| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9888268156424581, |
| "eval_steps": 500, |
| "global_step": 59, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 12902.000651041666, |
| "epoch": 0.01675977653631285, |
| "grad_norm": 0.21624265611171722, |
| "kl": 0.0, |
| "learning_rate": 5e-07, |
| "loss": 0.0, |
| "reward": 0.4761905123790105, |
| "reward_std": 0.43015046914418537, |
| "rewards/accuracy_reward": 0.4761905123790105, |
| "step": 1 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10812.238606770834, |
| "epoch": 0.0335195530726257, |
| "grad_norm": 0.2267991006374359, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "reward": 0.5714285870393118, |
| "reward_std": 0.48795002698898315, |
| "rewards/accuracy_reward": 0.5714285870393118, |
| "step": 2 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 5980.00048828125, |
| "epoch": 0.05027932960893855, |
| "grad_norm": 0.1943487972021103, |
| "kl": 7.70886739095052e-05, |
| "learning_rate": 1.5e-06, |
| "loss": 0.0, |
| "reward": 0.9047619501749674, |
| "reward_std": 0.2519763112068176, |
| "rewards/accuracy_reward": 0.9047619501749674, |
| "step": 3 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10902.095377604166, |
| "epoch": 0.0670391061452514, |
| "grad_norm": 0.16429206728935242, |
| "kl": 0.0001037915547688802, |
| "learning_rate": 2e-06, |
| "loss": 0.0, |
| "reward": 0.6666666865348816, |
| "reward_std": 0.35634831587473553, |
| "rewards/accuracy_reward": 0.6666666865348816, |
| "step": 4 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 13527.381510416666, |
| "epoch": 0.08379888268156424, |
| "grad_norm": 0.17159564793109894, |
| "kl": 8.392333984375e-05, |
| "learning_rate": 2.5e-06, |
| "loss": 0.0, |
| "reward": 0.3809524079163869, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.3809524079163869, |
| "step": 5 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10472.143391927084, |
| "epoch": 0.1005586592178771, |
| "grad_norm": 0.0006169096450321376, |
| "kl": 8.376439412434895e-05, |
| "learning_rate": 3e-06, |
| "loss": 0.0, |
| "reward": 0.6666666666666666, |
| "reward_std": 0.0, |
| "rewards/accuracy_reward": 0.6666666666666666, |
| "step": 6 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8165.190836588542, |
| "epoch": 0.11731843575418995, |
| "grad_norm": 0.16822050511837006, |
| "kl": 5.841255187988281e-05, |
| "learning_rate": 2.9973655976464826e-06, |
| "loss": 0.0, |
| "reward": 0.7142857313156128, |
| "reward_std": 0.3408241669336955, |
| "rewards/accuracy_reward": 0.7142857313156128, |
| "step": 7 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7482.38134765625, |
| "epoch": 0.1340782122905028, |
| "grad_norm": 0.18172389268875122, |
| "kl": 7.677078247070312e-05, |
| "learning_rate": 2.989471644020276e-06, |
| "loss": 0.0, |
| "reward": 0.8571428855260214, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.8571428855260214, |
| "step": 8 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 11130.000325520834, |
| "epoch": 0.15083798882681565, |
| "grad_norm": 0.17098551988601685, |
| "kl": 0.0001338322957356771, |
| "learning_rate": 2.976345866921395e-06, |
| "loss": 0.0, |
| "reward": 0.7142857412497202, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.7142857412497202, |
| "step": 9 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9002.905110677084, |
| "epoch": 0.16759776536312848, |
| "grad_norm": 0.0026507459115236998, |
| "kl": 0.0001624425252278646, |
| "learning_rate": 2.9580343711206164e-06, |
| "loss": 0.0, |
| "reward": 0.6666666666666666, |
| "reward_std": 0.0, |
| "rewards/accuracy_reward": 0.6666666666666666, |
| "step": 10 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9350.048014322916, |
| "epoch": 0.18435754189944134, |
| "grad_norm": 0.1485043615102768, |
| "kl": 0.0001201629638671875, |
| "learning_rate": 2.9346014764147837e-06, |
| "loss": 0.0, |
| "reward": 0.7142857412497202, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.7142857412497202, |
| "step": 11 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9505.04833984375, |
| "epoch": 0.2011173184357542, |
| "grad_norm": 0.24309073388576508, |
| "kl": 0.00019772847493489584, |
| "learning_rate": 2.9061294917011814e-06, |
| "loss": 0.0, |
| "reward": 0.6666666865348816, |
| "reward_std": 0.5034741759300232, |
| "rewards/accuracy_reward": 0.6666666865348816, |
| "step": 12 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10659.857747395834, |
| "epoch": 0.21787709497206703, |
| "grad_norm": 0.20960527658462524, |
| "kl": 0.00028260548909505207, |
| "learning_rate": 2.8727184258645274e-06, |
| "loss": 0.0, |
| "reward": 0.571428601940473, |
| "reward_std": 0.4146263202031453, |
| "rewards/accuracy_reward": 0.571428601940473, |
| "step": 13 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 6585.381184895833, |
| "epoch": 0.2346368715083799, |
| "grad_norm": 0.003225038992241025, |
| "kl": 0.0007921854654947916, |
| "learning_rate": 2.834485636492121e-06, |
| "loss": 0.0, |
| "reward": 1.0, |
| "reward_std": 0.0, |
| "rewards/accuracy_reward": 1.0, |
| "step": 14 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8817.381184895834, |
| "epoch": 0.25139664804469275, |
| "grad_norm": 0.20985934138298035, |
| "kl": 0.00016641616821289062, |
| "learning_rate": 2.7915654176510332e-06, |
| "loss": 0.0, |
| "reward": 0.5714285870393118, |
| "reward_std": 0.3408241669336955, |
| "rewards/accuracy_reward": 0.5714285870393118, |
| "step": 15 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10070.000651041666, |
| "epoch": 0.2681564245810056, |
| "grad_norm": 0.1750493049621582, |
| "kl": 0.0001678466796875, |
| "learning_rate": 2.744108528175303e-06, |
| "loss": 0.0, |
| "reward": 0.523809552192688, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.523809552192688, |
| "step": 16 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10897.857747395834, |
| "epoch": 0.2849162011173184, |
| "grad_norm": 0.15098980069160461, |
| "kl": 0.0002867380777994792, |
| "learning_rate": 2.6922816621200303e-06, |
| "loss": 0.0, |
| "reward": 0.6666666915019354, |
| "reward_std": 0.2519763112068176, |
| "rewards/accuracy_reward": 0.6666666915019354, |
| "step": 17 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10651.61962890625, |
| "epoch": 0.3016759776536313, |
| "grad_norm": 0.12570348381996155, |
| "kl": 0.0016530354817708333, |
| "learning_rate": 2.6362668632424304e-06, |
| "loss": 0.0001, |
| "reward": 0.9047619104385376, |
| "reward_std": 0.16265000899632773, |
| "rewards/accuracy_reward": 0.9047619104385376, |
| "step": 18 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8625.333984375, |
| "epoch": 0.31843575418994413, |
| "grad_norm": 0.11648119986057281, |
| "kl": 0.0003102620442708333, |
| "learning_rate": 2.5762608855664966e-06, |
| "loss": 0.0, |
| "reward": 0.5714285771052042, |
| "reward_std": 0.16265000899632773, |
| "rewards/accuracy_reward": 0.5714285771052042, |
| "step": 19 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 6595.95263671875, |
| "epoch": 0.33519553072625696, |
| "grad_norm": 0.16656093299388885, |
| "kl": 0.0021419525146484375, |
| "learning_rate": 2.5124745022773158e-06, |
| "loss": 0.0001, |
| "reward": 0.9523809750874838, |
| "reward_std": 0.1259881556034088, |
| "rewards/accuracy_reward": 0.9523809750874838, |
| "step": 20 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9921.238606770834, |
| "epoch": 0.35195530726256985, |
| "grad_norm": 0.19331984221935272, |
| "kl": 0.0006939570109049479, |
| "learning_rate": 2.445131765372567e-06, |
| "loss": 0.0, |
| "reward": 0.6190476417541504, |
| "reward_std": 0.5034741759300232, |
| "rewards/accuracy_reward": 0.6190476417541504, |
| "step": 21 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9879.524251302084, |
| "epoch": 0.3687150837988827, |
| "grad_norm": 0.18972796201705933, |
| "kl": 0.0016269683837890625, |
| "learning_rate": 2.374469218671708e-06, |
| "loss": 0.0001, |
| "reward": 0.4761905123790105, |
| "reward_std": 0.43015046914418537, |
| "rewards/accuracy_reward": 0.4761905123790105, |
| "step": 22 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 11783.905110677084, |
| "epoch": 0.3854748603351955, |
| "grad_norm": 0.20772391557693481, |
| "kl": 0.0015246073404947917, |
| "learning_rate": 2.3007350669471863e-06, |
| "loss": 0.0001, |
| "reward": 0.4761905123790105, |
| "reward_std": 0.43015046914418537, |
| "rewards/accuracy_reward": 0.4761905123790105, |
| "step": 23 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7978.476399739583, |
| "epoch": 0.4022346368715084, |
| "grad_norm": 0.19960026443004608, |
| "kl": 0.0011844635009765625, |
| "learning_rate": 2.2241883040961176e-06, |
| "loss": 0.0, |
| "reward": 0.7142857313156128, |
| "reward_std": 0.3408241669336955, |
| "rewards/accuracy_reward": 0.7142857313156128, |
| "step": 24 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7601.95263671875, |
| "epoch": 0.41899441340782123, |
| "grad_norm": 0.2679334580898285, |
| "kl": 0.0011603037516276042, |
| "learning_rate": 2.145097803414781e-06, |
| "loss": 0.0, |
| "reward": 0.7619048158327738, |
| "reward_std": 0.43015046914418537, |
| "rewards/accuracy_reward": 0.7619048158327738, |
| "step": 25 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9032.524251302084, |
| "epoch": 0.43575418994413406, |
| "grad_norm": 0.16398951411247253, |
| "kl": 0.0021718343098958335, |
| "learning_rate": 2.063741373171357e-06, |
| "loss": 0.0001, |
| "reward": 0.9047619501749674, |
| "reward_std": 0.2519763112068176, |
| "rewards/accuracy_reward": 0.9047619501749674, |
| "step": 26 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7896.76220703125, |
| "epoch": 0.45251396648044695, |
| "grad_norm": 0.16601108014583588, |
| "kl": 0.0002962748209635417, |
| "learning_rate": 1.980404780794256e-06, |
| "loss": 0.0, |
| "reward": 0.761904795964559, |
| "reward_std": 0.30416231354077655, |
| "rewards/accuracy_reward": 0.761904795964559, |
| "step": 27 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8936.71484375, |
| "epoch": 0.4692737430167598, |
| "grad_norm": 0.17378006875514984, |
| "kl": 0.0005385080973307291, |
| "learning_rate": 1.8953807491036015e-06, |
| "loss": 0.0, |
| "reward": 0.8571428855260214, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.8571428855260214, |
| "step": 28 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8249.333821614584, |
| "epoch": 0.4860335195530726, |
| "grad_norm": 0.1794224977493286, |
| "kl": 0.0006039937337239584, |
| "learning_rate": 1.8089679281116475e-06, |
| "loss": 0.0, |
| "reward": 0.9047619104385376, |
| "reward_std": 0.16265000899632773, |
| "rewards/accuracy_reward": 0.9047619104385376, |
| "step": 29 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 6461.4765625, |
| "epoch": 0.5027932960893855, |
| "grad_norm": 0.21347256004810333, |
| "kl": 0.0010026295979817708, |
| "learning_rate": 1.7214698460037221e-06, |
| "loss": 0.0, |
| "reward": 0.7142857313156128, |
| "reward_std": 0.3408241669336955, |
| "rewards/accuracy_reward": 0.7142857313156128, |
| "step": 30 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7002.904947916667, |
| "epoch": 0.5195530726256983, |
| "grad_norm": 0.23133651912212372, |
| "kl": 0.00049591064453125, |
| "learning_rate": 1.6331938429844024e-06, |
| "loss": 0.0, |
| "reward": 0.809523860613505, |
| "reward_std": 0.4146263202031453, |
| "rewards/accuracy_reward": 0.809523860613505, |
| "step": 31 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 12326.857747395834, |
| "epoch": 0.5363128491620112, |
| "grad_norm": 0.214329332113266, |
| "kl": 0.00101470947265625, |
| "learning_rate": 1.5444499917338397e-06, |
| "loss": 0.0, |
| "reward": 0.6190476616223654, |
| "reward_std": 0.48233647147814435, |
| "rewards/accuracy_reward": 0.6190476616223654, |
| "step": 32 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8316.57177734375, |
| "epoch": 0.553072625698324, |
| "grad_norm": 0.25241097807884216, |
| "kl": 0.0011425018310546875, |
| "learning_rate": 1.4555500082661604e-06, |
| "loss": 0.0, |
| "reward": 0.809523860613505, |
| "reward_std": 0.4146263202031453, |
| "rewards/accuracy_reward": 0.809523860613505, |
| "step": 33 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8084.76220703125, |
| "epoch": 0.5698324022346368, |
| "grad_norm": 0.21672719717025757, |
| "kl": 0.000568389892578125, |
| "learning_rate": 1.3668061570155979e-06, |
| "loss": 0.0, |
| "reward": 0.8095238407452902, |
| "reward_std": 0.30416231354077655, |
| "rewards/accuracy_reward": 0.8095238407452902, |
| "step": 34 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 11807.095540364584, |
| "epoch": 0.5865921787709497, |
| "grad_norm": 0.22705776989459991, |
| "kl": 0.000720977783203125, |
| "learning_rate": 1.2785301539962784e-06, |
| "loss": 0.0, |
| "reward": 0.3809523930152257, |
| "reward_std": 0.45128817359606427, |
| "rewards/accuracy_reward": 0.3809523930152257, |
| "step": 35 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8191.952799479167, |
| "epoch": 0.6033519553072626, |
| "grad_norm": 0.18633516132831573, |
| "kl": 0.0005601247151692709, |
| "learning_rate": 1.1910320718883526e-06, |
| "loss": 0.0, |
| "reward": 0.8095238208770752, |
| "reward_std": 0.32530001799265545, |
| "rewards/accuracy_reward": 0.8095238208770752, |
| "step": 36 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 12022.5244140625, |
| "epoch": 0.6201117318435754, |
| "grad_norm": 0.1912575364112854, |
| "kl": 0.000637054443359375, |
| "learning_rate": 1.104619250896399e-06, |
| "loss": 0.0, |
| "reward": 0.571428601940473, |
| "reward_std": 0.4146263202031453, |
| "rewards/accuracy_reward": 0.571428601940473, |
| "step": 37 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 11669.95263671875, |
| "epoch": 0.6368715083798883, |
| "grad_norm": 0.2329685539007187, |
| "kl": 0.0008748372395833334, |
| "learning_rate": 1.0195952192057437e-06, |
| "loss": 0.0, |
| "reward": 0.666666716337204, |
| "reward_std": 0.4146263202031453, |
| "rewards/accuracy_reward": 0.666666716337204, |
| "step": 38 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8310.952799479166, |
| "epoch": 0.6536312849162011, |
| "grad_norm": 0.17929613590240479, |
| "kl": 0.0005823771158854166, |
| "learning_rate": 9.362586268286431e-07, |
| "loss": 0.0, |
| "reward": 0.7142857313156128, |
| "reward_std": 0.3408241669336955, |
| "rewards/accuracy_reward": 0.7142857313156128, |
| "step": 39 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9006.0478515625, |
| "epoch": 0.6703910614525139, |
| "grad_norm": 0.20117197930812836, |
| "kl": 0.000732421875, |
| "learning_rate": 8.549021965852198e-07, |
| "loss": 0.0, |
| "reward": 0.761904795964559, |
| "reward_std": 0.30416231354077655, |
| "rewards/accuracy_reward": 0.761904795964559, |
| "step": 40 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7405.619303385417, |
| "epoch": 0.6871508379888268, |
| "grad_norm": 0.26455193758010864, |
| "kl": 0.000705718994140625, |
| "learning_rate": 7.758116959038829e-07, |
| "loss": 0.0, |
| "reward": 0.809523860613505, |
| "reward_std": 0.4146263202031453, |
| "rewards/accuracy_reward": 0.809523860613505, |
| "step": 41 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9592.6669921875, |
| "epoch": 0.7039106145251397, |
| "grad_norm": 0.27179741859436035, |
| "kl": 0.0007375081380208334, |
| "learning_rate": 6.992649330528146e-07, |
| "loss": 0.0, |
| "reward": 0.4761904974778493, |
| "reward_std": 0.5189983248710632, |
| "rewards/accuracy_reward": 0.4761904974778493, |
| "step": 42 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9534.6669921875, |
| "epoch": 0.7206703910614525, |
| "grad_norm": 0.20514477789402008, |
| "kl": 0.0008087158203125, |
| "learning_rate": 6.255307813282921e-07, |
| "loss": 0.0, |
| "reward": 0.6666666766007742, |
| "reward_std": 0.32530001799265545, |
| "rewards/accuracy_reward": 0.6666666766007742, |
| "step": 43 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9129.809895833334, |
| "epoch": 0.7374301675977654, |
| "grad_norm": 0.2465788871049881, |
| "kl": 0.0007044474283854166, |
| "learning_rate": 5.548682346274336e-07, |
| "loss": 0.0, |
| "reward": 0.6666667064030966, |
| "reward_std": 0.4668123225371043, |
| "rewards/accuracy_reward": 0.6666667064030966, |
| "step": 44 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 11743.572102864584, |
| "epoch": 0.7541899441340782, |
| "grad_norm": 0.2178698033094406, |
| "kl": 0.0009282430013020834, |
| "learning_rate": 4.875254977226845e-07, |
| "loss": 0.0, |
| "reward": 0.5238095422585806, |
| "reward_std": 0.5034741759300232, |
| "rewards/accuracy_reward": 0.5238095422585806, |
| "step": 45 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9390.333821614584, |
| "epoch": 0.770949720670391, |
| "grad_norm": 0.24005185067653656, |
| "kl": 0.0007349650065104166, |
| "learning_rate": 4.2373911443350296e-07, |
| "loss": 0.0, |
| "reward": 0.8571428855260214, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.8571428855260214, |
| "step": 46 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10397.190755208334, |
| "epoch": 0.7877094972067039, |
| "grad_norm": 0.2124994695186615, |
| "kl": 0.000736236572265625, |
| "learning_rate": 3.637331367575698e-07, |
| "loss": 0.0, |
| "reward": 0.6190476616223654, |
| "reward_std": 0.48233647147814435, |
| "rewards/accuracy_reward": 0.6190476616223654, |
| "step": 47 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9078.762369791666, |
| "epoch": 0.8044692737430168, |
| "grad_norm": 0.19195324182510376, |
| "kl": 0.0006663004557291666, |
| "learning_rate": 3.0771833787996994e-07, |
| "loss": 0.0, |
| "reward": 0.8571428855260214, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.8571428855260214, |
| "step": 48 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8796.428873697916, |
| "epoch": 0.8212290502793296, |
| "grad_norm": 0.16551946103572845, |
| "kl": 0.0008672078450520834, |
| "learning_rate": 2.5589147182469736e-07, |
| "loss": 0.0, |
| "reward": 0.6190476417541504, |
| "reward_std": 0.1259881556034088, |
| "rewards/accuracy_reward": 0.6190476417541504, |
| "step": 49 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9645.048014322916, |
| "epoch": 0.8379888268156425, |
| "grad_norm": 0.23935681581497192, |
| "kl": 0.000896453857421875, |
| "learning_rate": 2.0843458234896666e-07, |
| "loss": 0.0, |
| "reward": 0.4761905074119568, |
| "reward_std": 0.45128817359606427, |
| "rewards/accuracy_reward": 0.4761905074119568, |
| "step": 50 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 9876.571940104166, |
| "epoch": 0.8547486033519553, |
| "grad_norm": 0.19879846274852753, |
| "kl": 0.000865936279296875, |
| "learning_rate": 1.6551436350787918e-07, |
| "loss": 0.0, |
| "reward": 0.7142857412497202, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.7142857412497202, |
| "step": 51 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 12011.71484375, |
| "epoch": 0.8715083798882681, |
| "grad_norm": 0.2283555269241333, |
| "kl": 0.000827789306640625, |
| "learning_rate": 1.272815741354723e-07, |
| "loss": 0.0, |
| "reward": 0.7142857511838278, |
| "reward_std": 0.4668123225371043, |
| "rewards/accuracy_reward": 0.7142857511838278, |
| "step": 52 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7346.61962890625, |
| "epoch": 0.888268156424581, |
| "grad_norm": 0.23349399864673615, |
| "kl": 0.0007680257161458334, |
| "learning_rate": 9.387050829881866e-08, |
| "loss": 0.0, |
| "reward": 0.8571428855260214, |
| "reward_std": 0.2886381645997365, |
| "rewards/accuracy_reward": 0.8571428855260214, |
| "step": 53 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 10462.52392578125, |
| "epoch": 0.9050279329608939, |
| "grad_norm": 0.26301538944244385, |
| "kl": 0.001003265380859375, |
| "learning_rate": 6.539852358521637e-08, |
| "loss": 0.0, |
| "reward": 0.4761904925107956, |
| "reward_std": 0.4668123225371043, |
| "rewards/accuracy_reward": 0.4761904925107956, |
| "step": 54 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 8198.857421875, |
| "epoch": 0.9217877094972067, |
| "grad_norm": 0.16529662907123566, |
| "kl": 0.0009066263834635416, |
| "learning_rate": 4.1965628879383876e-08, |
| "loss": 0.0, |
| "reward": 0.9047619104385376, |
| "reward_std": 0.16265000899632773, |
| "rewards/accuracy_reward": 0.9047619104385376, |
| "step": 55 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 13437.333984375, |
| "epoch": 0.9385474860335196, |
| "grad_norm": 0.16132879257202148, |
| "kl": 0.000797271728515625, |
| "learning_rate": 2.3654133078604757e-08, |
| "loss": 0.0, |
| "reward": 0.3333333432674408, |
| "reward_std": 0.32530001799265545, |
| "rewards/accuracy_reward": 0.3333333432674408, |
| "step": 56 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 12116.429036458334, |
| "epoch": 0.9553072625698324, |
| "grad_norm": 0.16875086724758148, |
| "kl": 0.0007883707682291666, |
| "learning_rate": 1.0528355979724625e-08, |
| "loss": 0.0, |
| "reward": 0.7619048158327738, |
| "reward_std": 0.43015046914418537, |
| "rewards/accuracy_reward": 0.7619048158327738, |
| "step": 57 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 7293.76220703125, |
| "epoch": 0.9720670391061452, |
| "grad_norm": 0.1918724626302719, |
| "kl": 0.0012308756510416667, |
| "learning_rate": 2.634402353517973e-09, |
| "loss": 0.0, |
| "reward": 0.7142857313156128, |
| "reward_std": 0.3408241669336955, |
| "rewards/accuracy_reward": 0.7142857313156128, |
| "step": 58 |
| }, |
| { |
| "clip_ratio": 0.0, |
| "completion_length": 11295.5, |
| "epoch": 0.9888268156424581, |
| "grad_norm": 0.1678023487329483, |
| "kl": 0.0007527669270833334, |
| "learning_rate": 0.0, |
| "loss": 0.0, |
| "reward": 0.6190476417541504, |
| "reward_std": 0.35634831587473553, |
| "rewards/accuracy_reward": 0.6190476417541504, |
| "step": 59 |
| }, |
| { |
| "epoch": 0.9888268156424581, |
| "step": 59, |
| "total_flos": 0.0, |
| "train_loss": 2.769587171291079e-05, |
| "train_runtime": 35654.1932, |
| "train_samples_per_second": 0.005, |
| "train_steps_per_second": 0.002 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 59, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
