| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9985185185185185, |
| "eval_steps": 500, |
| "global_step": 337, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.002962962962962963, |
| "grad_norm": 314.5073547363281, |
| "learning_rate": 1.9940652818991096e-06, |
| "logits/chosen": -0.341796875, |
| "logits/rejected": -0.35546875, |
| "logps/chosen": -118.0, |
| "logps/rejected": -159.0, |
| "loss": 0.6914, |
| "rewards/accuracies": 0.0, |
| "rewards/chosen": 0.0, |
| "rewards/margins": 0.0, |
| "rewards/rejected": 0.0, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.005925925925925926, |
| "grad_norm": 291.4620056152344, |
| "learning_rate": 1.9881305637982194e-06, |
| "logits/chosen": -0.388671875, |
| "logits/rejected": -0.396484375, |
| "logps/chosen": -130.0, |
| "logps/rejected": -110.0, |
| "loss": 0.6969, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": -0.000782012939453125, |
| "rewards/margins": -0.003143310546875, |
| "rewards/rejected": 0.002349853515625, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.008888888888888889, |
| "grad_norm": 289.64862060546875, |
| "learning_rate": 1.9821958456973295e-06, |
| "logits/chosen": -0.361328125, |
| "logits/rejected": -0.369140625, |
| "logps/chosen": -116.0, |
| "logps/rejected": -114.0, |
| "loss": 0.6938, |
| "rewards/accuracies": 0.34375, |
| "rewards/chosen": -0.0133056640625, |
| "rewards/margins": -0.001556396484375, |
| "rewards/rejected": -0.01171875, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.011851851851851851, |
| "grad_norm": 295.88763427734375, |
| "learning_rate": 1.9762611275964393e-06, |
| "logits/chosen": -0.359375, |
| "logits/rejected": -0.3671875, |
| "logps/chosen": -108.5, |
| "logps/rejected": -129.0, |
| "loss": 0.6779, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.0172119140625, |
| "rewards/margins": 0.03515625, |
| "rewards/rejected": -0.0179443359375, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.014814814814814815, |
| "grad_norm": 268.0253601074219, |
| "learning_rate": 1.970326409495549e-06, |
| "logits/chosen": -0.330078125, |
| "logits/rejected": -0.3671875, |
| "logps/chosen": -108.0, |
| "logps/rejected": -111.5, |
| "loss": 0.6913, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": 0.006256103515625, |
| "rewards/margins": 0.0054931640625, |
| "rewards/rejected": 0.00078582763671875, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.017777777777777778, |
| "grad_norm": 262.7245178222656, |
| "learning_rate": 1.9643916913946587e-06, |
| "logits/chosen": -0.3671875, |
| "logits/rejected": -0.384765625, |
| "logps/chosen": -95.5, |
| "logps/rejected": -104.5, |
| "loss": 0.6852, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": 0.0133056640625, |
| "rewards/margins": 0.016357421875, |
| "rewards/rejected": -0.0031280517578125, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.02074074074074074, |
| "grad_norm": 292.5845642089844, |
| "learning_rate": 1.9584569732937684e-06, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -116.0, |
| "logps/rejected": -127.0, |
| "loss": 0.6946, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": -0.01019287109375, |
| "rewards/margins": -0.0031280517578125, |
| "rewards/rejected": -0.007049560546875, |
| "step": 7 |
| }, |
| { |
| "epoch": 0.023703703703703703, |
| "grad_norm": 273.1530456542969, |
| "learning_rate": 1.952522255192878e-06, |
| "logits/chosen": -0.3828125, |
| "logits/rejected": -0.3671875, |
| "logps/chosen": -121.5, |
| "logps/rejected": -116.0, |
| "loss": 0.6863, |
| "rewards/accuracies": 0.40625, |
| "rewards/chosen": 0.01019287109375, |
| "rewards/margins": 0.0179443359375, |
| "rewards/rejected": -0.0078125, |
| "step": 8 |
| }, |
| { |
| "epoch": 0.02666666666666667, |
| "grad_norm": 308.3035888671875, |
| "learning_rate": 1.946587537091988e-06, |
| "logits/chosen": -0.396484375, |
| "logits/rejected": -0.384765625, |
| "logps/chosen": -131.0, |
| "logps/rejected": -114.5, |
| "loss": 0.6702, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.042236328125, |
| "rewards/margins": 0.04296875, |
| "rewards/rejected": -0.000762939453125, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.02962962962962963, |
| "grad_norm": 295.3406982421875, |
| "learning_rate": 1.9406528189910976e-06, |
| "logits/chosen": -0.384765625, |
| "logits/rejected": -0.390625, |
| "logps/chosen": -120.5, |
| "logps/rejected": -118.5, |
| "loss": 0.6799, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.0179443359375, |
| "rewards/margins": 0.0296630859375, |
| "rewards/rejected": -0.01171875, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.03259259259259259, |
| "grad_norm": 260.79620361328125, |
| "learning_rate": 1.934718100890208e-06, |
| "logits/chosen": -0.3671875, |
| "logits/rejected": -0.3984375, |
| "logps/chosen": -99.5, |
| "logps/rejected": -83.5, |
| "loss": 0.6849, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0172119140625, |
| "rewards/margins": 0.0191650390625, |
| "rewards/rejected": -0.001953125, |
| "step": 11 |
| }, |
| { |
| "epoch": 0.035555555555555556, |
| "grad_norm": 284.2510986328125, |
| "learning_rate": 1.9287833827893175e-06, |
| "logits/chosen": -0.337890625, |
| "logits/rejected": -0.353515625, |
| "logps/chosen": -114.5, |
| "logps/rejected": -105.0, |
| "loss": 0.6869, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.030517578125, |
| "rewards/margins": 0.0164794921875, |
| "rewards/rejected": 0.0140380859375, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.03851851851851852, |
| "grad_norm": 240.7752685546875, |
| "learning_rate": 1.9228486646884273e-06, |
| "logits/chosen": -0.357421875, |
| "logits/rejected": -0.404296875, |
| "logps/chosen": -91.0, |
| "logps/rejected": -98.0, |
| "loss": 0.687, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.0179443359375, |
| "rewards/margins": 0.01446533203125, |
| "rewards/rejected": 0.003509521484375, |
| "step": 13 |
| }, |
| { |
| "epoch": 0.04148148148148148, |
| "grad_norm": 349.65118408203125, |
| "learning_rate": 1.916913946587537e-06, |
| "logits/chosen": -0.35546875, |
| "logits/rejected": -0.337890625, |
| "logps/chosen": -105.0, |
| "logps/rejected": -158.0, |
| "loss": 0.7042, |
| "rewards/accuracies": 0.3125, |
| "rewards/chosen": 0.01409912109375, |
| "rewards/margins": -0.01806640625, |
| "rewards/rejected": 0.031982421875, |
| "step": 14 |
| }, |
| { |
| "epoch": 0.044444444444444446, |
| "grad_norm": 274.92315673828125, |
| "learning_rate": 1.9109792284866467e-06, |
| "logits/chosen": -0.37890625, |
| "logits/rejected": -0.36328125, |
| "logps/chosen": -114.0, |
| "logps/rejected": -107.5, |
| "loss": 0.6887, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": 0.0234375, |
| "rewards/margins": 0.01409912109375, |
| "rewards/rejected": 0.0093994140625, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.047407407407407405, |
| "grad_norm": 284.9162292480469, |
| "learning_rate": 1.9050445103857567e-06, |
| "logits/chosen": -0.369140625, |
| "logits/rejected": -0.419921875, |
| "logps/chosen": -101.5, |
| "logps/rejected": -113.0, |
| "loss": 0.6893, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.039794921875, |
| "rewards/margins": 0.00860595703125, |
| "rewards/rejected": 0.03125, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.05037037037037037, |
| "grad_norm": 295.5323486328125, |
| "learning_rate": 1.8991097922848664e-06, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.408203125, |
| "logps/chosen": -124.0, |
| "logps/rejected": -167.0, |
| "loss": 0.6849, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0242919921875, |
| "rewards/margins": 0.0257568359375, |
| "rewards/rejected": -0.001556396484375, |
| "step": 17 |
| }, |
| { |
| "epoch": 0.05333333333333334, |
| "grad_norm": 357.27410888671875, |
| "learning_rate": 1.8931750741839763e-06, |
| "logits/chosen": -0.40625, |
| "logits/rejected": -0.400390625, |
| "logps/chosen": -136.0, |
| "logps/rejected": -220.0, |
| "loss": 0.6925, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": 0.0289306640625, |
| "rewards/margins": 0.0031280517578125, |
| "rewards/rejected": 0.0257568359375, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.056296296296296296, |
| "grad_norm": 286.5829772949219, |
| "learning_rate": 1.887240356083086e-06, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.439453125, |
| "logps/chosen": -110.5, |
| "logps/rejected": -138.0, |
| "loss": 0.6929, |
| "rewards/accuracies": 0.34375, |
| "rewards/chosen": 0.043701171875, |
| "rewards/margins": 0.0078125, |
| "rewards/rejected": 0.035888671875, |
| "step": 19 |
| }, |
| { |
| "epoch": 0.05925925925925926, |
| "grad_norm": 502.36614990234375, |
| "learning_rate": 1.8813056379821956e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.361328125, |
| "logps/chosen": -105.5, |
| "logps/rejected": -97.5, |
| "loss": 0.6826, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0172119140625, |
| "rewards/margins": 0.0289306640625, |
| "rewards/rejected": -0.01165771484375, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.06222222222222222, |
| "grad_norm": 258.4698486328125, |
| "learning_rate": 1.8753709198813055e-06, |
| "logits/chosen": -0.33203125, |
| "logits/rejected": -0.36328125, |
| "logps/chosen": -94.0, |
| "logps/rejected": -98.5, |
| "loss": 0.6742, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.05859375, |
| "rewards/margins": 0.038330078125, |
| "rewards/rejected": 0.0203857421875, |
| "step": 21 |
| }, |
| { |
| "epoch": 0.06518518518518518, |
| "grad_norm": 282.2925109863281, |
| "learning_rate": 1.8694362017804153e-06, |
| "logits/chosen": -0.390625, |
| "logits/rejected": -0.37109375, |
| "logps/chosen": -125.0, |
| "logps/rejected": -108.5, |
| "loss": 0.6752, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.04541015625, |
| "rewards/margins": 0.038330078125, |
| "rewards/rejected": 0.00701904296875, |
| "step": 22 |
| }, |
| { |
| "epoch": 0.06814814814814815, |
| "grad_norm": 291.8224182128906, |
| "learning_rate": 1.863501483679525e-06, |
| "logits/chosen": -0.3828125, |
| "logits/rejected": -0.388671875, |
| "logps/chosen": -111.0, |
| "logps/rejected": -108.5, |
| "loss": 0.6808, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.06396484375, |
| "rewards/margins": 0.0281982421875, |
| "rewards/rejected": 0.035888671875, |
| "step": 23 |
| }, |
| { |
| "epoch": 0.07111111111111111, |
| "grad_norm": 258.92022705078125, |
| "learning_rate": 1.857566765578635e-06, |
| "logits/chosen": -0.369140625, |
| "logits/rejected": -0.38671875, |
| "logps/chosen": -93.5, |
| "logps/rejected": -97.5, |
| "loss": 0.6665, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.076171875, |
| "rewards/margins": 0.058349609375, |
| "rewards/rejected": 0.0179443359375, |
| "step": 24 |
| }, |
| { |
| "epoch": 0.07407407407407407, |
| "grad_norm": 281.2393798828125, |
| "learning_rate": 1.8516320474777447e-06, |
| "logits/chosen": -0.37109375, |
| "logits/rejected": -0.369140625, |
| "logps/chosen": -106.5, |
| "logps/rejected": -115.0, |
| "loss": 0.6932, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": 0.03125, |
| "rewards/margins": 0.00628662109375, |
| "rewards/rejected": 0.0250244140625, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.07703703703703704, |
| "grad_norm": 279.8067626953125, |
| "learning_rate": 1.8456973293768546e-06, |
| "logits/chosen": -0.369140625, |
| "logits/rejected": -0.36328125, |
| "logps/chosen": -120.5, |
| "logps/rejected": -113.5, |
| "loss": 0.6689, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.061767578125, |
| "rewards/margins": 0.049072265625, |
| "rewards/rejected": 0.01287841796875, |
| "step": 26 |
| }, |
| { |
| "epoch": 0.08, |
| "grad_norm": 293.7424621582031, |
| "learning_rate": 1.8397626112759643e-06, |
| "logits/chosen": -0.396484375, |
| "logits/rejected": -0.416015625, |
| "logps/chosen": -131.0, |
| "logps/rejected": -135.0, |
| "loss": 0.6686, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.07763671875, |
| "rewards/margins": 0.052490234375, |
| "rewards/rejected": 0.0250244140625, |
| "step": 27 |
| }, |
| { |
| "epoch": 0.08296296296296296, |
| "grad_norm": 287.2871398925781, |
| "learning_rate": 1.833827893175074e-06, |
| "logits/chosen": -0.375, |
| "logits/rejected": -0.408203125, |
| "logps/chosen": -113.0, |
| "logps/rejected": -150.0, |
| "loss": 0.6848, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.053955078125, |
| "rewards/margins": 0.02978515625, |
| "rewards/rejected": 0.0242919921875, |
| "step": 28 |
| }, |
| { |
| "epoch": 0.08592592592592592, |
| "grad_norm": 275.70904541015625, |
| "learning_rate": 1.827893175074184e-06, |
| "logits/chosen": -0.3671875, |
| "logits/rejected": -0.36328125, |
| "logps/chosen": -108.0, |
| "logps/rejected": -106.5, |
| "loss": 0.6874, |
| "rewards/accuracies": 0.40625, |
| "rewards/chosen": 0.047607421875, |
| "rewards/margins": 0.014892578125, |
| "rewards/rejected": 0.032958984375, |
| "step": 29 |
| }, |
| { |
| "epoch": 0.08888888888888889, |
| "grad_norm": 275.59912109375, |
| "learning_rate": 1.8219584569732937e-06, |
| "logits/chosen": -0.341796875, |
| "logits/rejected": -0.326171875, |
| "logps/chosen": -102.0, |
| "logps/rejected": -109.0, |
| "loss": 0.6935, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": 0.03759765625, |
| "rewards/margins": -0.0022735595703125, |
| "rewards/rejected": 0.039794921875, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.09185185185185185, |
| "grad_norm": 288.207275390625, |
| "learning_rate": 1.8160237388724035e-06, |
| "logits/chosen": -0.357421875, |
| "logits/rejected": -0.328125, |
| "logps/chosen": -129.0, |
| "logps/rejected": -109.0, |
| "loss": 0.6677, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0791015625, |
| "rewards/margins": 0.05712890625, |
| "rewards/rejected": 0.0218505859375, |
| "step": 31 |
| }, |
| { |
| "epoch": 0.09481481481481481, |
| "grad_norm": 290.9600524902344, |
| "learning_rate": 1.8100890207715134e-06, |
| "logits/chosen": -0.37109375, |
| "logits/rejected": -0.380859375, |
| "logps/chosen": -124.0, |
| "logps/rejected": -112.5, |
| "loss": 0.6808, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.035888671875, |
| "rewards/margins": 0.034423828125, |
| "rewards/rejected": 0.001556396484375, |
| "step": 32 |
| }, |
| { |
| "epoch": 0.09777777777777778, |
| "grad_norm": 274.994384765625, |
| "learning_rate": 1.804154302670623e-06, |
| "logits/chosen": -0.384765625, |
| "logits/rejected": -0.419921875, |
| "logps/chosen": -108.0, |
| "logps/rejected": -108.0, |
| "loss": 0.6702, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.046875, |
| "rewards/margins": 0.04736328125, |
| "rewards/rejected": -0.0003910064697265625, |
| "step": 33 |
| }, |
| { |
| "epoch": 0.10074074074074074, |
| "grad_norm": 302.385009765625, |
| "learning_rate": 1.7982195845697327e-06, |
| "logits/chosen": -0.361328125, |
| "logits/rejected": -0.3828125, |
| "logps/chosen": -111.5, |
| "logps/rejected": -133.0, |
| "loss": 0.666, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.06640625, |
| "rewards/margins": 0.060302734375, |
| "rewards/rejected": 0.006256103515625, |
| "step": 34 |
| }, |
| { |
| "epoch": 0.1037037037037037, |
| "grad_norm": 276.3713684082031, |
| "learning_rate": 1.7922848664688426e-06, |
| "logits/chosen": -0.400390625, |
| "logits/rejected": -0.41015625, |
| "logps/chosen": -114.0, |
| "logps/rejected": -128.0, |
| "loss": 0.6787, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.032958984375, |
| "rewards/margins": 0.036376953125, |
| "rewards/rejected": -0.003509521484375, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.10666666666666667, |
| "grad_norm": 305.0614929199219, |
| "learning_rate": 1.7863501483679523e-06, |
| "logits/chosen": -0.369140625, |
| "logits/rejected": -0.400390625, |
| "logps/chosen": -110.5, |
| "logps/rejected": -165.0, |
| "loss": 0.6464, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.0791015625, |
| "rewards/margins": 0.1064453125, |
| "rewards/rejected": -0.02734375, |
| "step": 36 |
| }, |
| { |
| "epoch": 0.10962962962962963, |
| "grad_norm": 333.9048767089844, |
| "learning_rate": 1.7804154302670623e-06, |
| "logits/chosen": -0.341796875, |
| "logits/rejected": -0.373046875, |
| "logps/chosen": -112.0, |
| "logps/rejected": -118.5, |
| "loss": 0.6801, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.06103515625, |
| "rewards/margins": 0.03271484375, |
| "rewards/rejected": 0.0281982421875, |
| "step": 37 |
| }, |
| { |
| "epoch": 0.11259259259259259, |
| "grad_norm": 282.57940673828125, |
| "learning_rate": 1.774480712166172e-06, |
| "logits/chosen": -0.39453125, |
| "logits/rejected": -0.455078125, |
| "logps/chosen": -112.0, |
| "logps/rejected": -122.0, |
| "loss": 0.6456, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.08349609375, |
| "rewards/margins": 0.10546875, |
| "rewards/rejected": -0.02197265625, |
| "step": 38 |
| }, |
| { |
| "epoch": 0.11555555555555555, |
| "grad_norm": 296.46282958984375, |
| "learning_rate": 1.7685459940652817e-06, |
| "logits/chosen": -0.37109375, |
| "logits/rejected": -0.404296875, |
| "logps/chosen": -100.0, |
| "logps/rejected": -130.0, |
| "loss": 0.6799, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0289306640625, |
| "rewards/margins": 0.039794921875, |
| "rewards/rejected": -0.01092529296875, |
| "step": 39 |
| }, |
| { |
| "epoch": 0.11851851851851852, |
| "grad_norm": 283.2777404785156, |
| "learning_rate": 1.7626112759643917e-06, |
| "logits/chosen": -0.31640625, |
| "logits/rejected": -0.33984375, |
| "logps/chosen": -105.5, |
| "logps/rejected": -106.5, |
| "loss": 0.6694, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.04150390625, |
| "rewards/margins": 0.0517578125, |
| "rewards/rejected": -0.0101318359375, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.12148148148148148, |
| "grad_norm": 273.7195129394531, |
| "learning_rate": 1.7566765578635014e-06, |
| "logits/chosen": -0.40625, |
| "logits/rejected": -0.392578125, |
| "logps/chosen": -107.0, |
| "logps/rejected": -103.0, |
| "loss": 0.6556, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.03466796875, |
| "rewards/margins": 0.0810546875, |
| "rewards/rejected": -0.046142578125, |
| "step": 41 |
| }, |
| { |
| "epoch": 0.12444444444444444, |
| "grad_norm": 294.5924072265625, |
| "learning_rate": 1.7507418397626112e-06, |
| "logits/chosen": -0.33984375, |
| "logits/rejected": -0.36328125, |
| "logps/chosen": -105.0, |
| "logps/rejected": -156.0, |
| "loss": 0.6646, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.046142578125, |
| "rewards/margins": 0.07568359375, |
| "rewards/rejected": -0.029296875, |
| "step": 42 |
| }, |
| { |
| "epoch": 0.1274074074074074, |
| "grad_norm": 307.97613525390625, |
| "learning_rate": 1.744807121661721e-06, |
| "logits/chosen": -0.396484375, |
| "logits/rejected": -0.43359375, |
| "logps/chosen": -118.0, |
| "logps/rejected": -133.0, |
| "loss": 0.6605, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.047607421875, |
| "rewards/margins": 0.07421875, |
| "rewards/rejected": -0.026611328125, |
| "step": 43 |
| }, |
| { |
| "epoch": 0.13037037037037036, |
| "grad_norm": 281.07147216796875, |
| "learning_rate": 1.7388724035608308e-06, |
| "logits/chosen": -0.40234375, |
| "logits/rejected": -0.37109375, |
| "logps/chosen": -119.5, |
| "logps/rejected": -104.5, |
| "loss": 0.6449, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.06884765625, |
| "rewards/margins": 0.10107421875, |
| "rewards/rejected": -0.031982421875, |
| "step": 44 |
| }, |
| { |
| "epoch": 0.13333333333333333, |
| "grad_norm": 305.4576110839844, |
| "learning_rate": 1.7329376854599408e-06, |
| "logits/chosen": -0.345703125, |
| "logits/rejected": -0.33203125, |
| "logps/chosen": -116.0, |
| "logps/rejected": -128.0, |
| "loss": 0.6584, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.05078125, |
| "rewards/margins": 0.09228515625, |
| "rewards/rejected": -0.04150390625, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.1362962962962963, |
| "grad_norm": 296.2742614746094, |
| "learning_rate": 1.7270029673590503e-06, |
| "logits/chosen": -0.396484375, |
| "logits/rejected": -0.375, |
| "logps/chosen": -124.0, |
| "logps/rejected": -105.0, |
| "loss": 0.6591, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.08837890625, |
| "rewards/margins": 0.078125, |
| "rewards/rejected": 0.0101318359375, |
| "step": 46 |
| }, |
| { |
| "epoch": 0.13925925925925925, |
| "grad_norm": 271.832275390625, |
| "learning_rate": 1.72106824925816e-06, |
| "logits/chosen": -0.359375, |
| "logits/rejected": -0.384765625, |
| "logps/chosen": -109.5, |
| "logps/rejected": -119.5, |
| "loss": 0.6566, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.07373046875, |
| "rewards/margins": 0.08935546875, |
| "rewards/rejected": -0.015625, |
| "step": 47 |
| }, |
| { |
| "epoch": 0.14222222222222222, |
| "grad_norm": 300.9525146484375, |
| "learning_rate": 1.71513353115727e-06, |
| "logits/chosen": -0.38671875, |
| "logits/rejected": -0.390625, |
| "logps/chosen": -116.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6597, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.016357421875, |
| "rewards/margins": 0.091796875, |
| "rewards/rejected": -0.0751953125, |
| "step": 48 |
| }, |
| { |
| "epoch": 0.1451851851851852, |
| "grad_norm": 284.7471008300781, |
| "learning_rate": 1.7091988130563797e-06, |
| "logits/chosen": -0.361328125, |
| "logits/rejected": -0.369140625, |
| "logps/chosen": -116.0, |
| "logps/rejected": -107.0, |
| "loss": 0.6768, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.0164794921875, |
| "rewards/margins": 0.039794921875, |
| "rewards/rejected": -0.0234375, |
| "step": 49 |
| }, |
| { |
| "epoch": 0.14814814814814814, |
| "grad_norm": 244.4700469970703, |
| "learning_rate": 1.7032640949554894e-06, |
| "logits/chosen": -0.357421875, |
| "logits/rejected": -0.373046875, |
| "logps/chosen": -87.0, |
| "logps/rejected": -86.0, |
| "loss": 0.6724, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.005462646484375, |
| "rewards/margins": 0.055419921875, |
| "rewards/rejected": -0.050048828125, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.1511111111111111, |
| "grad_norm": 336.86126708984375, |
| "learning_rate": 1.6973293768545994e-06, |
| "logits/chosen": -0.373046875, |
| "logits/rejected": -0.3828125, |
| "logps/chosen": -101.5, |
| "logps/rejected": -122.0, |
| "loss": 0.6431, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.039794921875, |
| "rewards/margins": 0.11669921875, |
| "rewards/rejected": -0.07666015625, |
| "step": 51 |
| }, |
| { |
| "epoch": 0.15407407407407409, |
| "grad_norm": 302.233642578125, |
| "learning_rate": 1.691394658753709e-06, |
| "logits/chosen": -0.36328125, |
| "logits/rejected": -0.416015625, |
| "logps/chosen": -108.5, |
| "logps/rejected": -130.0, |
| "loss": 0.6747, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0101318359375, |
| "rewards/margins": 0.04541015625, |
| "rewards/rejected": -0.03515625, |
| "step": 52 |
| }, |
| { |
| "epoch": 0.15703703703703703, |
| "grad_norm": 298.18658447265625, |
| "learning_rate": 1.685459940652819e-06, |
| "logits/chosen": -0.359375, |
| "logits/rejected": -0.408203125, |
| "logps/chosen": -110.0, |
| "logps/rejected": -153.0, |
| "loss": 0.6451, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.03125, |
| "rewards/margins": 0.13671875, |
| "rewards/rejected": -0.10546875, |
| "step": 53 |
| }, |
| { |
| "epoch": 0.16, |
| "grad_norm": 262.4490051269531, |
| "learning_rate": 1.6795252225519288e-06, |
| "logits/chosen": -0.353515625, |
| "logits/rejected": -0.361328125, |
| "logps/chosen": -103.5, |
| "logps/rejected": -105.5, |
| "loss": 0.6473, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.06884765625, |
| "rewards/margins": 0.10107421875, |
| "rewards/rejected": -0.031982421875, |
| "step": 54 |
| }, |
| { |
| "epoch": 0.16296296296296298, |
| "grad_norm": 307.3597412109375, |
| "learning_rate": 1.6735905044510385e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -114.5, |
| "logps/rejected": -179.0, |
| "loss": 0.6312, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.05712890625, |
| "rewards/margins": 0.166015625, |
| "rewards/rejected": -0.10888671875, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.16592592592592592, |
| "grad_norm": 319.4726257324219, |
| "learning_rate": 1.6676557863501484e-06, |
| "logits/chosen": -0.3671875, |
| "logits/rejected": -0.3515625, |
| "logps/chosen": -100.5, |
| "logps/rejected": -96.5, |
| "loss": 0.6974, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.007781982421875, |
| "rewards/margins": 0.0015106201171875, |
| "rewards/rejected": -0.00933837890625, |
| "step": 56 |
| }, |
| { |
| "epoch": 0.1688888888888889, |
| "grad_norm": 315.0474548339844, |
| "learning_rate": 1.6617210682492582e-06, |
| "logits/chosen": -0.412109375, |
| "logits/rejected": -0.42578125, |
| "logps/chosen": -137.0, |
| "logps/rejected": -184.0, |
| "loss": 0.6286, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.00775146484375, |
| "rewards/margins": 0.1484375, |
| "rewards/rejected": -0.140625, |
| "step": 57 |
| }, |
| { |
| "epoch": 0.17185185185185184, |
| "grad_norm": 302.9765625, |
| "learning_rate": 1.655786350148368e-06, |
| "logits/chosen": -0.359375, |
| "logits/rejected": -0.3828125, |
| "logps/chosen": -103.5, |
| "logps/rejected": -110.5, |
| "loss": 0.6796, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.04296875, |
| "rewards/margins": 0.038330078125, |
| "rewards/rejected": 0.00469970703125, |
| "step": 58 |
| }, |
| { |
| "epoch": 0.1748148148148148, |
| "grad_norm": 272.1220703125, |
| "learning_rate": 1.6498516320474776e-06, |
| "logits/chosen": -0.388671875, |
| "logits/rejected": -0.390625, |
| "logps/chosen": -110.5, |
| "logps/rejected": -96.5, |
| "loss": 0.6531, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.0546875, |
| "rewards/margins": 0.09619140625, |
| "rewards/rejected": -0.04150390625, |
| "step": 59 |
| }, |
| { |
| "epoch": 0.17777777777777778, |
| "grad_norm": 307.90301513671875, |
| "learning_rate": 1.6439169139465874e-06, |
| "logits/chosen": -0.369140625, |
| "logits/rejected": -0.390625, |
| "logps/chosen": -112.0, |
| "logps/rejected": -124.0, |
| "loss": 0.6341, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.0625, |
| "rewards/margins": 0.126953125, |
| "rewards/rejected": -0.06494140625, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.18074074074074073, |
| "grad_norm": 312.32916259765625, |
| "learning_rate": 1.637982195845697e-06, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.376953125, |
| "logps/chosen": -132.0, |
| "logps/rejected": -113.0, |
| "loss": 0.647, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.04150390625, |
| "rewards/margins": 0.10888671875, |
| "rewards/rejected": -0.0673828125, |
| "step": 61 |
| }, |
| { |
| "epoch": 0.1837037037037037, |
| "grad_norm": 327.3997802734375, |
| "learning_rate": 1.632047477744807e-06, |
| "logits/chosen": -0.404296875, |
| "logits/rejected": -0.3828125, |
| "logps/chosen": -115.0, |
| "logps/rejected": -153.0, |
| "loss": 0.6488, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.0179443359375, |
| "rewards/margins": 0.11083984375, |
| "rewards/rejected": -0.09326171875, |
| "step": 62 |
| }, |
| { |
| "epoch": 0.18666666666666668, |
| "grad_norm": 290.80908203125, |
| "learning_rate": 1.6261127596439168e-06, |
| "logits/chosen": -0.337890625, |
| "logits/rejected": -0.39453125, |
| "logps/chosen": -114.0, |
| "logps/rejected": -120.0, |
| "loss": 0.6129, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.057861328125, |
| "rewards/margins": 0.1826171875, |
| "rewards/rejected": -0.12451171875, |
| "step": 63 |
| }, |
| { |
| "epoch": 0.18962962962962962, |
| "grad_norm": 266.57843017578125, |
| "learning_rate": 1.6201780415430267e-06, |
| "logits/chosen": -0.3515625, |
| "logits/rejected": -0.373046875, |
| "logps/chosen": -105.0, |
| "logps/rejected": -88.5, |
| "loss": 0.6377, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.05322265625, |
| "rewards/margins": 0.1181640625, |
| "rewards/rejected": -0.06494140625, |
| "step": 64 |
| }, |
| { |
| "epoch": 0.1925925925925926, |
| "grad_norm": 314.43389892578125, |
| "learning_rate": 1.6142433234421364e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -113.5, |
| "logps/rejected": -193.0, |
| "loss": 0.6267, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.061767578125, |
| "rewards/margins": 0.208984375, |
| "rewards/rejected": -0.1474609375, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.19555555555555557, |
| "grad_norm": 315.067626953125, |
| "learning_rate": 1.6083086053412462e-06, |
| "logits/chosen": -0.376953125, |
| "logits/rejected": -0.400390625, |
| "logps/chosen": -116.0, |
| "logps/rejected": -117.5, |
| "loss": 0.6625, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.0257568359375, |
| "rewards/margins": 0.07958984375, |
| "rewards/rejected": -0.053955078125, |
| "step": 66 |
| }, |
| { |
| "epoch": 0.1985185185185185, |
| "grad_norm": 309.8024597167969, |
| "learning_rate": 1.6023738872403561e-06, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.466796875, |
| "logps/chosen": -122.5, |
| "logps/rejected": -139.0, |
| "loss": 0.6585, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": 0.03759765625, |
| "rewards/margins": 0.09130859375, |
| "rewards/rejected": -0.053955078125, |
| "step": 67 |
| }, |
| { |
| "epoch": 0.20148148148148148, |
| "grad_norm": 303.5541687011719, |
| "learning_rate": 1.5964391691394659e-06, |
| "logits/chosen": -0.376953125, |
| "logits/rejected": -0.4375, |
| "logps/chosen": -124.5, |
| "logps/rejected": -142.0, |
| "loss": 0.6592, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.0289306640625, |
| "rewards/margins": 0.08056640625, |
| "rewards/rejected": -0.109375, |
| "step": 68 |
| }, |
| { |
| "epoch": 0.20444444444444446, |
| "grad_norm": 336.009033203125, |
| "learning_rate": 1.5905044510385756e-06, |
| "logits/chosen": -0.3515625, |
| "logits/rejected": -0.369140625, |
| "logps/chosen": -118.0, |
| "logps/rejected": -132.0, |
| "loss": 0.6651, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.00225830078125, |
| "rewards/margins": 0.07373046875, |
| "rewards/rejected": -0.07568359375, |
| "step": 69 |
| }, |
| { |
| "epoch": 0.2074074074074074, |
| "grad_norm": 295.90008544921875, |
| "learning_rate": 1.5845697329376855e-06, |
| "logits/chosen": -0.41015625, |
| "logits/rejected": -0.43359375, |
| "logps/chosen": -123.5, |
| "logps/rejected": -129.0, |
| "loss": 0.6312, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.04345703125, |
| "rewards/margins": 0.1455078125, |
| "rewards/rejected": -0.1025390625, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.21037037037037037, |
| "grad_norm": 308.5846862792969, |
| "learning_rate": 1.5786350148367953e-06, |
| "logits/chosen": -0.400390625, |
| "logits/rejected": -0.3984375, |
| "logps/chosen": -110.0, |
| "logps/rejected": -160.0, |
| "loss": 0.6731, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.01806640625, |
| "rewards/margins": 0.11767578125, |
| "rewards/rejected": -0.09912109375, |
| "step": 71 |
| }, |
| { |
| "epoch": 0.21333333333333335, |
| "grad_norm": 311.0722961425781, |
| "learning_rate": 1.5727002967359048e-06, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -137.0, |
| "logps/rejected": -214.0, |
| "loss": 0.593, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.064453125, |
| "rewards/margins": 0.294921875, |
| "rewards/rejected": -0.2294921875, |
| "step": 72 |
| }, |
| { |
| "epoch": 0.2162962962962963, |
| "grad_norm": 304.479248046875, |
| "learning_rate": 1.5667655786350147e-06, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -124.5, |
| "logps/rejected": -148.0, |
| "loss": 0.6428, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.022705078125, |
| "rewards/margins": 0.125, |
| "rewards/rejected": -0.1025390625, |
| "step": 73 |
| }, |
| { |
| "epoch": 0.21925925925925926, |
| "grad_norm": 280.2344665527344, |
| "learning_rate": 1.5608308605341244e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.373046875, |
| "logps/chosen": -107.5, |
| "logps/rejected": -119.0, |
| "loss": 0.5997, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.0908203125, |
| "rewards/margins": 0.236328125, |
| "rewards/rejected": -0.1455078125, |
| "step": 74 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 275.5752258300781, |
| "learning_rate": 1.5548961424332344e-06, |
| "logits/chosen": -0.40234375, |
| "logits/rejected": -0.44140625, |
| "logps/chosen": -116.0, |
| "logps/rejected": -110.5, |
| "loss": 0.6214, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.057861328125, |
| "rewards/margins": 0.154296875, |
| "rewards/rejected": -0.09619140625, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.22518518518518518, |
| "grad_norm": 262.7906188964844, |
| "learning_rate": 1.5489614243323441e-06, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.419921875, |
| "logps/chosen": -100.5, |
| "logps/rejected": -137.0, |
| "loss": 0.5972, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.050048828125, |
| "rewards/margins": 0.2392578125, |
| "rewards/rejected": -0.189453125, |
| "step": 76 |
| }, |
| { |
| "epoch": 0.22814814814814816, |
| "grad_norm": 308.56695556640625, |
| "learning_rate": 1.5430267062314539e-06, |
| "logits/chosen": -0.40234375, |
| "logits/rejected": -0.4140625, |
| "logps/chosen": -114.0, |
| "logps/rejected": -199.0, |
| "loss": 0.6227, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.0078125, |
| "rewards/margins": 0.2177734375, |
| "rewards/rejected": -0.2255859375, |
| "step": 77 |
| }, |
| { |
| "epoch": 0.2311111111111111, |
| "grad_norm": 320.17828369140625, |
| "learning_rate": 1.5370919881305638e-06, |
| "logits/chosen": -0.404296875, |
| "logits/rejected": -0.43359375, |
| "logps/chosen": -137.0, |
| "logps/rejected": -103.0, |
| "loss": 0.6665, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.0257568359375, |
| "rewards/margins": 0.07177734375, |
| "rewards/rejected": -0.046142578125, |
| "step": 78 |
| }, |
| { |
| "epoch": 0.23407407407407407, |
| "grad_norm": 325.34100341796875, |
| "learning_rate": 1.5311572700296735e-06, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -138.0, |
| "logps/rejected": -125.5, |
| "loss": 0.6537, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.031982421875, |
| "rewards/margins": 0.11669921875, |
| "rewards/rejected": -0.08447265625, |
| "step": 79 |
| }, |
| { |
| "epoch": 0.23703703703703705, |
| "grad_norm": 341.1995544433594, |
| "learning_rate": 1.5252225519287833e-06, |
| "logits/chosen": -0.380859375, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -112.0, |
| "logps/rejected": -179.0, |
| "loss": 0.6116, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.02734375, |
| "rewards/margins": 0.2109375, |
| "rewards/rejected": -0.18359375, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.24, |
| "grad_norm": 284.32147216796875, |
| "learning_rate": 1.5192878338278932e-06, |
| "logits/chosen": -0.404296875, |
| "logits/rejected": -0.390625, |
| "logps/chosen": -132.0, |
| "logps/rejected": -123.0, |
| "loss": 0.6152, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.109375, |
| "rewards/margins": 0.1845703125, |
| "rewards/rejected": -0.0751953125, |
| "step": 81 |
| }, |
| { |
| "epoch": 0.24296296296296296, |
| "grad_norm": 280.8007507324219, |
| "learning_rate": 1.513353115727003e-06, |
| "logits/chosen": -0.3515625, |
| "logits/rejected": -0.37890625, |
| "logps/chosen": -97.5, |
| "logps/rejected": -107.0, |
| "loss": 0.6219, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.06591796875, |
| "rewards/margins": 0.1650390625, |
| "rewards/rejected": -0.0986328125, |
| "step": 82 |
| }, |
| { |
| "epoch": 0.24592592592592594, |
| "grad_norm": 287.39727783203125, |
| "learning_rate": 1.5074183976261129e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -121.0, |
| "logps/rejected": -130.0, |
| "loss": 0.6078, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0693359375, |
| "rewards/margins": 0.212890625, |
| "rewards/rejected": -0.1435546875, |
| "step": 83 |
| }, |
| { |
| "epoch": 0.24888888888888888, |
| "grad_norm": 293.0041809082031, |
| "learning_rate": 1.5014836795252226e-06, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.412109375, |
| "logps/chosen": -95.5, |
| "logps/rejected": -104.5, |
| "loss": 0.655, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.032470703125, |
| "rewards/margins": 0.0966796875, |
| "rewards/rejected": -0.06396484375, |
| "step": 84 |
| }, |
| { |
| "epoch": 0.2518518518518518, |
| "grad_norm": 365.5411071777344, |
| "learning_rate": 1.4955489614243321e-06, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.41015625, |
| "logps/chosen": -159.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6666, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.0546875, |
| "rewards/margins": 0.1064453125, |
| "rewards/rejected": -0.1611328125, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.2548148148148148, |
| "grad_norm": 327.99237060546875, |
| "learning_rate": 1.489614243323442e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.396484375, |
| "logps/chosen": -126.0, |
| "logps/rejected": -171.0, |
| "loss": 0.6431, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.02978515625, |
| "rewards/margins": 0.1904296875, |
| "rewards/rejected": -0.1611328125, |
| "step": 86 |
| }, |
| { |
| "epoch": 0.2577777777777778, |
| "grad_norm": 299.70306396484375, |
| "learning_rate": 1.4836795252225518e-06, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.490234375, |
| "logps/chosen": -124.5, |
| "logps/rejected": -192.0, |
| "loss": 0.5707, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1396484375, |
| "rewards/margins": 0.373046875, |
| "rewards/rejected": -0.2333984375, |
| "step": 87 |
| }, |
| { |
| "epoch": 0.2607407407407407, |
| "grad_norm": 293.9344787597656, |
| "learning_rate": 1.4777448071216615e-06, |
| "logits/chosen": -0.390625, |
| "logits/rejected": -0.4140625, |
| "logps/chosen": -107.0, |
| "logps/rejected": -125.0, |
| "loss": 0.6241, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.08349609375, |
| "rewards/margins": 0.1875, |
| "rewards/rejected": -0.10400390625, |
| "step": 88 |
| }, |
| { |
| "epoch": 0.2637037037037037, |
| "grad_norm": 295.9560241699219, |
| "learning_rate": 1.4718100890207715e-06, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.400390625, |
| "logps/chosen": -136.0, |
| "logps/rejected": -129.0, |
| "loss": 0.5942, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.08837890625, |
| "rewards/margins": 0.2353515625, |
| "rewards/rejected": -0.146484375, |
| "step": 89 |
| }, |
| { |
| "epoch": 0.26666666666666666, |
| "grad_norm": 341.2995910644531, |
| "learning_rate": 1.4658753709198812e-06, |
| "logits/chosen": -0.40625, |
| "logits/rejected": -0.41796875, |
| "logps/chosen": -112.5, |
| "logps/rejected": -119.5, |
| "loss": 0.638, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.09228515625, |
| "rewards/margins": 0.1435546875, |
| "rewards/rejected": -0.051513671875, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.2696296296296296, |
| "grad_norm": 295.9945373535156, |
| "learning_rate": 1.459940652818991e-06, |
| "logits/chosen": -0.390625, |
| "logits/rejected": -0.40234375, |
| "logps/chosen": -97.5, |
| "logps/rejected": -122.5, |
| "loss": 0.6188, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.043701171875, |
| "rewards/margins": 0.181640625, |
| "rewards/rejected": -0.1376953125, |
| "step": 91 |
| }, |
| { |
| "epoch": 0.2725925925925926, |
| "grad_norm": 335.8014831542969, |
| "learning_rate": 1.4540059347181009e-06, |
| "logits/chosen": -0.423828125, |
| "logits/rejected": -0.404296875, |
| "logps/chosen": -141.0, |
| "logps/rejected": -124.5, |
| "loss": 0.6823, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.015625, |
| "rewards/margins": 0.10546875, |
| "rewards/rejected": -0.12109375, |
| "step": 92 |
| }, |
| { |
| "epoch": 0.27555555555555555, |
| "grad_norm": 316.69366455078125, |
| "learning_rate": 1.4480712166172106e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -132.0, |
| "logps/rejected": -121.0, |
| "loss": 0.5893, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.09326171875, |
| "rewards/margins": 0.251953125, |
| "rewards/rejected": -0.158203125, |
| "step": 93 |
| }, |
| { |
| "epoch": 0.2785185185185185, |
| "grad_norm": 287.4927062988281, |
| "learning_rate": 1.4421364985163206e-06, |
| "logits/chosen": -0.375, |
| "logits/rejected": -0.388671875, |
| "logps/chosen": -98.0, |
| "logps/rejected": -114.5, |
| "loss": 0.6371, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.08203125, |
| "rewards/margins": 0.142578125, |
| "rewards/rejected": -0.060302734375, |
| "step": 94 |
| }, |
| { |
| "epoch": 0.2814814814814815, |
| "grad_norm": 279.02618408203125, |
| "learning_rate": 1.4362017804154303e-06, |
| "logits/chosen": -0.3828125, |
| "logits/rejected": -0.416015625, |
| "logps/chosen": -104.5, |
| "logps/rejected": -145.0, |
| "loss": 0.5956, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.07568359375, |
| "rewards/margins": 0.2890625, |
| "rewards/rejected": -0.2138671875, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.28444444444444444, |
| "grad_norm": 387.8583068847656, |
| "learning_rate": 1.43026706231454e-06, |
| "logits/chosen": -0.404296875, |
| "logits/rejected": -0.396484375, |
| "logps/chosen": -108.5, |
| "logps/rejected": -106.0, |
| "loss": 0.6595, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.0133056640625, |
| "rewards/margins": 0.09130859375, |
| "rewards/rejected": -0.078125, |
| "step": 96 |
| }, |
| { |
| "epoch": 0.2874074074074074, |
| "grad_norm": 362.9851379394531, |
| "learning_rate": 1.42433234421365e-06, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.4375, |
| "logps/chosen": -139.0, |
| "logps/rejected": -120.0, |
| "loss": 0.673, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.020263671875, |
| "rewards/margins": 0.0751953125, |
| "rewards/rejected": -0.0546875, |
| "step": 97 |
| }, |
| { |
| "epoch": 0.2903703703703704, |
| "grad_norm": 308.3312072753906, |
| "learning_rate": 1.4183976261127595e-06, |
| "logits/chosen": -0.390625, |
| "logits/rejected": -0.439453125, |
| "logps/chosen": -113.0, |
| "logps/rejected": -119.5, |
| "loss": 0.6538, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.013916015625, |
| "rewards/margins": 0.126953125, |
| "rewards/rejected": -0.11328125, |
| "step": 98 |
| }, |
| { |
| "epoch": 0.29333333333333333, |
| "grad_norm": 360.1666259765625, |
| "learning_rate": 1.4124629080118692e-06, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -131.0, |
| "logps/rejected": -124.0, |
| "loss": 0.6746, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": -0.06982421875, |
| "rewards/margins": 0.053466796875, |
| "rewards/rejected": -0.12353515625, |
| "step": 99 |
| }, |
| { |
| "epoch": 0.2962962962962963, |
| "grad_norm": 285.48065185546875, |
| "learning_rate": 1.4065281899109791e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.4140625, |
| "logps/chosen": -101.5, |
| "logps/rejected": -109.0, |
| "loss": 0.5988, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.039794921875, |
| "rewards/margins": 0.228515625, |
| "rewards/rejected": -0.189453125, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.2992592592592593, |
| "grad_norm": 272.54583740234375, |
| "learning_rate": 1.4005934718100889e-06, |
| "logits/chosen": -0.3828125, |
| "logits/rejected": -0.37109375, |
| "logps/chosen": -94.0, |
| "logps/rejected": -100.5, |
| "loss": 0.6157, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.06884765625, |
| "rewards/margins": 0.1865234375, |
| "rewards/rejected": -0.1181640625, |
| "step": 101 |
| }, |
| { |
| "epoch": 0.3022222222222222, |
| "grad_norm": 261.4800109863281, |
| "learning_rate": 1.3946587537091988e-06, |
| "logits/chosen": -0.3671875, |
| "logits/rejected": -0.396484375, |
| "logps/chosen": -90.0, |
| "logps/rejected": -101.0, |
| "loss": 0.6438, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.018798828125, |
| "rewards/margins": 0.12353515625, |
| "rewards/rejected": -0.10498046875, |
| "step": 102 |
| }, |
| { |
| "epoch": 0.30518518518518517, |
| "grad_norm": 299.9373779296875, |
| "learning_rate": 1.3887240356083086e-06, |
| "logits/chosen": -0.443359375, |
| "logits/rejected": -0.435546875, |
| "logps/chosen": -109.0, |
| "logps/rejected": -102.0, |
| "loss": 0.6418, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.0361328125, |
| "rewards/margins": 0.1298828125, |
| "rewards/rejected": -0.09375, |
| "step": 103 |
| }, |
| { |
| "epoch": 0.30814814814814817, |
| "grad_norm": 315.3984069824219, |
| "learning_rate": 1.3827893175074183e-06, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.439453125, |
| "logps/chosen": -112.0, |
| "logps/rejected": -118.0, |
| "loss": 0.6638, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.024169921875, |
| "rewards/margins": 0.09716796875, |
| "rewards/rejected": -0.12109375, |
| "step": 104 |
| }, |
| { |
| "epoch": 0.3111111111111111, |
| "grad_norm": 309.1874084472656, |
| "learning_rate": 1.3768545994065282e-06, |
| "logits/chosen": -0.41015625, |
| "logits/rejected": -0.412109375, |
| "logps/chosen": -104.0, |
| "logps/rejected": -118.5, |
| "loss": 0.6292, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.038330078125, |
| "rewards/margins": 0.15234375, |
| "rewards/rejected": -0.1142578125, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.31407407407407406, |
| "grad_norm": 306.6035461425781, |
| "learning_rate": 1.370919881305638e-06, |
| "logits/chosen": -0.39453125, |
| "logits/rejected": -0.423828125, |
| "logps/chosen": -121.5, |
| "logps/rejected": -111.5, |
| "loss": 0.636, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.02294921875, |
| "rewards/margins": 0.1611328125, |
| "rewards/rejected": -0.1845703125, |
| "step": 106 |
| }, |
| { |
| "epoch": 0.31703703703703706, |
| "grad_norm": 267.6772766113281, |
| "learning_rate": 1.3649851632047477e-06, |
| "logits/chosen": -0.384765625, |
| "logits/rejected": -0.412109375, |
| "logps/chosen": -91.0, |
| "logps/rejected": -116.5, |
| "loss": 0.5857, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.0849609375, |
| "rewards/margins": 0.2734375, |
| "rewards/rejected": -0.1884765625, |
| "step": 107 |
| }, |
| { |
| "epoch": 0.32, |
| "grad_norm": 323.27239990234375, |
| "learning_rate": 1.3590504451038576e-06, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -98.0, |
| "logps/rejected": -118.5, |
| "loss": 0.6342, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.08056640625, |
| "rewards/margins": 0.1572265625, |
| "rewards/rejected": -0.0771484375, |
| "step": 108 |
| }, |
| { |
| "epoch": 0.32296296296296295, |
| "grad_norm": 303.9071044921875, |
| "learning_rate": 1.3531157270029674e-06, |
| "logits/chosen": -0.4609375, |
| "logits/rejected": -0.466796875, |
| "logps/chosen": -114.0, |
| "logps/rejected": -108.5, |
| "loss": 0.6272, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.028076171875, |
| "rewards/margins": 0.1650390625, |
| "rewards/rejected": -0.13671875, |
| "step": 109 |
| }, |
| { |
| "epoch": 0.32592592592592595, |
| "grad_norm": 309.7470397949219, |
| "learning_rate": 1.347181008902077e-06, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -132.0, |
| "logps/rejected": -168.0, |
| "loss": 0.5938, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.036865234375, |
| "rewards/margins": 0.30859375, |
| "rewards/rejected": -0.2734375, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.3288888888888889, |
| "grad_norm": 276.505615234375, |
| "learning_rate": 1.3412462908011868e-06, |
| "logits/chosen": -0.39453125, |
| "logits/rejected": -0.40234375, |
| "logps/chosen": -97.5, |
| "logps/rejected": -106.0, |
| "loss": 0.6151, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.1220703125, |
| "rewards/margins": 0.1875, |
| "rewards/rejected": -0.06591796875, |
| "step": 111 |
| }, |
| { |
| "epoch": 0.33185185185185184, |
| "grad_norm": 317.2270202636719, |
| "learning_rate": 1.3353115727002966e-06, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.439453125, |
| "logps/chosen": -117.0, |
| "logps/rejected": -106.5, |
| "loss": 0.6424, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.046142578125, |
| "rewards/margins": 0.1298828125, |
| "rewards/rejected": -0.08349609375, |
| "step": 112 |
| }, |
| { |
| "epoch": 0.3348148148148148, |
| "grad_norm": 283.0464172363281, |
| "learning_rate": 1.3293768545994065e-06, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.421875, |
| "logps/chosen": -88.5, |
| "logps/rejected": -97.5, |
| "loss": 0.6356, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0947265625, |
| "rewards/margins": 0.14453125, |
| "rewards/rejected": -0.050048828125, |
| "step": 113 |
| }, |
| { |
| "epoch": 0.3377777777777778, |
| "grad_norm": 280.1183166503906, |
| "learning_rate": 1.3234421364985162e-06, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.466796875, |
| "logps/chosen": -99.0, |
| "logps/rejected": -152.0, |
| "loss": 0.5299, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": 0.12353515625, |
| "rewards/margins": 0.404296875, |
| "rewards/rejected": -0.28125, |
| "step": 114 |
| }, |
| { |
| "epoch": 0.34074074074074073, |
| "grad_norm": 280.7152404785156, |
| "learning_rate": 1.317507418397626e-06, |
| "logits/chosen": -0.39453125, |
| "logits/rejected": -0.412109375, |
| "logps/chosen": -102.5, |
| "logps/rejected": -117.5, |
| "loss": 0.5846, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.10546875, |
| "rewards/margins": 0.27734375, |
| "rewards/rejected": -0.1728515625, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.3437037037037037, |
| "grad_norm": 311.7525329589844, |
| "learning_rate": 1.311572700296736e-06, |
| "logits/chosen": -0.384765625, |
| "logits/rejected": -0.40625, |
| "logps/chosen": -98.5, |
| "logps/rejected": -100.5, |
| "loss": 0.6609, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.038330078125, |
| "rewards/margins": 0.08837890625, |
| "rewards/rejected": -0.126953125, |
| "step": 116 |
| }, |
| { |
| "epoch": 0.3466666666666667, |
| "grad_norm": 299.9917297363281, |
| "learning_rate": 1.3056379821958456e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.42578125, |
| "logps/chosen": -97.5, |
| "logps/rejected": -112.0, |
| "loss": 0.6263, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.056884765625, |
| "rewards/margins": 0.166015625, |
| "rewards/rejected": -0.109375, |
| "step": 117 |
| }, |
| { |
| "epoch": 0.3496296296296296, |
| "grad_norm": 309.1261901855469, |
| "learning_rate": 1.2997032640949554e-06, |
| "logits/chosen": -0.42578125, |
| "logits/rejected": -0.416015625, |
| "logps/chosen": -113.5, |
| "logps/rejected": -113.5, |
| "loss": 0.621, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.052490234375, |
| "rewards/margins": 0.1826171875, |
| "rewards/rejected": -0.130859375, |
| "step": 118 |
| }, |
| { |
| "epoch": 0.35259259259259257, |
| "grad_norm": 272.764404296875, |
| "learning_rate": 1.2937685459940653e-06, |
| "logits/chosen": -0.384765625, |
| "logits/rejected": -0.431640625, |
| "logps/chosen": -105.0, |
| "logps/rejected": -113.5, |
| "loss": 0.6042, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.060302734375, |
| "rewards/margins": 0.2265625, |
| "rewards/rejected": -0.166015625, |
| "step": 119 |
| }, |
| { |
| "epoch": 0.35555555555555557, |
| "grad_norm": 268.743408203125, |
| "learning_rate": 1.287833827893175e-06, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.435546875, |
| "logps/chosen": -109.5, |
| "logps/rejected": -209.0, |
| "loss": 0.5652, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.0908203125, |
| "rewards/margins": 0.39453125, |
| "rewards/rejected": -0.3046875, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.3585185185185185, |
| "grad_norm": 328.0078125, |
| "learning_rate": 1.281899109792285e-06, |
| "logits/chosen": -0.40234375, |
| "logits/rejected": -0.412109375, |
| "logps/chosen": -110.0, |
| "logps/rejected": -120.5, |
| "loss": 0.6072, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.031494140625, |
| "rewards/margins": 0.2177734375, |
| "rewards/rejected": -0.1865234375, |
| "step": 121 |
| }, |
| { |
| "epoch": 0.36148148148148146, |
| "grad_norm": 291.20794677734375, |
| "learning_rate": 1.2759643916913947e-06, |
| "logits/chosen": -0.357421875, |
| "logits/rejected": -0.36328125, |
| "logps/chosen": -96.0, |
| "logps/rejected": -93.5, |
| "loss": 0.6382, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.047607421875, |
| "rewards/margins": 0.1435546875, |
| "rewards/rejected": -0.09619140625, |
| "step": 122 |
| }, |
| { |
| "epoch": 0.36444444444444446, |
| "grad_norm": 311.23870849609375, |
| "learning_rate": 1.2700296735905044e-06, |
| "logits/chosen": -0.37890625, |
| "logits/rejected": -0.380859375, |
| "logps/chosen": -101.5, |
| "logps/rejected": -191.0, |
| "loss": 0.6198, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.03271484375, |
| "rewards/margins": 0.349609375, |
| "rewards/rejected": -0.31640625, |
| "step": 123 |
| }, |
| { |
| "epoch": 0.3674074074074074, |
| "grad_norm": 327.720703125, |
| "learning_rate": 1.2640949554896142e-06, |
| "logits/chosen": -0.3984375, |
| "logits/rejected": -0.443359375, |
| "logps/chosen": -111.0, |
| "logps/rejected": -133.0, |
| "loss": 0.6528, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.01953125, |
| "rewards/margins": 0.1513671875, |
| "rewards/rejected": -0.1318359375, |
| "step": 124 |
| }, |
| { |
| "epoch": 0.37037037037037035, |
| "grad_norm": 301.95654296875, |
| "learning_rate": 1.258160237388724e-06, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.435546875, |
| "logps/chosen": -121.0, |
| "logps/rejected": -118.5, |
| "loss": 0.5869, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0732421875, |
| "rewards/margins": 0.28125, |
| "rewards/rejected": -0.20703125, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.37333333333333335, |
| "grad_norm": 349.1401672363281, |
| "learning_rate": 1.2522255192878336e-06, |
| "logits/chosen": -0.412109375, |
| "logits/rejected": -0.41015625, |
| "logps/chosen": -120.5, |
| "logps/rejected": -104.5, |
| "loss": 0.6876, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.083984375, |
| "rewards/margins": 0.052978515625, |
| "rewards/rejected": -0.13671875, |
| "step": 126 |
| }, |
| { |
| "epoch": 0.3762962962962963, |
| "grad_norm": 319.4087219238281, |
| "learning_rate": 1.2462908011869436e-06, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.49609375, |
| "logps/chosen": -115.5, |
| "logps/rejected": -129.0, |
| "loss": 0.6176, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.043701171875, |
| "rewards/margins": 0.2080078125, |
| "rewards/rejected": -0.1640625, |
| "step": 127 |
| }, |
| { |
| "epoch": 0.37925925925925924, |
| "grad_norm": 325.74456787109375, |
| "learning_rate": 1.2403560830860533e-06, |
| "logits/chosen": -0.412109375, |
| "logits/rejected": -0.4140625, |
| "logps/chosen": -109.5, |
| "logps/rejected": -177.0, |
| "loss": 0.6262, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.0322265625, |
| "rewards/margins": 0.2578125, |
| "rewards/rejected": -0.2255859375, |
| "step": 128 |
| }, |
| { |
| "epoch": 0.38222222222222224, |
| "grad_norm": 288.1813659667969, |
| "learning_rate": 1.234421364985163e-06, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -106.0, |
| "logps/rejected": -96.5, |
| "loss": 0.6172, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.0108642578125, |
| "rewards/margins": 0.208984375, |
| "rewards/rejected": -0.197265625, |
| "step": 129 |
| }, |
| { |
| "epoch": 0.3851851851851852, |
| "grad_norm": 291.5497131347656, |
| "learning_rate": 1.228486646884273e-06, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -110.5, |
| "logps/rejected": -158.0, |
| "loss": 0.5666, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.03662109375, |
| "rewards/margins": 0.36328125, |
| "rewards/rejected": -0.326171875, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.38814814814814813, |
| "grad_norm": 298.1041564941406, |
| "learning_rate": 1.2225519287833827e-06, |
| "logits/chosen": -0.404296875, |
| "logits/rejected": -0.419921875, |
| "logps/chosen": -115.0, |
| "logps/rejected": -133.0, |
| "loss": 0.6319, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.053955078125, |
| "rewards/margins": 0.162109375, |
| "rewards/rejected": -0.10791015625, |
| "step": 131 |
| }, |
| { |
| "epoch": 0.39111111111111113, |
| "grad_norm": 291.80596923828125, |
| "learning_rate": 1.2166172106824927e-06, |
| "logits/chosen": -0.400390625, |
| "logits/rejected": -0.3984375, |
| "logps/chosen": -100.5, |
| "logps/rejected": -109.0, |
| "loss": 0.6202, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0390625, |
| "rewards/margins": 0.171875, |
| "rewards/rejected": -0.1328125, |
| "step": 132 |
| }, |
| { |
| "epoch": 0.3940740740740741, |
| "grad_norm": 291.0092468261719, |
| "learning_rate": 1.2106824925816024e-06, |
| "logits/chosen": -0.3828125, |
| "logits/rejected": -0.421875, |
| "logps/chosen": -97.5, |
| "logps/rejected": -105.5, |
| "loss": 0.5925, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.03662109375, |
| "rewards/margins": 0.2578125, |
| "rewards/rejected": -0.2216796875, |
| "step": 133 |
| }, |
| { |
| "epoch": 0.397037037037037, |
| "grad_norm": 341.5547180175781, |
| "learning_rate": 1.2047477744807121e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.416015625, |
| "logps/chosen": -109.5, |
| "logps/rejected": -112.0, |
| "loss": 0.6822, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.025146484375, |
| "rewards/margins": 0.059326171875, |
| "rewards/rejected": -0.08447265625, |
| "step": 134 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 392.70574951171875, |
| "learning_rate": 1.198813056379822e-06, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.439453125, |
| "logps/chosen": -127.0, |
| "logps/rejected": -137.0, |
| "loss": 0.6592, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.046875, |
| "rewards/margins": 0.15625, |
| "rewards/rejected": -0.203125, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.40296296296296297, |
| "grad_norm": 306.712646484375, |
| "learning_rate": 1.1928783382789318e-06, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -107.0, |
| "logps/rejected": -115.5, |
| "loss": 0.6097, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.016357421875, |
| "rewards/margins": 0.2001953125, |
| "rewards/rejected": -0.2158203125, |
| "step": 136 |
| }, |
| { |
| "epoch": 0.4059259259259259, |
| "grad_norm": 303.9560546875, |
| "learning_rate": 1.1869436201780413e-06, |
| "logits/chosen": -0.37890625, |
| "logits/rejected": -0.423828125, |
| "logps/chosen": -101.5, |
| "logps/rejected": -131.0, |
| "loss": 0.6343, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.030517578125, |
| "rewards/margins": 0.15234375, |
| "rewards/rejected": -0.1826171875, |
| "step": 137 |
| }, |
| { |
| "epoch": 0.4088888888888889, |
| "grad_norm": 360.12445068359375, |
| "learning_rate": 1.1810089020771513e-06, |
| "logits/chosen": -0.423828125, |
| "logits/rejected": -0.412109375, |
| "logps/chosen": -122.0, |
| "logps/rejected": -104.5, |
| "loss": 0.6484, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.0281982421875, |
| "rewards/margins": 0.1591796875, |
| "rewards/rejected": -0.1865234375, |
| "step": 138 |
| }, |
| { |
| "epoch": 0.41185185185185186, |
| "grad_norm": 281.4104919433594, |
| "learning_rate": 1.175074183976261e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.40625, |
| "logps/chosen": -109.5, |
| "logps/rejected": -115.5, |
| "loss": 0.6265, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.06103515625, |
| "rewards/margins": 0.1845703125, |
| "rewards/rejected": -0.12353515625, |
| "step": 139 |
| }, |
| { |
| "epoch": 0.4148148148148148, |
| "grad_norm": 319.68548583984375, |
| "learning_rate": 1.169139465875371e-06, |
| "logits/chosen": -0.40234375, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -118.0, |
| "logps/rejected": -120.0, |
| "loss": 0.604, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.02490234375, |
| "rewards/margins": 0.2294921875, |
| "rewards/rejected": -0.205078125, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.4177777777777778, |
| "grad_norm": 285.38104248046875, |
| "learning_rate": 1.1632047477744807e-06, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -102.0, |
| "logps/rejected": -134.0, |
| "loss": 0.5652, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.05078125, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.296875, |
| "step": 141 |
| }, |
| { |
| "epoch": 0.42074074074074075, |
| "grad_norm": 295.56475830078125, |
| "learning_rate": 1.1572700296735904e-06, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.40625, |
| "logps/chosen": -100.5, |
| "logps/rejected": -97.5, |
| "loss": 0.6649, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.02197265625, |
| "rewards/margins": 0.083984375, |
| "rewards/rejected": -0.06201171875, |
| "step": 142 |
| }, |
| { |
| "epoch": 0.4237037037037037, |
| "grad_norm": 323.8175354003906, |
| "learning_rate": 1.1513353115727003e-06, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -136.0, |
| "logps/rejected": -125.0, |
| "loss": 0.6222, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.061767578125, |
| "rewards/margins": 0.2041015625, |
| "rewards/rejected": -0.142578125, |
| "step": 143 |
| }, |
| { |
| "epoch": 0.4266666666666667, |
| "grad_norm": 314.3506164550781, |
| "learning_rate": 1.14540059347181e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.42578125, |
| "logps/chosen": -115.5, |
| "logps/rejected": -104.0, |
| "loss": 0.6947, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.044677734375, |
| "rewards/margins": 0.047607421875, |
| "rewards/rejected": -0.09228515625, |
| "step": 144 |
| }, |
| { |
| "epoch": 0.42962962962962964, |
| "grad_norm": 324.5817565917969, |
| "learning_rate": 1.1394658753709198e-06, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.345703125, |
| "logps/chosen": -121.0, |
| "logps/rejected": -150.0, |
| "loss": 0.6394, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.039794921875, |
| "rewards/margins": 0.1982421875, |
| "rewards/rejected": -0.158203125, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.4325925925925926, |
| "grad_norm": 314.7122497558594, |
| "learning_rate": 1.1335311572700297e-06, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.41015625, |
| "logps/chosen": -105.5, |
| "logps/rejected": -96.0, |
| "loss": 0.6624, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.034423828125, |
| "rewards/margins": 0.09765625, |
| "rewards/rejected": -0.06298828125, |
| "step": 146 |
| }, |
| { |
| "epoch": 0.43555555555555553, |
| "grad_norm": 275.6307678222656, |
| "learning_rate": 1.1275964391691395e-06, |
| "logits/chosen": -0.423828125, |
| "logits/rejected": -0.39453125, |
| "logps/chosen": -93.0, |
| "logps/rejected": -100.5, |
| "loss": 0.5839, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.11083984375, |
| "rewards/margins": 0.271484375, |
| "rewards/rejected": -0.16015625, |
| "step": 147 |
| }, |
| { |
| "epoch": 0.43851851851851853, |
| "grad_norm": 279.8615417480469, |
| "learning_rate": 1.1216617210682492e-06, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.349609375, |
| "logps/chosen": -97.5, |
| "logps/rejected": -237.0, |
| "loss": 0.523, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": 0.1181640625, |
| "rewards/margins": 0.609375, |
| "rewards/rejected": -0.490234375, |
| "step": 148 |
| }, |
| { |
| "epoch": 0.4414814814814815, |
| "grad_norm": 311.0171813964844, |
| "learning_rate": 1.1157270029673591e-06, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.400390625, |
| "logps/chosen": -114.0, |
| "logps/rejected": -134.0, |
| "loss": 0.6284, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.04443359375, |
| "rewards/margins": 0.1884765625, |
| "rewards/rejected": -0.14453125, |
| "step": 149 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 301.62060546875, |
| "learning_rate": 1.1097922848664687e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.392578125, |
| "logps/chosen": -110.5, |
| "logps/rejected": -112.0, |
| "loss": 0.6028, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.12109375, |
| "rewards/margins": 0.2314453125, |
| "rewards/rejected": -0.1103515625, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.4474074074074074, |
| "grad_norm": 291.1073303222656, |
| "learning_rate": 1.1038575667655786e-06, |
| "logits/chosen": -0.400390625, |
| "logits/rejected": -0.373046875, |
| "logps/chosen": -91.5, |
| "logps/rejected": -101.5, |
| "loss": 0.6195, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.11865234375, |
| "rewards/margins": 0.18359375, |
| "rewards/rejected": -0.06396484375, |
| "step": 151 |
| }, |
| { |
| "epoch": 0.45037037037037037, |
| "grad_norm": 299.3499450683594, |
| "learning_rate": 1.0979228486646883e-06, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.453125, |
| "logps/chosen": -84.5, |
| "logps/rejected": -128.0, |
| "loss": 0.5974, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.0703125, |
| "rewards/margins": 0.2451171875, |
| "rewards/rejected": -0.1748046875, |
| "step": 152 |
| }, |
| { |
| "epoch": 0.4533333333333333, |
| "grad_norm": 306.00897216796875, |
| "learning_rate": 1.091988130563798e-06, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.451171875, |
| "logps/chosen": -109.5, |
| "logps/rejected": -103.0, |
| "loss": 0.6441, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.05859375, |
| "rewards/margins": 0.115234375, |
| "rewards/rejected": -0.056640625, |
| "step": 153 |
| }, |
| { |
| "epoch": 0.4562962962962963, |
| "grad_norm": 285.1234436035156, |
| "learning_rate": 1.086053412462908e-06, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.3828125, |
| "logps/chosen": -96.5, |
| "logps/rejected": -147.0, |
| "loss": 0.5937, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.13671875, |
| "rewards/margins": 0.33984375, |
| "rewards/rejected": -0.2041015625, |
| "step": 154 |
| }, |
| { |
| "epoch": 0.45925925925925926, |
| "grad_norm": 333.5130920410156, |
| "learning_rate": 1.0801186943620177e-06, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.41796875, |
| "logps/chosen": -115.5, |
| "logps/rejected": -108.5, |
| "loss": 0.6072, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.0771484375, |
| "rewards/margins": 0.21875, |
| "rewards/rejected": -0.1416015625, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.4622222222222222, |
| "grad_norm": 327.6874084472656, |
| "learning_rate": 1.0741839762611275e-06, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -117.0, |
| "logps/rejected": -145.0, |
| "loss": 0.5776, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.0849609375, |
| "rewards/margins": 0.310546875, |
| "rewards/rejected": -0.2255859375, |
| "step": 156 |
| }, |
| { |
| "epoch": 0.4651851851851852, |
| "grad_norm": 274.0838928222656, |
| "learning_rate": 1.0682492581602374e-06, |
| "logits/chosen": -0.447265625, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -94.5, |
| "logps/rejected": -164.0, |
| "loss": 0.5773, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.1064453125, |
| "rewards/margins": 0.38671875, |
| "rewards/rejected": -0.279296875, |
| "step": 157 |
| }, |
| { |
| "epoch": 0.46814814814814815, |
| "grad_norm": 309.1082458496094, |
| "learning_rate": 1.0623145400593471e-06, |
| "logits/chosen": -0.46484375, |
| "logits/rejected": -0.51953125, |
| "logps/chosen": -107.0, |
| "logps/rejected": -236.0, |
| "loss": 0.5369, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.0849609375, |
| "rewards/margins": 0.53125, |
| "rewards/rejected": -0.443359375, |
| "step": 158 |
| }, |
| { |
| "epoch": 0.4711111111111111, |
| "grad_norm": 328.0971374511719, |
| "learning_rate": 1.056379821958457e-06, |
| "logits/chosen": -0.392578125, |
| "logits/rejected": -0.392578125, |
| "logps/chosen": -122.0, |
| "logps/rejected": -205.0, |
| "loss": 0.5574, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.1328125, |
| "rewards/margins": 0.380859375, |
| "rewards/rejected": -0.248046875, |
| "step": 159 |
| }, |
| { |
| "epoch": 0.4740740740740741, |
| "grad_norm": 314.67706298828125, |
| "learning_rate": 1.0504451038575668e-06, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.40625, |
| "logps/chosen": -131.0, |
| "logps/rejected": -110.0, |
| "loss": 0.6024, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.07470703125, |
| "rewards/margins": 0.23828125, |
| "rewards/rejected": -0.1630859375, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.47703703703703704, |
| "grad_norm": 365.208984375, |
| "learning_rate": 1.0445103857566765e-06, |
| "logits/chosen": -0.4609375, |
| "logits/rejected": -0.4140625, |
| "logps/chosen": -142.0, |
| "logps/rejected": -162.0, |
| "loss": 0.5747, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.053955078125, |
| "rewards/margins": 0.310546875, |
| "rewards/rejected": -0.2578125, |
| "step": 161 |
| }, |
| { |
| "epoch": 0.48, |
| "grad_norm": 314.8697814941406, |
| "learning_rate": 1.0385756676557865e-06, |
| "logits/chosen": -0.388671875, |
| "logits/rejected": -0.404296875, |
| "logps/chosen": -96.0, |
| "logps/rejected": -108.0, |
| "loss": 0.6699, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.04931640625, |
| "rewards/margins": 0.0869140625, |
| "rewards/rejected": -0.03759765625, |
| "step": 162 |
| }, |
| { |
| "epoch": 0.482962962962963, |
| "grad_norm": 297.8361511230469, |
| "learning_rate": 1.032640949554896e-06, |
| "logits/chosen": -0.435546875, |
| "logits/rejected": -0.423828125, |
| "logps/chosen": -114.0, |
| "logps/rejected": -138.0, |
| "loss": 0.6028, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.053955078125, |
| "rewards/margins": 0.25390625, |
| "rewards/rejected": -0.19921875, |
| "step": 163 |
| }, |
| { |
| "epoch": 0.48592592592592593, |
| "grad_norm": 316.10369873046875, |
| "learning_rate": 1.0267062314540057e-06, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -100.0, |
| "logps/rejected": -119.0, |
| "loss": 0.6105, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.0830078125, |
| "rewards/margins": 0.2353515625, |
| "rewards/rejected": -0.1533203125, |
| "step": 164 |
| }, |
| { |
| "epoch": 0.4888888888888889, |
| "grad_norm": 282.8133239746094, |
| "learning_rate": 1.0207715133531157e-06, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -97.0, |
| "logps/rejected": -152.0, |
| "loss": 0.5733, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.072265625, |
| "rewards/margins": 0.421875, |
| "rewards/rejected": -0.349609375, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.4918518518518519, |
| "grad_norm": 299.11846923828125, |
| "learning_rate": 1.0148367952522254e-06, |
| "logits/chosen": -0.40625, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -104.5, |
| "logps/rejected": -185.0, |
| "loss": 0.6322, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.00164794921875, |
| "rewards/margins": 0.248046875, |
| "rewards/rejected": -0.2470703125, |
| "step": 166 |
| }, |
| { |
| "epoch": 0.4948148148148148, |
| "grad_norm": 327.81646728515625, |
| "learning_rate": 1.0089020771513351e-06, |
| "logits/chosen": -0.45703125, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -111.5, |
| "logps/rejected": -143.0, |
| "loss": 0.5774, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.052490234375, |
| "rewards/margins": 0.30859375, |
| "rewards/rejected": -0.255859375, |
| "step": 167 |
| }, |
| { |
| "epoch": 0.49777777777777776, |
| "grad_norm": 342.8192443847656, |
| "learning_rate": 1.002967359050445e-06, |
| "logits/chosen": -0.3671875, |
| "logits/rejected": -0.39453125, |
| "logps/chosen": -102.5, |
| "logps/rejected": -111.5, |
| "loss": 0.5883, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.057861328125, |
| "rewards/margins": 0.2451171875, |
| "rewards/rejected": -0.1875, |
| "step": 168 |
| }, |
| { |
| "epoch": 0.5007407407407407, |
| "grad_norm": 334.7403259277344, |
| "learning_rate": 9.970326409495548e-07, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -124.5, |
| "logps/rejected": -124.5, |
| "loss": 0.5725, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.0218505859375, |
| "rewards/margins": 0.3125, |
| "rewards/rejected": -0.2890625, |
| "step": 169 |
| }, |
| { |
| "epoch": 0.5037037037037037, |
| "grad_norm": 285.1996154785156, |
| "learning_rate": 9.910979228486648e-07, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -107.0, |
| "logps/rejected": -109.5, |
| "loss": 0.6043, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.07275390625, |
| "rewards/margins": 0.21875, |
| "rewards/rejected": -0.1455078125, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.5066666666666667, |
| "grad_norm": 326.8195495605469, |
| "learning_rate": 9.851632047477745e-07, |
| "logits/chosen": -0.408203125, |
| "logits/rejected": -0.43359375, |
| "logps/chosen": -105.0, |
| "logps/rejected": -107.5, |
| "loss": 0.6083, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.0888671875, |
| "rewards/margins": 0.2080078125, |
| "rewards/rejected": -0.11865234375, |
| "step": 171 |
| }, |
| { |
| "epoch": 0.5096296296296297, |
| "grad_norm": 390.0325012207031, |
| "learning_rate": 9.792284866468842e-07, |
| "logits/chosen": -0.41015625, |
| "logits/rejected": -0.3984375, |
| "logps/chosen": -133.0, |
| "logps/rejected": -104.5, |
| "loss": 0.6678, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.00787353515625, |
| "rewards/margins": 0.07666015625, |
| "rewards/rejected": -0.08447265625, |
| "step": 172 |
| }, |
| { |
| "epoch": 0.5125925925925926, |
| "grad_norm": 285.420166015625, |
| "learning_rate": 9.73293768545994e-07, |
| "logits/chosen": -0.4609375, |
| "logits/rejected": -0.455078125, |
| "logps/chosen": -101.5, |
| "logps/rejected": -104.5, |
| "loss": 0.5923, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.1416015625, |
| "rewards/margins": 0.2470703125, |
| "rewards/rejected": -0.1044921875, |
| "step": 173 |
| }, |
| { |
| "epoch": 0.5155555555555555, |
| "grad_norm": 377.0559387207031, |
| "learning_rate": 9.67359050445104e-07, |
| "logits/chosen": -0.412109375, |
| "logits/rejected": -0.388671875, |
| "logps/chosen": -104.5, |
| "logps/rejected": -105.0, |
| "loss": 0.6486, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.1455078125, |
| "rewards/margins": 0.12060546875, |
| "rewards/rejected": 0.0250244140625, |
| "step": 174 |
| }, |
| { |
| "epoch": 0.5185185185185185, |
| "grad_norm": 348.386962890625, |
| "learning_rate": 9.614243323442136e-07, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -116.0, |
| "logps/rejected": -126.5, |
| "loss": 0.6291, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.05859375, |
| "rewards/margins": 0.19140625, |
| "rewards/rejected": -0.1328125, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.5214814814814814, |
| "grad_norm": 308.8468017578125, |
| "learning_rate": 9.554896142433234e-07, |
| "logits/chosen": -0.435546875, |
| "logits/rejected": -0.419921875, |
| "logps/chosen": -110.5, |
| "logps/rejected": -105.0, |
| "loss": 0.5789, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1279296875, |
| "rewards/margins": 0.275390625, |
| "rewards/rejected": -0.1474609375, |
| "step": 176 |
| }, |
| { |
| "epoch": 0.5244444444444445, |
| "grad_norm": 309.856201171875, |
| "learning_rate": 9.495548961424332e-07, |
| "logits/chosen": -0.380859375, |
| "logits/rejected": -0.44140625, |
| "logps/chosen": -98.5, |
| "logps/rejected": -133.0, |
| "loss": 0.6226, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.0986328125, |
| "rewards/margins": 0.2265625, |
| "rewards/rejected": -0.1279296875, |
| "step": 177 |
| }, |
| { |
| "epoch": 0.5274074074074074, |
| "grad_norm": 296.9087219238281, |
| "learning_rate": 9.43620178041543e-07, |
| "logits/chosen": -0.3984375, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -103.5, |
| "logps/rejected": -111.0, |
| "loss": 0.5776, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.123046875, |
| "rewards/margins": 0.29296875, |
| "rewards/rejected": -0.169921875, |
| "step": 178 |
| }, |
| { |
| "epoch": 0.5303703703703704, |
| "grad_norm": 284.62908935546875, |
| "learning_rate": 9.376854599406528e-07, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -105.5, |
| "logps/rejected": -131.0, |
| "loss": 0.5454, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0771484375, |
| "rewards/margins": 0.41015625, |
| "rewards/rejected": -0.33203125, |
| "step": 179 |
| }, |
| { |
| "epoch": 0.5333333333333333, |
| "grad_norm": 303.474853515625, |
| "learning_rate": 9.317507418397625e-07, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.474609375, |
| "logps/chosen": -106.5, |
| "logps/rejected": -112.5, |
| "loss": 0.5734, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.134765625, |
| "rewards/margins": 0.3046875, |
| "rewards/rejected": -0.169921875, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.5362962962962963, |
| "grad_norm": 296.1676025390625, |
| "learning_rate": 9.258160237388723e-07, |
| "logits/chosen": -0.423828125, |
| "logits/rejected": -0.431640625, |
| "logps/chosen": -110.5, |
| "logps/rejected": -107.0, |
| "loss": 0.5694, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.1181640625, |
| "rewards/margins": 0.3359375, |
| "rewards/rejected": -0.21875, |
| "step": 181 |
| }, |
| { |
| "epoch": 0.5392592592592592, |
| "grad_norm": 275.0120849609375, |
| "learning_rate": 9.198813056379822e-07, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.41796875, |
| "logps/chosen": -94.0, |
| "logps/rejected": -100.0, |
| "loss": 0.6093, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1279296875, |
| "rewards/margins": 0.2080078125, |
| "rewards/rejected": -0.07958984375, |
| "step": 182 |
| }, |
| { |
| "epoch": 0.5422222222222223, |
| "grad_norm": 323.9147033691406, |
| "learning_rate": 9.13946587537092e-07, |
| "logits/chosen": -0.45703125, |
| "logits/rejected": -0.455078125, |
| "logps/chosen": -125.5, |
| "logps/rejected": -101.5, |
| "loss": 0.5729, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1220703125, |
| "rewards/margins": 0.28515625, |
| "rewards/rejected": -0.1630859375, |
| "step": 183 |
| }, |
| { |
| "epoch": 0.5451851851851852, |
| "grad_norm": 307.3267517089844, |
| "learning_rate": 9.080118694362017e-07, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.5, |
| "logps/chosen": -98.0, |
| "logps/rejected": -149.0, |
| "loss": 0.5919, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.08837890625, |
| "rewards/margins": 0.318359375, |
| "rewards/rejected": -0.2294921875, |
| "step": 184 |
| }, |
| { |
| "epoch": 0.5481481481481482, |
| "grad_norm": 321.1439514160156, |
| "learning_rate": 9.020771513353115e-07, |
| "logits/chosen": -0.384765625, |
| "logits/rejected": -0.431640625, |
| "logps/chosen": -95.5, |
| "logps/rejected": -104.5, |
| "loss": 0.6293, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.11572265625, |
| "rewards/margins": 0.1962890625, |
| "rewards/rejected": -0.08056640625, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.5511111111111111, |
| "grad_norm": 328.7718200683594, |
| "learning_rate": 8.961424332344213e-07, |
| "logits/chosen": -0.41015625, |
| "logits/rejected": -0.404296875, |
| "logps/chosen": -124.0, |
| "logps/rejected": -109.0, |
| "loss": 0.6006, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.125, |
| "rewards/margins": 0.248046875, |
| "rewards/rejected": -0.12353515625, |
| "step": 186 |
| }, |
| { |
| "epoch": 0.554074074074074, |
| "grad_norm": 342.34442138671875, |
| "learning_rate": 8.902077151335311e-07, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.40234375, |
| "logps/chosen": -116.0, |
| "logps/rejected": -117.5, |
| "loss": 0.6102, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.033203125, |
| "rewards/margins": 0.2275390625, |
| "rewards/rejected": -0.1943359375, |
| "step": 187 |
| }, |
| { |
| "epoch": 0.557037037037037, |
| "grad_norm": 340.3980712890625, |
| "learning_rate": 8.842729970326409e-07, |
| "logits/chosen": -0.396484375, |
| "logits/rejected": -0.38671875, |
| "logps/chosen": -122.5, |
| "logps/rejected": -111.0, |
| "loss": 0.5869, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.0576171875, |
| "rewards/margins": 0.287109375, |
| "rewards/rejected": -0.2294921875, |
| "step": 188 |
| }, |
| { |
| "epoch": 0.56, |
| "grad_norm": 355.1933288574219, |
| "learning_rate": 8.783382789317507e-07, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.421875, |
| "logps/chosen": -118.5, |
| "logps/rejected": -102.0, |
| "loss": 0.6305, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.0732421875, |
| "rewards/margins": 0.169921875, |
| "rewards/rejected": -0.09619140625, |
| "step": 189 |
| }, |
| { |
| "epoch": 0.562962962962963, |
| "grad_norm": 286.2369689941406, |
| "learning_rate": 8.724035608308605e-07, |
| "logits/chosen": -0.4453125, |
| "logits/rejected": -0.494140625, |
| "logps/chosen": -111.0, |
| "logps/rejected": -137.0, |
| "loss": 0.5476, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.173828125, |
| "rewards/margins": 0.37890625, |
| "rewards/rejected": -0.2041015625, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.5659259259259259, |
| "grad_norm": 362.5728759765625, |
| "learning_rate": 8.664688427299704e-07, |
| "logits/chosen": -0.373046875, |
| "logits/rejected": -0.44140625, |
| "logps/chosen": -101.0, |
| "logps/rejected": -130.0, |
| "loss": 0.5947, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.06103515625, |
| "rewards/margins": 0.259765625, |
| "rewards/rejected": -0.19921875, |
| "step": 191 |
| }, |
| { |
| "epoch": 0.5688888888888889, |
| "grad_norm": 316.38043212890625, |
| "learning_rate": 8.6053412462908e-07, |
| "logits/chosen": -0.388671875, |
| "logits/rejected": -0.41796875, |
| "logps/chosen": -93.5, |
| "logps/rejected": -99.5, |
| "loss": 0.6141, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.0869140625, |
| "rewards/margins": 0.2119140625, |
| "rewards/rejected": -0.125, |
| "step": 192 |
| }, |
| { |
| "epoch": 0.5718518518518518, |
| "grad_norm": 298.670166015625, |
| "learning_rate": 8.545994065281898e-07, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -101.0, |
| "logps/rejected": -122.0, |
| "loss": 0.5581, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.197265625, |
| "rewards/margins": 0.365234375, |
| "rewards/rejected": -0.166015625, |
| "step": 193 |
| }, |
| { |
| "epoch": 0.5748148148148148, |
| "grad_norm": 331.94580078125, |
| "learning_rate": 8.486646884272997e-07, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.44921875, |
| "logps/chosen": -107.0, |
| "logps/rejected": -160.0, |
| "loss": 0.5891, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.08203125, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.265625, |
| "step": 194 |
| }, |
| { |
| "epoch": 0.5777777777777777, |
| "grad_norm": 312.4453125, |
| "learning_rate": 8.427299703264095e-07, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -96.5, |
| "logps/rejected": -117.5, |
| "loss": 0.5737, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1201171875, |
| "rewards/margins": 0.306640625, |
| "rewards/rejected": -0.1865234375, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.5807407407407408, |
| "grad_norm": 300.3534240722656, |
| "learning_rate": 8.367952522255193e-07, |
| "logits/chosen": -0.435546875, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -110.5, |
| "logps/rejected": -131.0, |
| "loss": 0.5817, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.038330078125, |
| "rewards/margins": 0.333984375, |
| "rewards/rejected": -0.296875, |
| "step": 196 |
| }, |
| { |
| "epoch": 0.5837037037037037, |
| "grad_norm": 319.2505798339844, |
| "learning_rate": 8.308605341246291e-07, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -96.0, |
| "logps/rejected": -107.0, |
| "loss": 0.5994, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1171875, |
| "rewards/margins": 0.25390625, |
| "rewards/rejected": -0.13671875, |
| "step": 197 |
| }, |
| { |
| "epoch": 0.5866666666666667, |
| "grad_norm": 315.4450988769531, |
| "learning_rate": 8.249258160237388e-07, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -115.0, |
| "logps/rejected": -114.5, |
| "loss": 0.5748, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.0986328125, |
| "rewards/margins": 0.310546875, |
| "rewards/rejected": -0.212890625, |
| "step": 198 |
| }, |
| { |
| "epoch": 0.5896296296296296, |
| "grad_norm": 303.500732421875, |
| "learning_rate": 8.189910979228485e-07, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.48828125, |
| "logps/chosen": -108.0, |
| "logps/rejected": -122.5, |
| "loss": 0.5804, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.140625, |
| "rewards/margins": 0.29296875, |
| "rewards/rejected": -0.15234375, |
| "step": 199 |
| }, |
| { |
| "epoch": 0.5925925925925926, |
| "grad_norm": 319.0402526855469, |
| "learning_rate": 8.130563798219584e-07, |
| "logits/chosen": -0.48046875, |
| "logits/rejected": -0.474609375, |
| "logps/chosen": -113.5, |
| "logps/rejected": -111.0, |
| "loss": 0.6036, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.11865234375, |
| "rewards/margins": 0.2421875, |
| "rewards/rejected": -0.12353515625, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.5955555555555555, |
| "grad_norm": 320.802490234375, |
| "learning_rate": 8.071216617210682e-07, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.396484375, |
| "logps/chosen": -97.5, |
| "logps/rejected": -107.5, |
| "loss": 0.5582, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.134765625, |
| "rewards/margins": 0.365234375, |
| "rewards/rejected": -0.23046875, |
| "step": 201 |
| }, |
| { |
| "epoch": 0.5985185185185186, |
| "grad_norm": 341.16033935546875, |
| "learning_rate": 8.011869436201781e-07, |
| "logits/chosen": -0.451171875, |
| "logits/rejected": -0.453125, |
| "logps/chosen": -110.0, |
| "logps/rejected": -103.0, |
| "loss": 0.6088, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.0289306640625, |
| "rewards/margins": 0.2021484375, |
| "rewards/rejected": -0.173828125, |
| "step": 202 |
| }, |
| { |
| "epoch": 0.6014814814814815, |
| "grad_norm": 339.4845275878906, |
| "learning_rate": 7.952522255192878e-07, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -124.5, |
| "logps/rejected": -141.0, |
| "loss": 0.5481, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.146484375, |
| "rewards/margins": 0.3671875, |
| "rewards/rejected": -0.2216796875, |
| "step": 203 |
| }, |
| { |
| "epoch": 0.6044444444444445, |
| "grad_norm": 288.8636169433594, |
| "learning_rate": 7.893175074183976e-07, |
| "logits/chosen": -0.44921875, |
| "logits/rejected": -0.494140625, |
| "logps/chosen": -99.0, |
| "logps/rejected": -138.0, |
| "loss": 0.5414, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1494140625, |
| "rewards/margins": 0.44140625, |
| "rewards/rejected": -0.291015625, |
| "step": 204 |
| }, |
| { |
| "epoch": 0.6074074074074074, |
| "grad_norm": 297.95355224609375, |
| "learning_rate": 7.833827893175074e-07, |
| "logits/chosen": -0.447265625, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -105.5, |
| "logps/rejected": -147.0, |
| "loss": 0.5584, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.150390625, |
| "rewards/margins": 0.43359375, |
| "rewards/rejected": -0.283203125, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.6103703703703703, |
| "grad_norm": 259.13623046875, |
| "learning_rate": 7.774480712166172e-07, |
| "logits/chosen": -0.4921875, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -107.0, |
| "logps/rejected": -213.0, |
| "loss": 0.4939, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.1640625, |
| "rewards/margins": 0.6796875, |
| "rewards/rejected": -0.515625, |
| "step": 206 |
| }, |
| { |
| "epoch": 0.6133333333333333, |
| "grad_norm": 352.1375732421875, |
| "learning_rate": 7.715133531157269e-07, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.42578125, |
| "logps/chosen": -106.0, |
| "logps/rejected": -113.5, |
| "loss": 0.6303, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.036865234375, |
| "rewards/margins": 0.162109375, |
| "rewards/rejected": -0.1259765625, |
| "step": 207 |
| }, |
| { |
| "epoch": 0.6162962962962963, |
| "grad_norm": 309.4278869628906, |
| "learning_rate": 7.655786350148368e-07, |
| "logits/chosen": -0.455078125, |
| "logits/rejected": -0.48046875, |
| "logps/chosen": -121.5, |
| "logps/rejected": -150.0, |
| "loss": 0.5443, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.08740234375, |
| "rewards/margins": 0.3984375, |
| "rewards/rejected": -0.3125, |
| "step": 208 |
| }, |
| { |
| "epoch": 0.6192592592592593, |
| "grad_norm": 302.8489990234375, |
| "learning_rate": 7.596439169139466e-07, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.423828125, |
| "logps/chosen": -94.5, |
| "logps/rejected": -106.0, |
| "loss": 0.5962, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1259765625, |
| "rewards/margins": 0.28125, |
| "rewards/rejected": -0.1552734375, |
| "step": 209 |
| }, |
| { |
| "epoch": 0.6222222222222222, |
| "grad_norm": 315.8456115722656, |
| "learning_rate": 7.537091988130564e-07, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.443359375, |
| "logps/chosen": -108.5, |
| "logps/rejected": -136.0, |
| "loss": 0.5422, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1064453125, |
| "rewards/margins": 0.42578125, |
| "rewards/rejected": -0.3203125, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.6251851851851852, |
| "grad_norm": 327.0802307128906, |
| "learning_rate": 7.477744807121661e-07, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -97.5, |
| "logps/rejected": -109.0, |
| "loss": 0.5616, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.08056640625, |
| "rewards/margins": 0.3359375, |
| "rewards/rejected": -0.255859375, |
| "step": 211 |
| }, |
| { |
| "epoch": 0.6281481481481481, |
| "grad_norm": 317.4538879394531, |
| "learning_rate": 7.418397626112759e-07, |
| "logits/chosen": -0.46484375, |
| "logits/rejected": -0.486328125, |
| "logps/chosen": -106.5, |
| "logps/rejected": -114.0, |
| "loss": 0.5953, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.17578125, |
| "rewards/margins": 0.26953125, |
| "rewards/rejected": -0.09375, |
| "step": 212 |
| }, |
| { |
| "epoch": 0.6311111111111111, |
| "grad_norm": 334.5902099609375, |
| "learning_rate": 7.359050445103857e-07, |
| "logits/chosen": -0.435546875, |
| "logits/rejected": -0.4921875, |
| "logps/chosen": -110.5, |
| "logps/rejected": -127.0, |
| "loss": 0.55, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.1904296875, |
| "rewards/margins": 0.423828125, |
| "rewards/rejected": -0.2333984375, |
| "step": 213 |
| }, |
| { |
| "epoch": 0.6340740740740741, |
| "grad_norm": 275.5820007324219, |
| "learning_rate": 7.299703264094955e-07, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.45703125, |
| "logps/chosen": -85.5, |
| "logps/rejected": -114.5, |
| "loss": 0.5648, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1376953125, |
| "rewards/margins": 0.416015625, |
| "rewards/rejected": -0.27734375, |
| "step": 214 |
| }, |
| { |
| "epoch": 0.6370370370370371, |
| "grad_norm": 326.2808532714844, |
| "learning_rate": 7.240356083086053e-07, |
| "logits/chosen": -0.423828125, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -101.5, |
| "logps/rejected": -113.0, |
| "loss": 0.6071, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.0830078125, |
| "rewards/margins": 0.2333984375, |
| "rewards/rejected": -0.150390625, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.64, |
| "grad_norm": 345.86163330078125, |
| "learning_rate": 7.181008902077151e-07, |
| "logits/chosen": -0.455078125, |
| "logits/rejected": -0.494140625, |
| "logps/chosen": -117.0, |
| "logps/rejected": -119.0, |
| "loss": 0.5734, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.024169921875, |
| "rewards/margins": 0.314453125, |
| "rewards/rejected": -0.291015625, |
| "step": 216 |
| }, |
| { |
| "epoch": 0.642962962962963, |
| "grad_norm": 302.5231018066406, |
| "learning_rate": 7.12166172106825e-07, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -113.5, |
| "logps/rejected": -122.0, |
| "loss": 0.55, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.1123046875, |
| "rewards/margins": 0.38671875, |
| "rewards/rejected": -0.2734375, |
| "step": 217 |
| }, |
| { |
| "epoch": 0.6459259259259259, |
| "grad_norm": 321.6429443359375, |
| "learning_rate": 7.062314540059346e-07, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.431640625, |
| "logps/chosen": -114.0, |
| "logps/rejected": -110.5, |
| "loss": 0.6326, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0810546875, |
| "rewards/margins": 0.185546875, |
| "rewards/rejected": -0.1044921875, |
| "step": 218 |
| }, |
| { |
| "epoch": 0.6488888888888888, |
| "grad_norm": 321.2485656738281, |
| "learning_rate": 7.002967359050444e-07, |
| "logits/chosen": -0.40234375, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -104.0, |
| "logps/rejected": -146.0, |
| "loss": 0.5269, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.09228515625, |
| "rewards/margins": 0.453125, |
| "rewards/rejected": -0.361328125, |
| "step": 219 |
| }, |
| { |
| "epoch": 0.6518518518518519, |
| "grad_norm": 351.099609375, |
| "learning_rate": 6.943620178041543e-07, |
| "logits/chosen": -0.484375, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -125.5, |
| "logps/rejected": -114.0, |
| "loss": 0.585, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.11962890625, |
| "rewards/margins": 0.333984375, |
| "rewards/rejected": -0.21484375, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.6548148148148148, |
| "grad_norm": 319.5416259765625, |
| "learning_rate": 6.884272997032641e-07, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.423828125, |
| "logps/chosen": -94.0, |
| "logps/rejected": -102.5, |
| "loss": 0.5757, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.173828125, |
| "rewards/margins": 0.3125, |
| "rewards/rejected": -0.138671875, |
| "step": 221 |
| }, |
| { |
| "epoch": 0.6577777777777778, |
| "grad_norm": 315.0635681152344, |
| "learning_rate": 6.824925816023738e-07, |
| "logits/chosen": -0.455078125, |
| "logits/rejected": -0.486328125, |
| "logps/chosen": -99.5, |
| "logps/rejected": -109.5, |
| "loss": 0.5692, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.1181640625, |
| "rewards/margins": 0.328125, |
| "rewards/rejected": -0.2099609375, |
| "step": 222 |
| }, |
| { |
| "epoch": 0.6607407407407407, |
| "grad_norm": 403.2724304199219, |
| "learning_rate": 6.765578635014837e-07, |
| "logits/chosen": -0.451171875, |
| "logits/rejected": -0.490234375, |
| "logps/chosen": -134.0, |
| "logps/rejected": -116.5, |
| "loss": 0.6002, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.0771484375, |
| "rewards/margins": 0.30859375, |
| "rewards/rejected": -0.232421875, |
| "step": 223 |
| }, |
| { |
| "epoch": 0.6637037037037037, |
| "grad_norm": 296.415771484375, |
| "learning_rate": 6.706231454005934e-07, |
| "logits/chosen": -0.46875, |
| "logits/rejected": -0.5, |
| "logps/chosen": -110.5, |
| "logps/rejected": -144.0, |
| "loss": 0.5433, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.11279296875, |
| "rewards/margins": 0.48828125, |
| "rewards/rejected": -0.376953125, |
| "step": 224 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 289.037841796875, |
| "learning_rate": 6.646884272997032e-07, |
| "logits/chosen": -0.486328125, |
| "logits/rejected": -0.498046875, |
| "logps/chosen": -125.0, |
| "logps/rejected": -127.0, |
| "loss": 0.5747, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.1298828125, |
| "rewards/margins": 0.306640625, |
| "rewards/rejected": -0.17578125, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.6696296296296296, |
| "grad_norm": 308.6339111328125, |
| "learning_rate": 6.58753709198813e-07, |
| "logits/chosen": -0.45703125, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -89.0, |
| "logps/rejected": -125.0, |
| "loss": 0.5129, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.302734375, |
| "rewards/margins": 0.470703125, |
| "rewards/rejected": -0.1689453125, |
| "step": 226 |
| }, |
| { |
| "epoch": 0.6725925925925926, |
| "grad_norm": 345.19512939453125, |
| "learning_rate": 6.528189910979228e-07, |
| "logits/chosen": -0.498046875, |
| "logits/rejected": -0.50390625, |
| "logps/chosen": -122.0, |
| "logps/rejected": -111.0, |
| "loss": 0.6402, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.0673828125, |
| "rewards/margins": 0.16796875, |
| "rewards/rejected": -0.10107421875, |
| "step": 227 |
| }, |
| { |
| "epoch": 0.6755555555555556, |
| "grad_norm": 309.3340148925781, |
| "learning_rate": 6.468842729970327e-07, |
| "logits/chosen": -0.478515625, |
| "logits/rejected": -0.46484375, |
| "logps/chosen": -101.0, |
| "logps/rejected": -105.5, |
| "loss": 0.5455, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.19921875, |
| "rewards/margins": 0.369140625, |
| "rewards/rejected": -0.1689453125, |
| "step": 228 |
| }, |
| { |
| "epoch": 0.6785185185185185, |
| "grad_norm": 342.4759216308594, |
| "learning_rate": 6.409495548961425e-07, |
| "logits/chosen": -0.484375, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -124.5, |
| "logps/rejected": -125.5, |
| "loss": 0.5702, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.1689453125, |
| "rewards/margins": 0.318359375, |
| "rewards/rejected": -0.150390625, |
| "step": 229 |
| }, |
| { |
| "epoch": 0.6814814814814815, |
| "grad_norm": 359.9955749511719, |
| "learning_rate": 6.350148367952522e-07, |
| "logits/chosen": -0.46484375, |
| "logits/rejected": -0.50390625, |
| "logps/chosen": -116.5, |
| "logps/rejected": -132.0, |
| "loss": 0.5958, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": 0.1533203125, |
| "rewards/margins": 0.298828125, |
| "rewards/rejected": -0.1455078125, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.6844444444444444, |
| "grad_norm": 413.5058898925781, |
| "learning_rate": 6.29080118694362e-07, |
| "logits/chosen": -0.46875, |
| "logits/rejected": -0.46484375, |
| "logps/chosen": -123.0, |
| "logps/rejected": -115.5, |
| "loss": 0.6223, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.0625, |
| "rewards/margins": 0.236328125, |
| "rewards/rejected": -0.173828125, |
| "step": 231 |
| }, |
| { |
| "epoch": 0.6874074074074074, |
| "grad_norm": 396.54669189453125, |
| "learning_rate": 6.231454005934718e-07, |
| "logits/chosen": -0.455078125, |
| "logits/rejected": -0.44921875, |
| "logps/chosen": -109.0, |
| "logps/rejected": -109.0, |
| "loss": 0.6149, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.1845703125, |
| "rewards/margins": 0.220703125, |
| "rewards/rejected": -0.03515625, |
| "step": 232 |
| }, |
| { |
| "epoch": 0.6903703703703704, |
| "grad_norm": 324.2456359863281, |
| "learning_rate": 6.172106824925815e-07, |
| "logits/chosen": -0.48046875, |
| "logits/rejected": -0.466796875, |
| "logps/chosen": -121.0, |
| "logps/rejected": -114.5, |
| "loss": 0.6149, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.130859375, |
| "rewards/margins": 0.193359375, |
| "rewards/rejected": -0.0625, |
| "step": 233 |
| }, |
| { |
| "epoch": 0.6933333333333334, |
| "grad_norm": 378.22955322265625, |
| "learning_rate": 6.112759643916914e-07, |
| "logits/chosen": -0.474609375, |
| "logits/rejected": -0.474609375, |
| "logps/chosen": -126.0, |
| "logps/rejected": -117.0, |
| "loss": 0.6306, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.0986328125, |
| "rewards/margins": 0.1865234375, |
| "rewards/rejected": -0.08740234375, |
| "step": 234 |
| }, |
| { |
| "epoch": 0.6962962962962963, |
| "grad_norm": 286.87713623046875, |
| "learning_rate": 6.053412462908012e-07, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -96.0, |
| "logps/rejected": -106.0, |
| "loss": 0.5373, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.216796875, |
| "rewards/margins": 0.412109375, |
| "rewards/rejected": -0.1953125, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.6992592592592592, |
| "grad_norm": 287.8903503417969, |
| "learning_rate": 5.99406528189911e-07, |
| "logits/chosen": -0.462890625, |
| "logits/rejected": -0.435546875, |
| "logps/chosen": -100.0, |
| "logps/rejected": -167.0, |
| "loss": 0.5497, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1533203125, |
| "rewards/margins": 0.5546875, |
| "rewards/rejected": -0.404296875, |
| "step": 236 |
| }, |
| { |
| "epoch": 0.7022222222222222, |
| "grad_norm": 389.0058288574219, |
| "learning_rate": 5.934718100890207e-07, |
| "logits/chosen": -0.5078125, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -115.5, |
| "logps/rejected": -118.5, |
| "loss": 0.5813, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1435546875, |
| "rewards/margins": 0.291015625, |
| "rewards/rejected": -0.1474609375, |
| "step": 237 |
| }, |
| { |
| "epoch": 0.7051851851851851, |
| "grad_norm": 322.4640197753906, |
| "learning_rate": 5.875370919881305e-07, |
| "logits/chosen": -0.455078125, |
| "logits/rejected": -0.48046875, |
| "logps/chosen": -116.5, |
| "logps/rejected": -127.5, |
| "loss": 0.5622, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.12890625, |
| "rewards/margins": 0.3671875, |
| "rewards/rejected": -0.23828125, |
| "step": 238 |
| }, |
| { |
| "epoch": 0.7081481481481482, |
| "grad_norm": 356.6217346191406, |
| "learning_rate": 5.816023738872403e-07, |
| "logits/chosen": -0.50390625, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -134.0, |
| "logps/rejected": -116.5, |
| "loss": 0.5994, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.15234375, |
| "rewards/margins": 0.2431640625, |
| "rewards/rejected": -0.0908203125, |
| "step": 239 |
| }, |
| { |
| "epoch": 0.7111111111111111, |
| "grad_norm": 352.3586120605469, |
| "learning_rate": 5.756676557863502e-07, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.453125, |
| "logps/chosen": -120.0, |
| "logps/rejected": -137.0, |
| "loss": 0.557, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.1376953125, |
| "rewards/margins": 0.404296875, |
| "rewards/rejected": -0.267578125, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.7140740740740741, |
| "grad_norm": 319.5504455566406, |
| "learning_rate": 5.697329376854599e-07, |
| "logits/chosen": -0.419921875, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -108.0, |
| "logps/rejected": -166.0, |
| "loss": 0.4836, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.287109375, |
| "rewards/margins": 0.63671875, |
| "rewards/rejected": -0.349609375, |
| "step": 241 |
| }, |
| { |
| "epoch": 0.717037037037037, |
| "grad_norm": 279.9657287597656, |
| "learning_rate": 5.637982195845697e-07, |
| "logits/chosen": -0.443359375, |
| "logits/rejected": -0.478515625, |
| "logps/chosen": -102.0, |
| "logps/rejected": -115.0, |
| "loss": 0.5181, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.21484375, |
| "rewards/margins": 0.4609375, |
| "rewards/rejected": -0.24609375, |
| "step": 242 |
| }, |
| { |
| "epoch": 0.72, |
| "grad_norm": 352.0918273925781, |
| "learning_rate": 5.578635014836796e-07, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.416015625, |
| "logps/chosen": -103.5, |
| "logps/rejected": -174.0, |
| "loss": 0.6038, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.09619140625, |
| "rewards/margins": 0.375, |
| "rewards/rejected": -0.279296875, |
| "step": 243 |
| }, |
| { |
| "epoch": 0.7229629629629629, |
| "grad_norm": 338.6324157714844, |
| "learning_rate": 5.519287833827893e-07, |
| "logits/chosen": -0.486328125, |
| "logits/rejected": -0.53125, |
| "logps/chosen": -115.0, |
| "logps/rejected": -143.0, |
| "loss": 0.5329, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.17578125, |
| "rewards/margins": 0.44921875, |
| "rewards/rejected": -0.2734375, |
| "step": 244 |
| }, |
| { |
| "epoch": 0.725925925925926, |
| "grad_norm": 355.90570068359375, |
| "learning_rate": 5.45994065281899e-07, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.44921875, |
| "logps/chosen": -122.5, |
| "logps/rejected": -116.0, |
| "loss": 0.6306, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.10302734375, |
| "rewards/margins": 0.1962890625, |
| "rewards/rejected": -0.0927734375, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.7288888888888889, |
| "grad_norm": 315.7834777832031, |
| "learning_rate": 5.400593471810089e-07, |
| "logits/chosen": -0.458984375, |
| "logits/rejected": -0.48046875, |
| "logps/chosen": -116.5, |
| "logps/rejected": -216.0, |
| "loss": 0.5527, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.1572265625, |
| "rewards/margins": 0.486328125, |
| "rewards/rejected": -0.330078125, |
| "step": 246 |
| }, |
| { |
| "epoch": 0.7318518518518519, |
| "grad_norm": 355.6773376464844, |
| "learning_rate": 5.341246290801187e-07, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -99.5, |
| "logps/rejected": -128.0, |
| "loss": 0.5726, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.25390625, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.09375, |
| "step": 247 |
| }, |
| { |
| "epoch": 0.7348148148148148, |
| "grad_norm": 345.4238586425781, |
| "learning_rate": 5.281899109792285e-07, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.482421875, |
| "logps/chosen": -114.5, |
| "logps/rejected": -103.0, |
| "loss": 0.5754, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1767578125, |
| "rewards/margins": 0.326171875, |
| "rewards/rejected": -0.150390625, |
| "step": 248 |
| }, |
| { |
| "epoch": 0.7377777777777778, |
| "grad_norm": 347.1054992675781, |
| "learning_rate": 5.222551928783383e-07, |
| "logits/chosen": -0.484375, |
| "logits/rejected": -0.48828125, |
| "logps/chosen": -117.0, |
| "logps/rejected": -155.0, |
| "loss": 0.5783, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1904296875, |
| "rewards/margins": 0.419921875, |
| "rewards/rejected": -0.2275390625, |
| "step": 249 |
| }, |
| { |
| "epoch": 0.7407407407407407, |
| "grad_norm": 320.8048400878906, |
| "learning_rate": 5.16320474777448e-07, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.462890625, |
| "logps/chosen": -97.0, |
| "logps/rejected": -112.5, |
| "loss": 0.5957, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.18359375, |
| "rewards/margins": 0.283203125, |
| "rewards/rejected": -0.10009765625, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.7437037037037038, |
| "grad_norm": 338.7988586425781, |
| "learning_rate": 5.103857566765578e-07, |
| "logits/chosen": -0.447265625, |
| "logits/rejected": -0.453125, |
| "logps/chosen": -102.5, |
| "logps/rejected": -118.0, |
| "loss": 0.5697, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.27734375, |
| "rewards/margins": 0.361328125, |
| "rewards/rejected": -0.08349609375, |
| "step": 251 |
| }, |
| { |
| "epoch": 0.7466666666666667, |
| "grad_norm": 294.05718994140625, |
| "learning_rate": 5.044510385756676e-07, |
| "logits/chosen": -0.42578125, |
| "logits/rejected": -0.478515625, |
| "logps/chosen": -105.0, |
| "logps/rejected": -107.5, |
| "loss": 0.5615, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.162109375, |
| "rewards/margins": 0.3515625, |
| "rewards/rejected": -0.1904296875, |
| "step": 252 |
| }, |
| { |
| "epoch": 0.7496296296296296, |
| "grad_norm": 338.3261413574219, |
| "learning_rate": 4.985163204747774e-07, |
| "logits/chosen": -0.474609375, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -115.5, |
| "logps/rejected": -118.0, |
| "loss": 0.5705, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.27734375, |
| "rewards/margins": 0.35546875, |
| "rewards/rejected": -0.07958984375, |
| "step": 253 |
| }, |
| { |
| "epoch": 0.7525925925925926, |
| "grad_norm": 303.49481201171875, |
| "learning_rate": 4.925816023738872e-07, |
| "logits/chosen": -0.4765625, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -105.0, |
| "logps/rejected": -107.0, |
| "loss": 0.5983, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.236328125, |
| "rewards/margins": 0.251953125, |
| "rewards/rejected": -0.0166015625, |
| "step": 254 |
| }, |
| { |
| "epoch": 0.7555555555555555, |
| "grad_norm": 314.61431884765625, |
| "learning_rate": 4.86646884272997e-07, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.384765625, |
| "logps/chosen": -108.0, |
| "logps/rejected": -248.0, |
| "loss": 0.5377, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.140625, |
| "rewards/margins": 0.74609375, |
| "rewards/rejected": -0.60546875, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.7585185185185185, |
| "grad_norm": 338.1864318847656, |
| "learning_rate": 4.807121661721068e-07, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -102.0, |
| "logps/rejected": -111.5, |
| "loss": 0.655, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": 0.1650390625, |
| "rewards/margins": 0.12353515625, |
| "rewards/rejected": 0.04150390625, |
| "step": 256 |
| }, |
| { |
| "epoch": 0.7614814814814815, |
| "grad_norm": 338.2643127441406, |
| "learning_rate": 4.747774480712166e-07, |
| "logits/chosen": -0.443359375, |
| "logits/rejected": -0.447265625, |
| "logps/chosen": -118.5, |
| "logps/rejected": -122.0, |
| "loss": 0.56, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.2578125, |
| "rewards/margins": 0.39453125, |
| "rewards/rejected": -0.13671875, |
| "step": 257 |
| }, |
| { |
| "epoch": 0.7644444444444445, |
| "grad_norm": 358.0396423339844, |
| "learning_rate": 4.688427299703264e-07, |
| "logits/chosen": -0.4609375, |
| "logits/rejected": -0.4921875, |
| "logps/chosen": -123.5, |
| "logps/rejected": -131.0, |
| "loss": 0.5951, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.173828125, |
| "rewards/margins": 0.31640625, |
| "rewards/rejected": -0.1435546875, |
| "step": 258 |
| }, |
| { |
| "epoch": 0.7674074074074074, |
| "grad_norm": 310.6124267578125, |
| "learning_rate": 4.6290801186943617e-07, |
| "logits/chosen": -0.5, |
| "logits/rejected": -0.51171875, |
| "logps/chosen": -91.5, |
| "logps/rejected": -120.5, |
| "loss": 0.5808, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.2177734375, |
| "rewards/margins": 0.357421875, |
| "rewards/rejected": -0.138671875, |
| "step": 259 |
| }, |
| { |
| "epoch": 0.7703703703703704, |
| "grad_norm": 335.0301513671875, |
| "learning_rate": 4.56973293768546e-07, |
| "logits/chosen": -0.451171875, |
| "logits/rejected": -0.451171875, |
| "logps/chosen": -87.5, |
| "logps/rejected": -109.0, |
| "loss": 0.5374, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.29296875, |
| "rewards/margins": 0.40234375, |
| "rewards/rejected": -0.1083984375, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.7733333333333333, |
| "grad_norm": 335.72308349609375, |
| "learning_rate": 4.5103857566765573e-07, |
| "logits/chosen": -0.404296875, |
| "logits/rejected": -0.427734375, |
| "logps/chosen": -90.5, |
| "logps/rejected": -96.5, |
| "loss": 0.6355, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.1826171875, |
| "rewards/margins": 0.166015625, |
| "rewards/rejected": 0.0164794921875, |
| "step": 261 |
| }, |
| { |
| "epoch": 0.7762962962962963, |
| "grad_norm": 341.7281799316406, |
| "learning_rate": 4.4510385756676557e-07, |
| "logits/chosen": -0.462890625, |
| "logits/rejected": -0.4453125, |
| "logps/chosen": -108.0, |
| "logps/rejected": -99.0, |
| "loss": 0.6155, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.15234375, |
| "rewards/margins": 0.208984375, |
| "rewards/rejected": -0.05615234375, |
| "step": 262 |
| }, |
| { |
| "epoch": 0.7792592592592592, |
| "grad_norm": 344.0790710449219, |
| "learning_rate": 4.3916913946587536e-07, |
| "logits/chosen": -0.458984375, |
| "logits/rejected": -0.494140625, |
| "logps/chosen": -92.5, |
| "logps/rejected": -112.5, |
| "loss": 0.5807, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.2216796875, |
| "rewards/margins": 0.3125, |
| "rewards/rejected": -0.09130859375, |
| "step": 263 |
| }, |
| { |
| "epoch": 0.7822222222222223, |
| "grad_norm": 352.5789794921875, |
| "learning_rate": 4.332344213649852e-07, |
| "logits/chosen": -0.451171875, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -115.0, |
| "logps/rejected": -139.0, |
| "loss": 0.5781, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.2294921875, |
| "rewards/margins": 0.333984375, |
| "rewards/rejected": -0.10498046875, |
| "step": 264 |
| }, |
| { |
| "epoch": 0.7851851851851852, |
| "grad_norm": 297.2604675292969, |
| "learning_rate": 4.272997032640949e-07, |
| "logits/chosen": -0.486328125, |
| "logits/rejected": -0.53125, |
| "logps/chosen": -122.5, |
| "logps/rejected": -175.0, |
| "loss": 0.5029, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.2470703125, |
| "rewards/margins": 0.640625, |
| "rewards/rejected": -0.392578125, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.7881481481481482, |
| "grad_norm": 311.6089782714844, |
| "learning_rate": 4.2136498516320476e-07, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -122.5, |
| "logps/rejected": -125.0, |
| "loss": 0.5132, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.2216796875, |
| "rewards/margins": 0.4921875, |
| "rewards/rejected": -0.271484375, |
| "step": 266 |
| }, |
| { |
| "epoch": 0.7911111111111111, |
| "grad_norm": 313.684326171875, |
| "learning_rate": 4.1543026706231454e-07, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.484375, |
| "logps/chosen": -107.5, |
| "logps/rejected": -106.5, |
| "loss": 0.5455, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.30859375, |
| "rewards/margins": 0.380859375, |
| "rewards/rejected": -0.0712890625, |
| "step": 267 |
| }, |
| { |
| "epoch": 0.794074074074074, |
| "grad_norm": 336.37811279296875, |
| "learning_rate": 4.094955489614243e-07, |
| "logits/chosen": -0.46875, |
| "logits/rejected": -0.51171875, |
| "logps/chosen": -130.0, |
| "logps/rejected": -112.5, |
| "loss": 0.5732, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.134765625, |
| "rewards/margins": 0.322265625, |
| "rewards/rejected": -0.1884765625, |
| "step": 268 |
| }, |
| { |
| "epoch": 0.797037037037037, |
| "grad_norm": 357.6194152832031, |
| "learning_rate": 4.035608308605341e-07, |
| "logits/chosen": -0.486328125, |
| "logits/rejected": -0.4921875, |
| "logps/chosen": -115.0, |
| "logps/rejected": -132.0, |
| "loss": 0.6066, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1748046875, |
| "rewards/margins": 0.267578125, |
| "rewards/rejected": -0.0927734375, |
| "step": 269 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 362.451171875, |
| "learning_rate": 3.976261127596439e-07, |
| "logits/chosen": -0.470703125, |
| "logits/rejected": -0.478515625, |
| "logps/chosen": -119.0, |
| "logps/rejected": -117.5, |
| "loss": 0.5926, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.189453125, |
| "rewards/margins": 0.31640625, |
| "rewards/rejected": -0.126953125, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.802962962962963, |
| "grad_norm": 324.0819091796875, |
| "learning_rate": 3.916913946587537e-07, |
| "logits/chosen": -0.484375, |
| "logits/rejected": -0.423828125, |
| "logps/chosen": -113.5, |
| "logps/rejected": -137.0, |
| "loss": 0.5917, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.12890625, |
| "rewards/margins": 0.328125, |
| "rewards/rejected": -0.2001953125, |
| "step": 271 |
| }, |
| { |
| "epoch": 0.8059259259259259, |
| "grad_norm": 333.343505859375, |
| "learning_rate": 3.8575667655786346e-07, |
| "logits/chosen": -0.447265625, |
| "logits/rejected": -0.455078125, |
| "logps/chosen": -112.0, |
| "logps/rejected": -133.0, |
| "loss": 0.5617, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.203125, |
| "rewards/margins": 0.345703125, |
| "rewards/rejected": -0.142578125, |
| "step": 272 |
| }, |
| { |
| "epoch": 0.8088888888888889, |
| "grad_norm": 379.63433837890625, |
| "learning_rate": 3.798219584569733e-07, |
| "logits/chosen": -0.44140625, |
| "logits/rejected": -0.466796875, |
| "logps/chosen": -93.0, |
| "logps/rejected": -99.5, |
| "loss": 0.6309, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.1884765625, |
| "rewards/margins": 0.1748046875, |
| "rewards/rejected": 0.01336669921875, |
| "step": 273 |
| }, |
| { |
| "epoch": 0.8118518518518518, |
| "grad_norm": 336.7275390625, |
| "learning_rate": 3.7388724035608303e-07, |
| "logits/chosen": -0.466796875, |
| "logits/rejected": -0.48046875, |
| "logps/chosen": -107.5, |
| "logps/rejected": -112.0, |
| "loss": 0.5811, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.291015625, |
| "rewards/margins": 0.33203125, |
| "rewards/rejected": -0.04150390625, |
| "step": 274 |
| }, |
| { |
| "epoch": 0.8148148148148148, |
| "grad_norm": 308.87542724609375, |
| "learning_rate": 3.6795252225519287e-07, |
| "logits/chosen": -0.47265625, |
| "logits/rejected": -0.49609375, |
| "logps/chosen": -108.5, |
| "logps/rejected": -120.0, |
| "loss": 0.532, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.259765625, |
| "rewards/margins": 0.40234375, |
| "rewards/rejected": -0.1435546875, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.8177777777777778, |
| "grad_norm": 349.4025573730469, |
| "learning_rate": 3.6201780415430265e-07, |
| "logits/chosen": -0.4140625, |
| "logits/rejected": -0.439453125, |
| "logps/chosen": -116.0, |
| "logps/rejected": -111.0, |
| "loss": 0.6223, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.154296875, |
| "rewards/margins": 0.1728515625, |
| "rewards/rejected": -0.017822265625, |
| "step": 276 |
| }, |
| { |
| "epoch": 0.8207407407407408, |
| "grad_norm": 327.4034423828125, |
| "learning_rate": 3.560830860534125e-07, |
| "logits/chosen": -0.447265625, |
| "logits/rejected": -0.474609375, |
| "logps/chosen": -114.0, |
| "logps/rejected": -105.0, |
| "loss": 0.5864, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.2431640625, |
| "rewards/margins": 0.28515625, |
| "rewards/rejected": -0.04150390625, |
| "step": 277 |
| }, |
| { |
| "epoch": 0.8237037037037037, |
| "grad_norm": 326.1805725097656, |
| "learning_rate": 3.501483679525222e-07, |
| "logits/chosen": -0.466796875, |
| "logits/rejected": -0.515625, |
| "logps/chosen": -112.0, |
| "logps/rejected": -139.0, |
| "loss": 0.4851, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.306640625, |
| "rewards/margins": 0.54296875, |
| "rewards/rejected": -0.2373046875, |
| "step": 278 |
| }, |
| { |
| "epoch": 0.8266666666666667, |
| "grad_norm": 337.7111511230469, |
| "learning_rate": 3.4421364985163206e-07, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.45703125, |
| "logps/chosen": -107.0, |
| "logps/rejected": -108.5, |
| "loss": 0.5501, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.23828125, |
| "rewards/margins": 0.400390625, |
| "rewards/rejected": -0.162109375, |
| "step": 279 |
| }, |
| { |
| "epoch": 0.8296296296296296, |
| "grad_norm": 305.23504638671875, |
| "learning_rate": 3.3827893175074184e-07, |
| "logits/chosen": -0.421875, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -115.5, |
| "logps/rejected": -117.0, |
| "loss": 0.5378, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.259765625, |
| "rewards/margins": 0.431640625, |
| "rewards/rejected": -0.1728515625, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.8325925925925926, |
| "grad_norm": 303.7311096191406, |
| "learning_rate": 3.323442136498516e-07, |
| "logits/chosen": -0.47265625, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -108.5, |
| "logps/rejected": -119.0, |
| "loss": 0.5375, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": 0.2255859375, |
| "rewards/margins": 0.4140625, |
| "rewards/rejected": -0.189453125, |
| "step": 281 |
| }, |
| { |
| "epoch": 0.8355555555555556, |
| "grad_norm": 306.6491394042969, |
| "learning_rate": 3.264094955489614e-07, |
| "logits/chosen": -0.45703125, |
| "logits/rejected": -0.443359375, |
| "logps/chosen": -113.5, |
| "logps/rejected": -191.0, |
| "loss": 0.5292, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.28125, |
| "rewards/margins": 0.58203125, |
| "rewards/rejected": -0.302734375, |
| "step": 282 |
| }, |
| { |
| "epoch": 0.8385185185185186, |
| "grad_norm": 344.1654052734375, |
| "learning_rate": 3.2047477744807125e-07, |
| "logits/chosen": -0.458984375, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -124.5, |
| "logps/rejected": -153.0, |
| "loss": 0.4954, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": 0.30859375, |
| "rewards/margins": 0.58984375, |
| "rewards/rejected": -0.279296875, |
| "step": 283 |
| }, |
| { |
| "epoch": 0.8414814814814815, |
| "grad_norm": 387.68011474609375, |
| "learning_rate": 3.14540059347181e-07, |
| "logits/chosen": -0.48046875, |
| "logits/rejected": -0.470703125, |
| "logps/chosen": -130.0, |
| "logps/rejected": -146.0, |
| "loss": 0.597, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.1298828125, |
| "rewards/margins": 0.326171875, |
| "rewards/rejected": -0.1962890625, |
| "step": 284 |
| }, |
| { |
| "epoch": 0.8444444444444444, |
| "grad_norm": 360.8407287597656, |
| "learning_rate": 3.0860534124629076e-07, |
| "logits/chosen": -0.46484375, |
| "logits/rejected": -0.48046875, |
| "logps/chosen": -100.0, |
| "logps/rejected": -107.0, |
| "loss": 0.5915, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1865234375, |
| "rewards/margins": 0.28125, |
| "rewards/rejected": -0.09423828125, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.8474074074074074, |
| "grad_norm": 328.58441162109375, |
| "learning_rate": 3.026706231454006e-07, |
| "logits/chosen": -0.4765625, |
| "logits/rejected": -0.478515625, |
| "logps/chosen": -105.0, |
| "logps/rejected": -110.0, |
| "loss": 0.5646, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.212890625, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.1337890625, |
| "step": 286 |
| }, |
| { |
| "epoch": 0.8503703703703703, |
| "grad_norm": 351.5137634277344, |
| "learning_rate": 2.9673590504451033e-07, |
| "logits/chosen": -0.412109375, |
| "logits/rejected": -0.40234375, |
| "logps/chosen": -98.5, |
| "logps/rejected": -155.0, |
| "loss": 0.6265, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.197265625, |
| "rewards/margins": 0.33203125, |
| "rewards/rejected": -0.1337890625, |
| "step": 287 |
| }, |
| { |
| "epoch": 0.8533333333333334, |
| "grad_norm": 312.41815185546875, |
| "learning_rate": 2.9080118694362016e-07, |
| "logits/chosen": -0.5078125, |
| "logits/rejected": -0.53515625, |
| "logps/chosen": -120.0, |
| "logps/rejected": -133.0, |
| "loss": 0.5293, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.1884765625, |
| "rewards/margins": 0.48828125, |
| "rewards/rejected": -0.30078125, |
| "step": 288 |
| }, |
| { |
| "epoch": 0.8562962962962963, |
| "grad_norm": 340.3605041503906, |
| "learning_rate": 2.8486646884272995e-07, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.474609375, |
| "logps/chosen": -114.5, |
| "logps/rejected": -107.5, |
| "loss": 0.572, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1748046875, |
| "rewards/margins": 0.330078125, |
| "rewards/rejected": -0.1552734375, |
| "step": 289 |
| }, |
| { |
| "epoch": 0.8592592592592593, |
| "grad_norm": 362.09100341796875, |
| "learning_rate": 2.789317507418398e-07, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.4296875, |
| "logps/chosen": -110.5, |
| "logps/rejected": -113.0, |
| "loss": 0.6093, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.1513671875, |
| "rewards/margins": 0.2138671875, |
| "rewards/rejected": -0.0625, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.8622222222222222, |
| "grad_norm": 329.35833740234375, |
| "learning_rate": 2.729970326409495e-07, |
| "logits/chosen": -0.48046875, |
| "logits/rejected": -0.51953125, |
| "logps/chosen": -106.5, |
| "logps/rejected": -138.0, |
| "loss": 0.4774, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": 0.30859375, |
| "rewards/margins": 0.578125, |
| "rewards/rejected": -0.26953125, |
| "step": 291 |
| }, |
| { |
| "epoch": 0.8651851851851852, |
| "grad_norm": 323.2444763183594, |
| "learning_rate": 2.6706231454005935e-07, |
| "logits/chosen": -0.470703125, |
| "logits/rejected": -0.5234375, |
| "logps/chosen": -93.0, |
| "logps/rejected": -116.0, |
| "loss": 0.6023, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.2216796875, |
| "rewards/margins": 0.2578125, |
| "rewards/rejected": -0.035888671875, |
| "step": 292 |
| }, |
| { |
| "epoch": 0.8681481481481481, |
| "grad_norm": 350.0760498046875, |
| "learning_rate": 2.6112759643916914e-07, |
| "logits/chosen": -0.46484375, |
| "logits/rejected": -0.52734375, |
| "logps/chosen": -116.5, |
| "logps/rejected": -153.0, |
| "loss": 0.5517, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.154296875, |
| "rewards/margins": 0.4765625, |
| "rewards/rejected": -0.322265625, |
| "step": 293 |
| }, |
| { |
| "epoch": 0.8711111111111111, |
| "grad_norm": 304.44775390625, |
| "learning_rate": 2.551928783382789e-07, |
| "logits/chosen": -0.46875, |
| "logits/rejected": -0.50390625, |
| "logps/chosen": -101.0, |
| "logps/rejected": -109.0, |
| "loss": 0.5915, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.318359375, |
| "rewards/margins": 0.294921875, |
| "rewards/rejected": 0.0235595703125, |
| "step": 294 |
| }, |
| { |
| "epoch": 0.8740740740740741, |
| "grad_norm": 324.566162109375, |
| "learning_rate": 2.492581602373887e-07, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.47265625, |
| "logps/chosen": -110.0, |
| "logps/rejected": -112.5, |
| "loss": 0.5785, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.26953125, |
| "rewards/margins": 0.318359375, |
| "rewards/rejected": -0.04833984375, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.8770370370370371, |
| "grad_norm": 328.0279541015625, |
| "learning_rate": 2.433234421364985e-07, |
| "logits/chosen": -0.462890625, |
| "logits/rejected": -0.431640625, |
| "logps/chosen": -110.5, |
| "logps/rejected": -108.0, |
| "loss": 0.5424, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.26171875, |
| "rewards/margins": 0.431640625, |
| "rewards/rejected": -0.1689453125, |
| "step": 296 |
| }, |
| { |
| "epoch": 0.88, |
| "grad_norm": 302.4769287109375, |
| "learning_rate": 2.373887240356083e-07, |
| "logits/chosen": -0.48828125, |
| "logits/rejected": -0.5, |
| "logps/chosen": -117.5, |
| "logps/rejected": -158.0, |
| "loss": 0.5054, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.2578125, |
| "rewards/margins": 0.51953125, |
| "rewards/rejected": -0.259765625, |
| "step": 297 |
| }, |
| { |
| "epoch": 0.882962962962963, |
| "grad_norm": 312.2156677246094, |
| "learning_rate": 2.3145400593471808e-07, |
| "logits/chosen": -0.416015625, |
| "logits/rejected": -0.443359375, |
| "logps/chosen": -87.0, |
| "logps/rejected": -102.5, |
| "loss": 0.5788, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.251953125, |
| "rewards/margins": 0.30859375, |
| "rewards/rejected": -0.057373046875, |
| "step": 298 |
| }, |
| { |
| "epoch": 0.8859259259259259, |
| "grad_norm": 295.56170654296875, |
| "learning_rate": 2.2551928783382787e-07, |
| "logits/chosen": -0.451171875, |
| "logits/rejected": -0.4375, |
| "logps/chosen": -118.0, |
| "logps/rejected": -108.5, |
| "loss": 0.5136, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.287109375, |
| "rewards/margins": 0.4609375, |
| "rewards/rejected": -0.1748046875, |
| "step": 299 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 366.7168884277344, |
| "learning_rate": 2.1958456973293768e-07, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -87.0, |
| "logps/rejected": -111.0, |
| "loss": 0.635, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.26171875, |
| "rewards/margins": 0.1767578125, |
| "rewards/rejected": 0.0849609375, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.8918518518518519, |
| "grad_norm": 333.2161865234375, |
| "learning_rate": 2.1364985163204746e-07, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.435546875, |
| "logps/chosen": -95.5, |
| "logps/rejected": -136.0, |
| "loss": 0.6377, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.2392578125, |
| "rewards/margins": 0.294921875, |
| "rewards/rejected": -0.05615234375, |
| "step": 301 |
| }, |
| { |
| "epoch": 0.8948148148148148, |
| "grad_norm": 296.5155334472656, |
| "learning_rate": 2.0771513353115727e-07, |
| "logits/chosen": -0.48828125, |
| "logits/rejected": -0.458984375, |
| "logps/chosen": -97.0, |
| "logps/rejected": -149.0, |
| "loss": 0.4789, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.345703125, |
| "rewards/margins": 0.62890625, |
| "rewards/rejected": -0.283203125, |
| "step": 302 |
| }, |
| { |
| "epoch": 0.8977777777777778, |
| "grad_norm": 309.7770690917969, |
| "learning_rate": 2.0178041543026706e-07, |
| "logits/chosen": -0.4296875, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -99.5, |
| "logps/rejected": -142.0, |
| "loss": 0.5024, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.263671875, |
| "rewards/margins": 0.5078125, |
| "rewards/rejected": -0.2431640625, |
| "step": 303 |
| }, |
| { |
| "epoch": 0.9007407407407407, |
| "grad_norm": 381.62310791015625, |
| "learning_rate": 1.9584569732937684e-07, |
| "logits/chosen": -0.462890625, |
| "logits/rejected": -0.46484375, |
| "logps/chosen": -115.5, |
| "logps/rejected": -96.0, |
| "loss": 0.6429, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.22265625, |
| "rewards/margins": 0.1767578125, |
| "rewards/rejected": 0.046142578125, |
| "step": 304 |
| }, |
| { |
| "epoch": 0.9037037037037037, |
| "grad_norm": 328.7142639160156, |
| "learning_rate": 1.8991097922848665e-07, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.451171875, |
| "logps/chosen": -88.0, |
| "logps/rejected": -99.5, |
| "loss": 0.6071, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.267578125, |
| "rewards/margins": 0.26953125, |
| "rewards/rejected": -0.0022430419921875, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.9066666666666666, |
| "grad_norm": 359.6824951171875, |
| "learning_rate": 1.8397626112759643e-07, |
| "logits/chosen": -0.515625, |
| "logits/rejected": -0.5, |
| "logps/chosen": -129.0, |
| "logps/rejected": -103.5, |
| "loss": 0.5629, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.236328125, |
| "rewards/margins": 0.35546875, |
| "rewards/rejected": -0.1181640625, |
| "step": 306 |
| }, |
| { |
| "epoch": 0.9096296296296297, |
| "grad_norm": 362.33837890625, |
| "learning_rate": 1.7804154302670624e-07, |
| "logits/chosen": -0.396484375, |
| "logits/rejected": -0.380859375, |
| "logps/chosen": -110.5, |
| "logps/rejected": -109.5, |
| "loss": 0.5921, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.263671875, |
| "rewards/margins": 0.275390625, |
| "rewards/rejected": -0.0111083984375, |
| "step": 307 |
| }, |
| { |
| "epoch": 0.9125925925925926, |
| "grad_norm": 323.42071533203125, |
| "learning_rate": 1.7210682492581603e-07, |
| "logits/chosen": -0.47265625, |
| "logits/rejected": -0.4921875, |
| "logps/chosen": -87.5, |
| "logps/rejected": -102.0, |
| "loss": 0.5582, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.2734375, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.07421875, |
| "step": 308 |
| }, |
| { |
| "epoch": 0.9155555555555556, |
| "grad_norm": 311.4526672363281, |
| "learning_rate": 1.661721068249258e-07, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.4375, |
| "logps/chosen": -108.5, |
| "logps/rejected": -104.5, |
| "loss": 0.5549, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.310546875, |
| "rewards/margins": 0.3984375, |
| "rewards/rejected": -0.087890625, |
| "step": 309 |
| }, |
| { |
| "epoch": 0.9185185185185185, |
| "grad_norm": 333.0624694824219, |
| "learning_rate": 1.6023738872403562e-07, |
| "logits/chosen": -0.46875, |
| "logits/rejected": -0.453125, |
| "logps/chosen": -119.0, |
| "logps/rejected": -106.5, |
| "loss": 0.5909, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.1650390625, |
| "rewards/margins": 0.26953125, |
| "rewards/rejected": -0.10498046875, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.9214814814814815, |
| "grad_norm": 333.5960998535156, |
| "learning_rate": 1.5430267062314538e-07, |
| "logits/chosen": -0.458984375, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -119.5, |
| "logps/rejected": -179.0, |
| "loss": 0.5661, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.193359375, |
| "rewards/margins": 0.40234375, |
| "rewards/rejected": -0.208984375, |
| "step": 311 |
| }, |
| { |
| "epoch": 0.9244444444444444, |
| "grad_norm": 347.4278259277344, |
| "learning_rate": 1.4836795252225516e-07, |
| "logits/chosen": -0.4765625, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -125.5, |
| "logps/rejected": -181.0, |
| "loss": 0.5573, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.27734375, |
| "rewards/margins": 0.4453125, |
| "rewards/rejected": -0.1669921875, |
| "step": 312 |
| }, |
| { |
| "epoch": 0.9274074074074075, |
| "grad_norm": 373.39276123046875, |
| "learning_rate": 1.4243323442136497e-07, |
| "logits/chosen": -0.427734375, |
| "logits/rejected": -0.451171875, |
| "logps/chosen": -99.0, |
| "logps/rejected": -134.0, |
| "loss": 0.5681, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.255859375, |
| "rewards/margins": 0.361328125, |
| "rewards/rejected": -0.1064453125, |
| "step": 313 |
| }, |
| { |
| "epoch": 0.9303703703703704, |
| "grad_norm": 302.965576171875, |
| "learning_rate": 1.3649851632047476e-07, |
| "logits/chosen": -0.5078125, |
| "logits/rejected": -0.4765625, |
| "logps/chosen": -127.0, |
| "logps/rejected": -210.0, |
| "loss": 0.5341, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.208984375, |
| "rewards/margins": 0.5625, |
| "rewards/rejected": -0.3515625, |
| "step": 314 |
| }, |
| { |
| "epoch": 0.9333333333333333, |
| "grad_norm": 305.0396728515625, |
| "learning_rate": 1.3056379821958457e-07, |
| "logits/chosen": -0.462890625, |
| "logits/rejected": -0.51171875, |
| "logps/chosen": -86.0, |
| "logps/rejected": -108.5, |
| "loss": 0.5503, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.271484375, |
| "rewards/margins": 0.341796875, |
| "rewards/rejected": -0.0703125, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.9362962962962963, |
| "grad_norm": 350.77337646484375, |
| "learning_rate": 1.2462908011869435e-07, |
| "logits/chosen": -0.375, |
| "logits/rejected": -0.4375, |
| "logps/chosen": -112.5, |
| "logps/rejected": -138.0, |
| "loss": 0.6021, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.193359375, |
| "rewards/margins": 0.267578125, |
| "rewards/rejected": -0.07470703125, |
| "step": 316 |
| }, |
| { |
| "epoch": 0.9392592592592592, |
| "grad_norm": 320.9561462402344, |
| "learning_rate": 1.1869436201780415e-07, |
| "logits/chosen": -0.47265625, |
| "logits/rejected": -0.466796875, |
| "logps/chosen": -104.0, |
| "logps/rejected": -183.0, |
| "loss": 0.531, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.287109375, |
| "rewards/margins": 0.55078125, |
| "rewards/rejected": -0.265625, |
| "step": 317 |
| }, |
| { |
| "epoch": 0.9422222222222222, |
| "grad_norm": 322.83856201171875, |
| "learning_rate": 1.1275964391691393e-07, |
| "logits/chosen": -0.41796875, |
| "logits/rejected": -0.4609375, |
| "logps/chosen": -99.5, |
| "logps/rejected": -124.5, |
| "loss": 0.5408, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.244140625, |
| "rewards/margins": 0.392578125, |
| "rewards/rejected": -0.1474609375, |
| "step": 318 |
| }, |
| { |
| "epoch": 0.9451851851851852, |
| "grad_norm": 306.716796875, |
| "learning_rate": 1.0682492581602373e-07, |
| "logits/chosen": -0.4765625, |
| "logits/rejected": -0.5078125, |
| "logps/chosen": -99.0, |
| "logps/rejected": -133.0, |
| "loss": 0.5377, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.2373046875, |
| "rewards/margins": 0.42578125, |
| "rewards/rejected": -0.1884765625, |
| "step": 319 |
| }, |
| { |
| "epoch": 0.9481481481481482, |
| "grad_norm": 345.20361328125, |
| "learning_rate": 1.0089020771513353e-07, |
| "logits/chosen": -0.470703125, |
| "logits/rejected": -0.46875, |
| "logps/chosen": -131.0, |
| "logps/rejected": -163.0, |
| "loss": 0.5181, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.25, |
| "rewards/margins": 0.53125, |
| "rewards/rejected": -0.279296875, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.9511111111111111, |
| "grad_norm": 340.1747741699219, |
| "learning_rate": 9.495548961424333e-08, |
| "logits/chosen": -0.451171875, |
| "logits/rejected": -0.48828125, |
| "logps/chosen": -97.0, |
| "logps/rejected": -98.5, |
| "loss": 0.6364, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": 0.15625, |
| "rewards/margins": 0.17578125, |
| "rewards/rejected": -0.0198974609375, |
| "step": 321 |
| }, |
| { |
| "epoch": 0.9540740740740741, |
| "grad_norm": 363.9776916503906, |
| "learning_rate": 8.902077151335312e-08, |
| "logits/chosen": -0.48046875, |
| "logits/rejected": -0.5078125, |
| "logps/chosen": -105.0, |
| "logps/rejected": -127.0, |
| "loss": 0.5558, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.26953125, |
| "rewards/margins": 0.431640625, |
| "rewards/rejected": -0.162109375, |
| "step": 322 |
| }, |
| { |
| "epoch": 0.957037037037037, |
| "grad_norm": 305.4539794921875, |
| "learning_rate": 8.30860534124629e-08, |
| "logits/chosen": -0.5, |
| "logits/rejected": -0.5078125, |
| "logps/chosen": -102.5, |
| "logps/rejected": -90.5, |
| "loss": 0.5662, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.2021484375, |
| "rewards/margins": 0.326171875, |
| "rewards/rejected": -0.125, |
| "step": 323 |
| }, |
| { |
| "epoch": 0.96, |
| "grad_norm": 326.83056640625, |
| "learning_rate": 7.715133531157269e-08, |
| "logits/chosen": -0.51171875, |
| "logits/rejected": -0.515625, |
| "logps/chosen": -115.5, |
| "logps/rejected": -112.5, |
| "loss": 0.6027, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.2197265625, |
| "rewards/margins": 0.259765625, |
| "rewards/rejected": -0.0400390625, |
| "step": 324 |
| }, |
| { |
| "epoch": 0.9629629629629629, |
| "grad_norm": 316.9124755859375, |
| "learning_rate": 7.121661721068249e-08, |
| "logits/chosen": -0.458984375, |
| "logits/rejected": -0.5, |
| "logps/chosen": -116.0, |
| "logps/rejected": -161.0, |
| "loss": 0.5458, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.208984375, |
| "rewards/margins": 0.4375, |
| "rewards/rejected": -0.2294921875, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.965925925925926, |
| "grad_norm": 319.8908386230469, |
| "learning_rate": 6.528189910979228e-08, |
| "logits/chosen": -0.494140625, |
| "logits/rejected": -0.515625, |
| "logps/chosen": -104.0, |
| "logps/rejected": -96.5, |
| "loss": 0.5284, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.345703125, |
| "rewards/margins": 0.4296875, |
| "rewards/rejected": -0.08447265625, |
| "step": 326 |
| }, |
| { |
| "epoch": 0.9688888888888889, |
| "grad_norm": 300.6716003417969, |
| "learning_rate": 5.9347181008902075e-08, |
| "logits/chosen": -0.474609375, |
| "logits/rejected": -0.462890625, |
| "logps/chosen": -92.0, |
| "logps/rejected": -108.5, |
| "loss": 0.5072, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.365234375, |
| "rewards/margins": 0.47265625, |
| "rewards/rejected": -0.10693359375, |
| "step": 327 |
| }, |
| { |
| "epoch": 0.9718518518518519, |
| "grad_norm": 345.5276794433594, |
| "learning_rate": 5.3412462908011865e-08, |
| "logits/chosen": -0.431640625, |
| "logits/rejected": -0.46484375, |
| "logps/chosen": -110.5, |
| "logps/rejected": -119.5, |
| "loss": 0.6319, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.1005859375, |
| "rewards/margins": 0.1806640625, |
| "rewards/rejected": -0.07958984375, |
| "step": 328 |
| }, |
| { |
| "epoch": 0.9748148148148148, |
| "grad_norm": 322.6155090332031, |
| "learning_rate": 4.747774480712166e-08, |
| "logits/chosen": -0.43359375, |
| "logits/rejected": -0.48828125, |
| "logps/chosen": -100.0, |
| "logps/rejected": -130.0, |
| "loss": 0.5236, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": 0.298828125, |
| "rewards/margins": 0.466796875, |
| "rewards/rejected": -0.1689453125, |
| "step": 329 |
| }, |
| { |
| "epoch": 0.9777777777777777, |
| "grad_norm": 331.7804260253906, |
| "learning_rate": 4.154302670623145e-08, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.48828125, |
| "logps/chosen": -108.0, |
| "logps/rejected": -146.0, |
| "loss": 0.5869, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": 0.158203125, |
| "rewards/margins": 0.390625, |
| "rewards/rejected": -0.2333984375, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.9807407407407407, |
| "grad_norm": 346.5263366699219, |
| "learning_rate": 3.5608308605341244e-08, |
| "logits/chosen": -0.482421875, |
| "logits/rejected": -0.478515625, |
| "logps/chosen": -129.0, |
| "logps/rejected": -124.0, |
| "loss": 0.5934, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": 0.130859375, |
| "rewards/margins": 0.279296875, |
| "rewards/rejected": -0.1474609375, |
| "step": 331 |
| }, |
| { |
| "epoch": 0.9837037037037037, |
| "grad_norm": 328.090576171875, |
| "learning_rate": 2.9673590504451037e-08, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.47265625, |
| "logps/chosen": -135.0, |
| "logps/rejected": -149.0, |
| "loss": 0.5096, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.1630859375, |
| "rewards/margins": 0.48828125, |
| "rewards/rejected": -0.32421875, |
| "step": 332 |
| }, |
| { |
| "epoch": 0.9866666666666667, |
| "grad_norm": 357.4970397949219, |
| "learning_rate": 2.373887240356083e-08, |
| "logits/chosen": -0.5, |
| "logits/rejected": -0.484375, |
| "logps/chosen": -112.0, |
| "logps/rejected": -102.5, |
| "loss": 0.5899, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": 0.181640625, |
| "rewards/margins": 0.30078125, |
| "rewards/rejected": -0.11865234375, |
| "step": 333 |
| }, |
| { |
| "epoch": 0.9896296296296296, |
| "grad_norm": 324.227783203125, |
| "learning_rate": 1.7804154302670622e-08, |
| "logits/chosen": -0.453125, |
| "logits/rejected": -0.494140625, |
| "logps/chosen": -102.5, |
| "logps/rejected": -120.0, |
| "loss": 0.4937, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": 0.26171875, |
| "rewards/margins": 0.53515625, |
| "rewards/rejected": -0.2734375, |
| "step": 334 |
| }, |
| { |
| "epoch": 0.9925925925925926, |
| "grad_norm": 320.154296875, |
| "learning_rate": 1.1869436201780416e-08, |
| "logits/chosen": -0.439453125, |
| "logits/rejected": -0.462890625, |
| "logps/chosen": -100.0, |
| "logps/rejected": -111.5, |
| "loss": 0.5704, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 0.2353515625, |
| "rewards/margins": 0.326171875, |
| "rewards/rejected": -0.08984375, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.9955555555555555, |
| "grad_norm": 321.2402038574219, |
| "learning_rate": 5.934718100890208e-09, |
| "logits/chosen": -0.443359375, |
| "logits/rejected": -0.5234375, |
| "logps/chosen": -115.5, |
| "logps/rejected": -121.5, |
| "loss": 0.5919, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.189453125, |
| "rewards/margins": 0.322265625, |
| "rewards/rejected": -0.1328125, |
| "step": 336 |
| }, |
| { |
| "epoch": 0.9985185185185185, |
| "grad_norm": 356.6434326171875, |
| "learning_rate": 0.0, |
| "logits/chosen": -0.46875, |
| "logits/rejected": -0.482421875, |
| "logps/chosen": -102.5, |
| "logps/rejected": -120.5, |
| "loss": 0.634, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": 0.125, |
| "rewards/margins": 0.1982421875, |
| "rewards/rejected": -0.07275390625, |
| "step": 337 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 337, |
| "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 |
| } |
|
|
