296875, | |
| "learning_rate": 5.978156821084987e-08, | |
| "loss": 0.011, | |
| "reward": 1.0625, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 352 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 304.84375, | |
| "epoch": 0.06293333333333333, | |
| "grad_norm": 3.260671051240977, | |
| "kl": 7.114013671875, | |
| "learning_rate": 5.7463580319254853e-08, | |
| "loss": 0.078, | |
| "reward": 1.09375, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 354 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 281.296875, | |
| "epoch": 0.0632888888888889, | |
| "grad_norm": 2.0339670428860477, | |
| "kl": 2.173095703125, | |
| "learning_rate": 5.518557697039081e-08, | |
| "loss": 0.0621, | |
| "reward": 1.140625, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 356 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 294.5625, | |
| "epoch": 0.06364444444444445, | |
| "grad_norm": 3.3851967041509154, | |
| "kl": 2.346923828125, | |
| "learning_rate": 5.294803124318145e-08, | |
| "loss": -0.008, | |
| "reward": 1.046875, | |
| "reward_std": 0.06733439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 358 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 319.390625, | |
| "epoch": 0.064, | |
| "grad_norm": 1.6809943974752608, | |
| "kl": 2.185791015625, | |
| "learning_rate": 5.07514078146106e-08, | |
| "loss": 0.0332, | |
| "reward": 1.109375, | |
| "reward_std": 0.19053421169519424, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 360 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 261.21875, | |
| "epoch": 0.06435555555555555, | |
| "grad_norm": 1.105930554758287, | |
| "kl": 1.884521484375, | |
| "learning_rate": 4.859616286322094e-08, | |
| "loss": 0.0335, | |
| "reward": 1.0625, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 362 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 293.71875, | |
| "epoch": 0.06471111111111111, | |
| "grad_norm": 2.075449397863208, | |
| "kl": 2.149169921875, | |
| "learning_rate": 4.648274397437829e-08, | |
| "loss": -0.0078, | |
| "reward": 1.109375, | |
| "reward_std": 0.15625, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 364 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 299.0625, | |
| "epoch": 0.06506666666666666, | |
| "grad_norm": 2.440666439161297, | |
| "kl": 2.136474609375, | |
| "learning_rate": 4.4411590047320617e-08, | |
| "loss": 0.0491, | |
| "reward": 1.140625, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 366 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 320.890625, | |
| "epoch": 0.06542222222222223, | |
| "grad_norm": 4.619211137682817, | |
| "kl": 1.549072265625, | |
| "learning_rate": 4.2383131204010494e-08, | |
| "loss": 0.0484, | |
| "reward": 1.078125, | |
| "reward_std": 0.26978103816509247, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 368 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 288.234375, | |
| "epoch": 0.06577777777777778, | |
| "grad_norm": 2.1383834468441396, | |
| "kl": 1.26123046875, | |
| "learning_rate": 4.039778869981064e-08, | |
| "loss": 0.017, | |
| "reward": 1.078125, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 370 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 316.71875, | |
| "epoch": 0.06613333333333334, | |
| "grad_norm": 2.9751121792977497, | |
| "kl": 1.9384765625, | |
| "learning_rate": 3.845597483600049e-08, | |
| "loss": 0.0462, | |
| "reward": 1.0625, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 372 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 302.296875, | |
| "epoch": 0.06648888888888889, | |
| "grad_norm": 1.7788468837628437, | |
| "kl": 1.295654296875, | |
| "learning_rate": 3.655809287415284e-08, | |
| "loss": -0.0181, | |
| "reward": 1.09375, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 374 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 282.734375, | |
| "epoch": 0.06684444444444444, | |
| "grad_norm": 1.7936851877118818, | |
| "kl": 1.361083984375, | |
| "learning_rate": 3.4704536952387285e-08, | |
| "loss": 0.0238, | |
| "reward": 1.078125, | |
| "reward_std": 0.16591878235340118, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 376 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 310.375, | |
| "epoch": 0.0672, | |
| "grad_norm": 4.685423790152177, | |
| "kl": 3.947998046875, | |
| "learning_rate": 3.2895692003518575e-08, | |
| "loss": 0.0234, | |
| "reward": 1.03125, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 378 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 298.34375, | |
| "epoch": 0.06755555555555555, | |
| "grad_norm": 3.1205307234975543, | |
| "kl": 1.413330078125, | |
| "learning_rate": 3.113193367511635e-08, | |
| "loss": 0.0445, | |
| "reward": 1.046875, | |
| "reward_std": 0.21875, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 380 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 319.703125, | |
| "epoch": 0.06791111111111112, | |
| "grad_norm": 1.1720140167792736, | |
| "kl": 1.87060546875, | |
| "learning_rate": 2.9413628251493934e-08, | |
| "loss": 0.0189, | |
| "reward": 1.03125, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 382 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 310.0, | |
| "epoch": 0.06826666666666667, | |
| "grad_norm": 2.6084931637781725, | |
| "kl": 3.1181640625, | |
| "learning_rate": 2.774113257764066e-08, | |
| "loss": 0.0243, | |
| "reward": 1.046875, | |
| "reward_std": 0.06733439117670059, | |
| "rewards/equation_reward_func": 0.046875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 384 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 313.3125, | |
| "epoch": 0.06862222222222222, | |
| "grad_norm": 4.358043629813717, | |
| "kl": 2.083251953125, | |
| "learning_rate": 2.611479398511518e-08, | |
| "loss": 0.0497, | |
| "reward": 1.15625, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.171875, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 386 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 311.5625, | |
| "epoch": 0.06897777777777778, | |
| "grad_norm": 1.7314295380277487, | |
| "kl": 2.2861328125, | |
| "learning_rate": 2.4534950219914057e-08, | |
| "loss": 0.0184, | |
| "reward": 1.03125, | |
| "reward_std": 0.0625, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 388 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 315.078125, | |
| "epoch": 0.06933333333333333, | |
| "grad_norm": 3.085044311379691, | |
| "kl": 1.55078125, | |
| "learning_rate": 2.300192937233128e-08, | |
| "loss": 0.0237, | |
| "reward": 1.125, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 390 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 321.1875, | |
| "epoch": 0.06968888888888888, | |
| "grad_norm": 2.1144825124318696, | |
| "kl": 1.6259765625, | |
| "learning_rate": 2.1516049808822935e-08, | |
| "loss": 0.0228, | |
| "reward": 1.078125, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 392 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 296.390625, | |
| "epoch": 0.07004444444444445, | |
| "grad_norm": 2.2495222116539715, | |
| "kl": 1.490234375, | |
| "learning_rate": 2.007762010589098e-08, | |
| "loss": 0.0232, | |
| "reward": 1.078125, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 394 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 286.09375, | |
| "epoch": 0.0704, | |
| "grad_norm": 3.5477202571420716, | |
| "kl": 3.229248046875, | |
| "learning_rate": 1.8686938986000627e-08, | |
| "loss": -0.0136, | |
| "reward": 1.109375, | |
| "reward_std": 0.20200317353010178, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 396 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 292.890625, | |
| "epoch": 0.07075555555555556, | |
| "grad_norm": 3.401342482083026, | |
| "kl": 1.608642578125, | |
| "learning_rate": 1.734429525554365e-08, | |
| "loss": -0.0173, | |
| "reward": 1.015625, | |
| "reward_std": 0.08228103816509247, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 398 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 290.140625, | |
| "epoch": 0.07111111111111111, | |
| "grad_norm": 2.4346116264171203, | |
| "kl": 2.428955078125, | |
| "learning_rate": 1.604996774486145e-08, | |
| "loss": 0.0619, | |
| "reward": 1.046875, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 400 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 327.09375, | |
| "epoch": 0.07146666666666666, | |
| "grad_norm": 1.4096767097340954, | |
| "kl": 2.732421875, | |
| "learning_rate": 1.4804225250339281e-08, | |
| "loss": 0.0063, | |
| "reward": 1.09375, | |
| "reward_std": 0.09858439117670059, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 402 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 308.53125, | |
| "epoch": 0.07182222222222222, | |
| "grad_norm": 11.032142152384289, | |
| "kl": 3.81396484375, | |
| "learning_rate": 1.360732647858498e-08, | |
| "loss": 0.0763, | |
| "reward": 1.203125, | |
| "reward_std": 0.2645031735301018, | |
| "rewards/equation_reward_func": 0.203125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 404 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 308.3125, | |
| "epoch": 0.07217777777777777, | |
| "grad_norm": 2.3580842283452412, | |
| "kl": 2.40478515625, | |
| "learning_rate": 1.2459519992702311e-08, | |
| "loss": 0.0281, | |
| "reward": 1.015625, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 406 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 276.265625, | |
| "epoch": 0.07253333333333334, | |
| "grad_norm": 1.807591388336414, | |
| "kl": 2.02685546875, | |
| "learning_rate": 1.1361044160671629e-08, | |
| "loss": 0.0073, | |
| "reward": 1.15625, | |
| "reward_std": 0.19716878235340118, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 408 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 305.59375, | |
| "epoch": 0.07288888888888889, | |
| "grad_norm": 397.3498098291439, | |
| "kl": 50.008544921875, | |
| "learning_rate": 1.0312127105846947e-08, | |
| "loss": 0.1459, | |
| "reward": 1.03125, | |
| "reward_std": 0.23853103816509247, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 410 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 307.015625, | |
| "epoch": 0.07324444444444445, | |
| "grad_norm": 2.197104654614589, | |
| "kl": 2.2783203125, | |
| "learning_rate": 9.312986659581301e-09, | |
| "loss": 0.0373, | |
| "reward": 1.09375, | |
| "reward_std": 0.1610843911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 412 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 307.390625, | |
| "epoch": 0.0736, | |
| "grad_norm": 2.2032918507669015, | |
| "kl": 1.651611328125, | |
| "learning_rate": 8.363830315988945e-09, | |
| "loss": 0.0537, | |
| "reward": 1.078125, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 414 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 296.71875, | |
| "epoch": 0.07395555555555555, | |
| "grad_norm": 2.6002311285352175, | |
| "kl": 2.895751953125, | |
| "learning_rate": 7.46485518885462e-09, | |
| "loss": 0.0162, | |
| "reward": 1.09375, | |
| "reward_std": 0.125, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 416 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 314.734375, | |
| "epoch": 0.0743111111111111, | |
| "grad_norm": 1.4366325174606742, | |
| "kl": 1.68115234375, | |
| "learning_rate": 6.616247970698319e-09, | |
| "loss": 0.0527, | |
| "reward": 1.0625, | |
| "reward_std": 0.22575797885656357, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 0.953125, | |
| "step": 418 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 274.125, | |
| "epoch": 0.07466666666666667, | |
| "grad_norm": 2.832174519961065, | |
| "kl": 1.653076171875, | |
| "learning_rate": 5.8181848940044855e-09, | |
| "loss": 0.0479, | |
| "reward": 1.15625, | |
| "reward_std": 0.25, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 420 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 293.578125, | |
| "epoch": 0.07502222222222223, | |
| "grad_norm": 3.2917629073928247, | |
| "kl": 1.63916015625, | |
| "learning_rate": 5.070831694623135e-09, | |
| "loss": 0.0314, | |
| "reward": 1.1875, | |
| "reward_std": 0.3221687823534012, | |
| "rewards/equation_reward_func": 0.1875, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 422 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 268.71875, | |
| "epoch": 0.07537777777777778, | |
| "grad_norm": 1.3697523023561404, | |
| "kl": 2.22021484375, | |
| "learning_rate": 4.374343577351336e-09, | |
| "loss": 0.0136, | |
| "reward": 1.046875, | |
| "reward_std": 0.1298343911767006, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 424 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 312.59375, | |
| "epoch": 0.07573333333333333, | |
| "grad_norm": 2.896324603562049, | |
| "kl": 1.552734375, | |
| "learning_rate": 3.7288651837012745e-09, | |
| "loss": 0.0834, | |
| "reward": 1.140625, | |
| "reward_std": 0.2909187823534012, | |
| "rewards/equation_reward_func": 0.15625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 426 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 279.921875, | |
| "epoch": 0.07608888888888889, | |
| "grad_norm": 2.2563282918149166, | |
| "kl": 2.084228515625, | |
| "learning_rate": 3.134530561862081e-09, | |
| "loss": 0.0184, | |
| "reward": 1.109375, | |
| "reward_std": 0.10341878235340118, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 428 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 299.8125, | |
| "epoch": 0.07644444444444444, | |
| "grad_norm": 2.805702161500462, | |
| "kl": 1.2802734375, | |
| "learning_rate": 2.5914631388619103e-09, | |
| "loss": 0.022, | |
| "reward": 1.0625, | |
| "reward_std": 0.0625, | |
| "rewards/equation_reward_func": 0.0625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 430 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 280.1875, | |
| "epoch": 0.0768, | |
| "grad_norm": 2.3897542479107328, | |
| "kl": 1.429931640625, | |
| "learning_rate": 2.0997756949353297e-09, | |
| "loss": 0.0494, | |
| "reward": 1.109375, | |
| "reward_std": 0.24336542934179306, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 432 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 309.828125, | |
| "epoch": 0.07715555555555556, | |
| "grad_norm": 3.547940603348454, | |
| "kl": 1.690185546875, | |
| "learning_rate": 1.6595703401020844e-09, | |
| "loss": 0.0557, | |
| "reward": 1.109375, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.109375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 434 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 295.828125, | |
| "epoch": 0.07751111111111111, | |
| "grad_norm": 109.41721675323444, | |
| "kl": 28.513916015625, | |
| "learning_rate": 1.2709384929615596e-09, | |
| "loss": 0.0721, | |
| "reward": 1.09375, | |
| "reward_std": 0.13466878235340118, | |
| "rewards/equation_reward_func": 0.09375, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 436 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 280.578125, | |
| "epoch": 0.07786666666666667, | |
| "grad_norm": 1.9228717734559753, | |
| "kl": 2.058837890625, | |
| "learning_rate": 9.339608617077165e-10, | |
| "loss": -0.034, | |
| "reward": 1.015625, | |
| "reward_std": 0.09375, | |
| "rewards/equation_reward_func": 0.03125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 438 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 292.5, | |
| "epoch": 0.07822222222222222, | |
| "grad_norm": 3.2983736045582552, | |
| "kl": 1.70654296875, | |
| "learning_rate": 6.487074273681114e-10, | |
| "loss": 0.0022, | |
| "reward": 1.140625, | |
| "reward_std": 0.22841878235340118, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 440 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 280.296875, | |
| "epoch": 0.07857777777777777, | |
| "grad_norm": 2.7643778642220154, | |
| "kl": 1.887939453125, | |
| "learning_rate": 4.152374292708538e-10, | |
| "loss": 0.0083, | |
| "reward": 1.140625, | |
| "reward_std": 0.1923343911767006, | |
| "rewards/equation_reward_func": 0.140625, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 442 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 287.140625, | |
| "epoch": 0.07893333333333333, | |
| "grad_norm": 2.995996483175849, | |
| "kl": 1.871337890625, | |
| "learning_rate": 2.3359935274214204e-10, | |
| "loss": 0.0448, | |
| "reward": 1.265625, | |
| "reward_std": 0.33667195588350296, | |
| "rewards/equation_reward_func": 0.28125, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 444 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 307.890625, | |
| "epoch": 0.0792888888888889, | |
| "grad_norm": 2.600208741694735, | |
| "kl": 2.14111328125, | |
| "learning_rate": 1.0383091903720665e-10, | |
| "loss": 0.0051, | |
| "reward": 1.0, | |
| "reward_std": 0.0625, | |
| "rewards/equation_reward_func": 0.015625, | |
| "rewards/format_reward_func": 0.984375, | |
| "step": 446 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 283.359375, | |
| "epoch": 0.07964444444444445, | |
| "grad_norm": 2.5889730669425286, | |
| "kl": 377.768310546875, | |
| "learning_rate": 2.595907750671533e-11, | |
| "loss": 0.4151, | |
| "reward": 1.046875, | |
| "reward_std": 0.20728103816509247, | |
| "rewards/equation_reward_func": 0.078125, | |
| "rewards/format_reward_func": 0.96875, | |
| "step": 448 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 283.171875, | |
| "epoch": 0.08, | |
| "grad_norm": 2.8580107740928953, | |
| "kl": 2.1181640625, | |
| "learning_rate": 0.0, | |
| "loss": 0.0826, | |
| "reward": 1.125, | |
| "reward_std": 0.2235843911767006, | |
| "rewards/equation_reward_func": 0.125, | |
| "rewards/format_reward_func": 1.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "step": 450, | |
| "total_flos": 0.0, | |
| "train_loss": 0.04409932630071934, | |
| "train_runtime": 12236.3143, | |
| "train_samples_per_second": 1.177, | |
| "train_steps_per_second": 0.037 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "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": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |