| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9989708404802744, |
| "eval_steps": 500, |
| "global_step": 273, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.003659233847913093, |
| "grad_norm": 346.33831787109375, |
| "learning_rate": 1.9926739926739927e-06, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -162.0, |
| "logps/rejected": -163.0, |
| "loss": 0.6914, |
| "rewards/accuracies": 0.0, |
| "rewards/chosen": 0.0, |
| "rewards/margins": 0.0, |
| "rewards/rejected": 0.0, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.007318467695826186, |
| "grad_norm": 344.1603698730469, |
| "learning_rate": 1.9853479853479855e-06, |
| "logits/chosen": -0.609375, |
| "logits/rejected": -0.625, |
| "logps/chosen": -139.0, |
| "logps/rejected": -188.0, |
| "loss": 0.7092, |
| "rewards/accuracies": 0.25, |
| "rewards/chosen": -0.00860595703125, |
| "rewards/margins": -0.038330078125, |
| "rewards/rejected": 0.0296630859375, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.010977701543739279, |
| "grad_norm": 321.5267028808594, |
| "learning_rate": 1.978021978021978e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -145.0, |
| "logps/rejected": -188.0, |
| "loss": 0.6973, |
| "rewards/accuracies": 0.28125, |
| "rewards/chosen": -0.005462646484375, |
| "rewards/margins": -0.00860595703125, |
| "rewards/rejected": 0.0031280517578125, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.014636935391652372, |
| "grad_norm": 338.7471923828125, |
| "learning_rate": 1.9706959706959706e-06, |
| "logits/chosen": -0.5625, |
| "logits/rejected": -0.60546875, |
| "logps/chosen": -160.0, |
| "logps/rejected": -189.0, |
| "loss": 0.6735, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.0218505859375, |
| "rewards/margins": 0.035888671875, |
| "rewards/rejected": -0.01409912109375, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.018296169239565466, |
| "grad_norm": 315.54010009765625, |
| "learning_rate": 1.9633699633699634e-06, |
| "logits/chosen": -0.55859375, |
| "logits/rejected": -0.58203125, |
| "logps/chosen": -181.0, |
| "logps/rejected": -195.0, |
| "loss": 0.6837, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": 0.0031280517578125, |
| "rewards/margins": 0.0218505859375, |
| "rewards/rejected": -0.018798828125, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.021955403087478557, |
| "grad_norm": 321.4845886230469, |
| "learning_rate": 1.9560439560439557e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -177.0, |
| "logps/rejected": -179.0, |
| "loss": 0.6964, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": -0.0234375, |
| "rewards/margins": -7.62939453125e-06, |
| "rewards/rejected": -0.0234375, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.025614636935391653, |
| "grad_norm": 330.94757080078125, |
| "learning_rate": 1.9487179487179485e-06, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.63671875, |
| "logps/chosen": -145.0, |
| "logps/rejected": -167.0, |
| "loss": 0.6996, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": -0.0218505859375, |
| "rewards/margins": -0.01251220703125, |
| "rewards/rejected": -0.0093994140625, |
| "step": 7 |
| }, |
| { |
| "epoch": 0.029273870783304744, |
| "grad_norm": 348.64080810546875, |
| "learning_rate": 1.9413919413919413e-06, |
| "logits/chosen": -0.59765625, |
| "logits/rejected": -0.609375, |
| "logps/chosen": -169.0, |
| "logps/rejected": -186.0, |
| "loss": 0.692, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": -0.016357421875, |
| "rewards/margins": 0.01019287109375, |
| "rewards/rejected": -0.026611328125, |
| "step": 8 |
| }, |
| { |
| "epoch": 0.03293310463121784, |
| "grad_norm": 331.8152770996094, |
| "learning_rate": 1.934065934065934e-06, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -173.0, |
| "logps/rejected": -182.0, |
| "loss": 0.6917, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.039794921875, |
| "rewards/margins": 0.0015106201171875, |
| "rewards/rejected": -0.04150390625, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.03659233847913093, |
| "grad_norm": 341.0348205566406, |
| "learning_rate": 1.926739926739927e-06, |
| "logits/chosen": -0.58984375, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -156.0, |
| "logps/rejected": -216.0, |
| "loss": 0.6727, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.034423828125, |
| "rewards/margins": 0.042236328125, |
| "rewards/rejected": -0.07666015625, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.04025157232704402, |
| "grad_norm": 314.3051452636719, |
| "learning_rate": 1.9194139194139196e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -152.0, |
| "logps/rejected": -176.0, |
| "loss": 0.6743, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.0250244140625, |
| "rewards/margins": 0.042236328125, |
| "rewards/rejected": -0.0673828125, |
| "step": 11 |
| }, |
| { |
| "epoch": 0.043910806174957115, |
| "grad_norm": 315.991943359375, |
| "learning_rate": 1.912087912087912e-06, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -137.0, |
| "logps/rejected": -167.0, |
| "loss": 0.6733, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.044677734375, |
| "rewards/margins": 0.038330078125, |
| "rewards/rejected": -0.0830078125, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.047570040022870214, |
| "grad_norm": 302.81024169921875, |
| "learning_rate": 1.9047619047619045e-06, |
| "logits/chosen": -0.62890625, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -162.0, |
| "logps/rejected": -162.0, |
| "loss": 0.7054, |
| "rewards/accuracies": 0.28125, |
| "rewards/chosen": -0.078125, |
| "rewards/margins": -0.01806640625, |
| "rewards/rejected": -0.060302734375, |
| "step": 13 |
| }, |
| { |
| "epoch": 0.051229273870783305, |
| "grad_norm": 312.2680969238281, |
| "learning_rate": 1.8974358974358973e-06, |
| "logits/chosen": -0.5859375, |
| "logits/rejected": -0.5625, |
| "logps/chosen": -141.0, |
| "logps/rejected": -140.0, |
| "loss": 0.6847, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.036865234375, |
| "rewards/margins": 0.01953125, |
| "rewards/rejected": -0.056396484375, |
| "step": 14 |
| }, |
| { |
| "epoch": 0.0548885077186964, |
| "grad_norm": 450.512451171875, |
| "learning_rate": 1.89010989010989e-06, |
| "logits/chosen": -0.5703125, |
| "logits/rejected": -0.5703125, |
| "logps/chosen": -170.0, |
| "logps/rejected": -186.0, |
| "loss": 0.703, |
| "rewards/accuracies": 0.3125, |
| "rewards/chosen": -0.09912109375, |
| "rewards/margins": -0.0194091796875, |
| "rewards/rejected": -0.07958984375, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.05854774156660949, |
| "grad_norm": 319.3954772949219, |
| "learning_rate": 1.8827838827838826e-06, |
| "logits/chosen": -0.62890625, |
| "logits/rejected": -0.62109375, |
| "logps/chosen": -183.0, |
| "logps/rejected": -161.0, |
| "loss": 0.6943, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.06591796875, |
| "rewards/margins": 0.004730224609375, |
| "rewards/rejected": -0.0703125, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.06220697541452259, |
| "grad_norm": 328.090087890625, |
| "learning_rate": 1.8754578754578754e-06, |
| "logits/chosen": -0.58984375, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -158.0, |
| "logps/rejected": -169.0, |
| "loss": 0.6694, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.02587890625, |
| "rewards/margins": 0.053955078125, |
| "rewards/rejected": -0.07958984375, |
| "step": 17 |
| }, |
| { |
| "epoch": 0.06586620926243568, |
| "grad_norm": 339.8540954589844, |
| "learning_rate": 1.8681318681318681e-06, |
| "logits/chosen": -0.6328125, |
| "logits/rejected": -0.59375, |
| "logps/chosen": -147.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6788, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.046875, |
| "rewards/margins": 0.035888671875, |
| "rewards/rejected": -0.0830078125, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.06952544311034876, |
| "grad_norm": 316.8853759765625, |
| "learning_rate": 1.8608058608058607e-06, |
| "logits/chosen": -0.60546875, |
| "logits/rejected": -0.609375, |
| "logps/chosen": -150.0, |
| "logps/rejected": -171.0, |
| "loss": 0.682, |
| "rewards/accuracies": 0.40625, |
| "rewards/chosen": -0.050048828125, |
| "rewards/margins": 0.0289306640625, |
| "rewards/rejected": -0.0791015625, |
| "step": 19 |
| }, |
| { |
| "epoch": 0.07318467695826186, |
| "grad_norm": 318.4516296386719, |
| "learning_rate": 1.8534798534798535e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -138.0, |
| "logps/rejected": -179.0, |
| "loss": 0.6903, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.05322265625, |
| "rewards/margins": 0.010986328125, |
| "rewards/rejected": -0.06396484375, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.07684391080617496, |
| "grad_norm": 382.6263122558594, |
| "learning_rate": 1.8461538461538462e-06, |
| "logits/chosen": -0.58203125, |
| "logits/rejected": -0.5546875, |
| "logps/chosen": -143.0, |
| "logps/rejected": -256.0, |
| "loss": 0.6578, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.08154296875, |
| "rewards/rejected": -0.126953125, |
| "step": 21 |
| }, |
| { |
| "epoch": 0.08050314465408805, |
| "grad_norm": 351.90972900390625, |
| "learning_rate": 1.8388278388278386e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.578125, |
| "logps/chosen": -152.0, |
| "logps/rejected": -213.0, |
| "loss": 0.6783, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.059326171875, |
| "rewards/margins": 0.0296630859375, |
| "rewards/rejected": -0.08935546875, |
| "step": 22 |
| }, |
| { |
| "epoch": 0.08416237850200115, |
| "grad_norm": 306.3103332519531, |
| "learning_rate": 1.8315018315018314e-06, |
| "logits/chosen": -0.59765625, |
| "logits/rejected": -0.578125, |
| "logps/chosen": -140.0, |
| "logps/rejected": -173.0, |
| "loss": 0.6982, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": -0.07177734375, |
| "rewards/margins": 0.00628662109375, |
| "rewards/rejected": -0.078125, |
| "step": 23 |
| }, |
| { |
| "epoch": 0.08782161234991423, |
| "grad_norm": 322.9337158203125, |
| "learning_rate": 1.8241758241758241e-06, |
| "logits/chosen": -0.6015625, |
| "logits/rejected": -0.57421875, |
| "logps/chosen": -149.0, |
| "logps/rejected": -160.0, |
| "loss": 0.6774, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.0595703125, |
| "rewards/margins": 0.03759765625, |
| "rewards/rejected": -0.09716796875, |
| "step": 24 |
| }, |
| { |
| "epoch": 0.09148084619782733, |
| "grad_norm": 348.9863586425781, |
| "learning_rate": 1.8168498168498167e-06, |
| "logits/chosen": -0.59765625, |
| "logits/rejected": -0.57421875, |
| "logps/chosen": -183.0, |
| "logps/rejected": -224.0, |
| "loss": 0.6976, |
| "rewards/accuracies": 0.34375, |
| "rewards/chosen": -0.10498046875, |
| "rewards/margins": 0.0078125, |
| "rewards/rejected": -0.11279296875, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.09514008004574043, |
| "grad_norm": 312.5710144042969, |
| "learning_rate": 1.8095238095238095e-06, |
| "logits/chosen": -0.59375, |
| "logits/rejected": -0.57421875, |
| "logps/chosen": -160.0, |
| "logps/rejected": -156.0, |
| "loss": 0.6707, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.078125, |
| "rewards/margins": 0.0361328125, |
| "rewards/rejected": -0.1142578125, |
| "step": 26 |
| }, |
| { |
| "epoch": 0.09879931389365351, |
| "grad_norm": 312.729736328125, |
| "learning_rate": 1.8021978021978023e-06, |
| "logits/chosen": -0.58984375, |
| "logits/rejected": -0.56640625, |
| "logps/chosen": -152.0, |
| "logps/rejected": -177.0, |
| "loss": 0.6608, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.06884765625, |
| "rewards/margins": 0.0703125, |
| "rewards/rejected": -0.1396484375, |
| "step": 27 |
| }, |
| { |
| "epoch": 0.10245854774156661, |
| "grad_norm": 335.3136291503906, |
| "learning_rate": 1.7948717948717948e-06, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -178.0, |
| "logps/rejected": -159.0, |
| "loss": 0.7061, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": -0.142578125, |
| "rewards/margins": -0.0186767578125, |
| "rewards/rejected": -0.12353515625, |
| "step": 28 |
| }, |
| { |
| "epoch": 0.10611778158947971, |
| "grad_norm": 308.06744384765625, |
| "learning_rate": 1.7875457875457876e-06, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -177.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6785, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.10009765625, |
| "rewards/margins": 0.035888671875, |
| "rewards/rejected": -0.1357421875, |
| "step": 29 |
| }, |
| { |
| "epoch": 0.1097770154373928, |
| "grad_norm": 324.392333984375, |
| "learning_rate": 1.78021978021978e-06, |
| "logits/chosen": -0.62890625, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -169.0, |
| "logps/rejected": -166.0, |
| "loss": 0.6895, |
| "rewards/accuracies": 0.375, |
| "rewards/chosen": -0.0908203125, |
| "rewards/margins": 0.01409912109375, |
| "rewards/rejected": -0.10498046875, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.11343624928530589, |
| "grad_norm": 348.1455993652344, |
| "learning_rate": 1.7728937728937727e-06, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -150.0, |
| "logps/rejected": -173.0, |
| "loss": 0.6512, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.057861328125, |
| "rewards/margins": 0.0859375, |
| "rewards/rejected": -0.1435546875, |
| "step": 31 |
| }, |
| { |
| "epoch": 0.11709548313321898, |
| "grad_norm": 310.3366394042969, |
| "learning_rate": 1.7655677655677655e-06, |
| "logits/chosen": -0.6171875, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -185.0, |
| "logps/rejected": -176.0, |
| "loss": 0.6638, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.08154296875, |
| "rewards/margins": 0.06494140625, |
| "rewards/rejected": -0.146484375, |
| "step": 32 |
| }, |
| { |
| "epoch": 0.12075471698113208, |
| "grad_norm": 367.6607971191406, |
| "learning_rate": 1.758241758241758e-06, |
| "logits/chosen": -0.609375, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -213.0, |
| "logps/rejected": -152.0, |
| "loss": 0.7106, |
| "rewards/accuracies": 0.40625, |
| "rewards/chosen": -0.15625, |
| "rewards/margins": -0.0234375, |
| "rewards/rejected": -0.1328125, |
| "step": 33 |
| }, |
| { |
| "epoch": 0.12441395082904518, |
| "grad_norm": 304.85638427734375, |
| "learning_rate": 1.7509157509157508e-06, |
| "logits/chosen": -0.55859375, |
| "logits/rejected": -0.58203125, |
| "logps/chosen": -155.0, |
| "logps/rejected": -179.0, |
| "loss": 0.676, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.0859375, |
| "rewards/margins": 0.0390625, |
| "rewards/rejected": -0.125, |
| "step": 34 |
| }, |
| { |
| "epoch": 0.12807318467695827, |
| "grad_norm": 344.142333984375, |
| "learning_rate": 1.7435897435897436e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -148.0, |
| "logps/rejected": -158.0, |
| "loss": 0.7004, |
| "rewards/accuracies": 0.28125, |
| "rewards/chosen": -0.10693359375, |
| "rewards/margins": -0.0133056640625, |
| "rewards/rejected": -0.09375, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.13173241852487136, |
| "grad_norm": 330.11029052734375, |
| "learning_rate": 1.7362637362637362e-06, |
| "logits/chosen": -0.5859375, |
| "logits/rejected": -0.59765625, |
| "logps/chosen": -138.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6927, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.0859375, |
| "rewards/margins": 0.010986328125, |
| "rewards/rejected": -0.09716796875, |
| "step": 36 |
| }, |
| { |
| "epoch": 0.13539165237278444, |
| "grad_norm": 367.31817626953125, |
| "learning_rate": 1.728937728937729e-06, |
| "logits/chosen": -0.61328125, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -150.0, |
| "logps/rejected": -159.0, |
| "loss": 0.6768, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.08544921875, |
| "rewards/margins": 0.046142578125, |
| "rewards/rejected": -0.1318359375, |
| "step": 37 |
| }, |
| { |
| "epoch": 0.13905088622069753, |
| "grad_norm": 332.1564025878906, |
| "learning_rate": 1.7216117216117217e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.59375, |
| "logps/chosen": -158.0, |
| "logps/rejected": -179.0, |
| "loss": 0.6858, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": -0.0859375, |
| "rewards/margins": 0.02587890625, |
| "rewards/rejected": -0.11181640625, |
| "step": 38 |
| }, |
| { |
| "epoch": 0.14271012006861064, |
| "grad_norm": 358.6282653808594, |
| "learning_rate": 1.714285714285714e-06, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -151.0, |
| "logps/rejected": -169.0, |
| "loss": 0.6743, |
| "rewards/accuracies": 0.4375, |
| "rewards/chosen": -0.0830078125, |
| "rewards/margins": 0.0390625, |
| "rewards/rejected": -0.1220703125, |
| "step": 39 |
| }, |
| { |
| "epoch": 0.14636935391652373, |
| "grad_norm": 344.2862548828125, |
| "learning_rate": 1.7069597069597068e-06, |
| "logits/chosen": -0.5859375, |
| "logits/rejected": -0.56640625, |
| "logps/chosen": -144.0, |
| "logps/rejected": -224.0, |
| "loss": 0.653, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.0625, |
| "rewards/margins": 0.10791015625, |
| "rewards/rejected": -0.1708984375, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.1500285877644368, |
| "grad_norm": 317.97698974609375, |
| "learning_rate": 1.6996336996336996e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -157.0, |
| "logps/rejected": -177.0, |
| "loss": 0.6595, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.051513671875, |
| "rewards/margins": 0.07763671875, |
| "rewards/rejected": -0.12890625, |
| "step": 41 |
| }, |
| { |
| "epoch": 0.15368782161234992, |
| "grad_norm": 336.11773681640625, |
| "learning_rate": 1.6923076923076922e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -172.0, |
| "logps/rejected": -168.0, |
| "loss": 0.6805, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": -0.09521484375, |
| "rewards/margins": 0.03125, |
| "rewards/rejected": -0.126953125, |
| "step": 42 |
| }, |
| { |
| "epoch": 0.157347055460263, |
| "grad_norm": 306.1403503417969, |
| "learning_rate": 1.684981684981685e-06, |
| "logits/chosen": -0.6328125, |
| "logits/rejected": -0.6015625, |
| "logps/chosen": -146.0, |
| "logps/rejected": -149.0, |
| "loss": 0.6506, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.06103515625, |
| "rewards/margins": 0.0947265625, |
| "rewards/rejected": -0.1552734375, |
| "step": 43 |
| }, |
| { |
| "epoch": 0.1610062893081761, |
| "grad_norm": 330.2728271484375, |
| "learning_rate": 1.6776556776556777e-06, |
| "logits/chosen": -0.609375, |
| "logits/rejected": -0.55859375, |
| "logps/chosen": -141.0, |
| "logps/rejected": -180.0, |
| "loss": 0.6661, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.0703125, |
| "rewards/margins": 0.06103515625, |
| "rewards/rejected": -0.1318359375, |
| "step": 44 |
| }, |
| { |
| "epoch": 0.1646655231560892, |
| "grad_norm": 300.4102783203125, |
| "learning_rate": 1.6703296703296703e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.66015625, |
| "logps/chosen": -148.0, |
| "logps/rejected": -156.0, |
| "loss": 0.6313, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.046142578125, |
| "rewards/margins": 0.1357421875, |
| "rewards/rejected": -0.181640625, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.1683247570040023, |
| "grad_norm": 314.258056640625, |
| "learning_rate": 1.663003663003663e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -145.0, |
| "logps/rejected": -164.0, |
| "loss": 0.658, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.08935546875, |
| "rewards/margins": 0.08056640625, |
| "rewards/rejected": -0.169921875, |
| "step": 46 |
| }, |
| { |
| "epoch": 0.17198399085191537, |
| "grad_norm": 300.92364501953125, |
| "learning_rate": 1.6556776556776554e-06, |
| "logits/chosen": -0.625, |
| "logits/rejected": -0.625, |
| "logps/chosen": -152.0, |
| "logps/rejected": -189.0, |
| "loss": 0.6508, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.09716796875, |
| "rewards/rejected": -0.142578125, |
| "step": 47 |
| }, |
| { |
| "epoch": 0.17564322469982846, |
| "grad_norm": 281.744384765625, |
| "learning_rate": 1.6483516483516482e-06, |
| "logits/chosen": -0.609375, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -131.0, |
| "logps/rejected": -165.0, |
| "loss": 0.655, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.07373046875, |
| "rewards/margins": 0.0888671875, |
| "rewards/rejected": -0.1630859375, |
| "step": 48 |
| }, |
| { |
| "epoch": 0.17930245854774157, |
| "grad_norm": 373.736328125, |
| "learning_rate": 1.641025641025641e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -147.0, |
| "logps/rejected": -176.0, |
| "loss": 0.6635, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.08447265625, |
| "rewards/margins": 0.0751953125, |
| "rewards/rejected": -0.1591796875, |
| "step": 49 |
| }, |
| { |
| "epoch": 0.18296169239565466, |
| "grad_norm": 319.23675537109375, |
| "learning_rate": 1.6336996336996335e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -168.0, |
| "logps/rejected": -183.0, |
| "loss": 0.6371, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.0751953125, |
| "rewards/margins": 0.134765625, |
| "rewards/rejected": -0.2099609375, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.18662092624356774, |
| "grad_norm": 454.0714416503906, |
| "learning_rate": 1.6263736263736263e-06, |
| "logits/chosen": -0.62890625, |
| "logits/rejected": -0.59375, |
| "logps/chosen": -171.0, |
| "logps/rejected": -196.0, |
| "loss": 0.6519, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.08740234375, |
| "rewards/margins": 0.0908203125, |
| "rewards/rejected": -0.1787109375, |
| "step": 51 |
| }, |
| { |
| "epoch": 0.19028016009148085, |
| "grad_norm": 316.0301513671875, |
| "learning_rate": 1.619047619047619e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -144.0, |
| "logps/rejected": -183.0, |
| "loss": 0.6611, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": -0.107421875, |
| "rewards/margins": 0.08544921875, |
| "rewards/rejected": -0.1923828125, |
| "step": 52 |
| }, |
| { |
| "epoch": 0.19393939393939394, |
| "grad_norm": 345.9140625, |
| "learning_rate": 1.6117216117216116e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.625, |
| "logps/chosen": -169.0, |
| "logps/rejected": -173.0, |
| "loss": 0.658, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.09375, |
| "rewards/margins": 0.08740234375, |
| "rewards/rejected": -0.181640625, |
| "step": 53 |
| }, |
| { |
| "epoch": 0.19759862778730702, |
| "grad_norm": 334.124755859375, |
| "learning_rate": 1.6043956043956044e-06, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -178.0, |
| "logps/rejected": -202.0, |
| "loss": 0.6633, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.1630859375, |
| "rewards/margins": 0.0751953125, |
| "rewards/rejected": -0.2373046875, |
| "step": 54 |
| }, |
| { |
| "epoch": 0.20125786163522014, |
| "grad_norm": 330.6783142089844, |
| "learning_rate": 1.5970695970695972e-06, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -163.0, |
| "logps/rejected": -140.0, |
| "loss": 0.698, |
| "rewards/accuracies": 0.46875, |
| "rewards/chosen": -0.1533203125, |
| "rewards/margins": -0.010986328125, |
| "rewards/rejected": -0.142578125, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.20491709548313322, |
| "grad_norm": 370.80706787109375, |
| "learning_rate": 1.5897435897435895e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -166.0, |
| "logps/rejected": -181.0, |
| "loss": 0.6512, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1689453125, |
| "rewards/margins": 0.09375, |
| "rewards/rejected": -0.263671875, |
| "step": 56 |
| }, |
| { |
| "epoch": 0.2085763293310463, |
| "grad_norm": 323.4024353027344, |
| "learning_rate": 1.5824175824175823e-06, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -171.0, |
| "logps/rejected": -157.0, |
| "loss": 0.6428, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1318359375, |
| "rewards/margins": 0.115234375, |
| "rewards/rejected": -0.24609375, |
| "step": 57 |
| }, |
| { |
| "epoch": 0.21223556317895942, |
| "grad_norm": 325.7928771972656, |
| "learning_rate": 1.575091575091575e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.59765625, |
| "logps/chosen": -154.0, |
| "logps/rejected": -167.0, |
| "loss": 0.6663, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.12451171875, |
| "rewards/margins": 0.07421875, |
| "rewards/rejected": -0.1982421875, |
| "step": 58 |
| }, |
| { |
| "epoch": 0.2158947970268725, |
| "grad_norm": 365.8585510253906, |
| "learning_rate": 1.5677655677655676e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.61328125, |
| "logps/chosen": -192.0, |
| "logps/rejected": -202.0, |
| "loss": 0.6435, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1611328125, |
| "rewards/margins": 0.125, |
| "rewards/rejected": -0.287109375, |
| "step": 59 |
| }, |
| { |
| "epoch": 0.2195540308747856, |
| "grad_norm": 339.77593994140625, |
| "learning_rate": 1.5604395604395604e-06, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -162.0, |
| "logps/rejected": -207.0, |
| "loss": 0.6537, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.15234375, |
| "rewards/margins": 0.10595703125, |
| "rewards/rejected": -0.2578125, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.22321326472269867, |
| "grad_norm": 316.12646484375, |
| "learning_rate": 1.5531135531135532e-06, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -137.0, |
| "logps/rejected": -166.0, |
| "loss": 0.6564, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.1494140625, |
| "rewards/margins": 0.0908203125, |
| "rewards/rejected": -0.240234375, |
| "step": 61 |
| }, |
| { |
| "epoch": 0.22687249857061179, |
| "grad_norm": 315.9810485839844, |
| "learning_rate": 1.5457875457875457e-06, |
| "logits/chosen": -0.5546875, |
| "logits/rejected": -0.5703125, |
| "logps/chosen": -173.0, |
| "logps/rejected": -170.0, |
| "loss": 0.6639, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": -0.166015625, |
| "rewards/margins": 0.0703125, |
| "rewards/rejected": -0.236328125, |
| "step": 62 |
| }, |
| { |
| "epoch": 0.23053173241852487, |
| "grad_norm": 312.3990478515625, |
| "learning_rate": 1.5384615384615385e-06, |
| "logits/chosen": -0.58984375, |
| "logits/rejected": -0.58984375, |
| "logps/chosen": -160.0, |
| "logps/rejected": -205.0, |
| "loss": 0.6407, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.1416015625, |
| "rewards/margins": 0.1416015625, |
| "rewards/rejected": -0.283203125, |
| "step": 63 |
| }, |
| { |
| "epoch": 0.23419096626643796, |
| "grad_norm": 374.50360107421875, |
| "learning_rate": 1.5311355311355309e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -229.0, |
| "logps/rejected": -189.0, |
| "loss": 0.6736, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.1962890625, |
| "rewards/margins": 0.056640625, |
| "rewards/rejected": -0.251953125, |
| "step": 64 |
| }, |
| { |
| "epoch": 0.23785020011435107, |
| "grad_norm": 322.1225280761719, |
| "learning_rate": 1.5238095238095236e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -168.0, |
| "logps/rejected": -151.0, |
| "loss": 0.6749, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.125, |
| "rewards/margins": 0.0390625, |
| "rewards/rejected": -0.1640625, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.24150943396226415, |
| "grad_norm": 322.7266845703125, |
| "learning_rate": 1.5164835164835164e-06, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -165.0, |
| "logps/rejected": -178.0, |
| "loss": 0.6255, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0673828125, |
| "rewards/margins": 0.1455078125, |
| "rewards/rejected": -0.212890625, |
| "step": 66 |
| }, |
| { |
| "epoch": 0.24516866781017724, |
| "grad_norm": 317.9302978515625, |
| "learning_rate": 1.509157509157509e-06, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -163.0, |
| "logps/rejected": -211.0, |
| "loss": 0.6194, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.11279296875, |
| "rewards/margins": 0.158203125, |
| "rewards/rejected": -0.26953125, |
| "step": 67 |
| }, |
| { |
| "epoch": 0.24882790165809035, |
| "grad_norm": 328.2178955078125, |
| "learning_rate": 1.5018315018315017e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -162.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6609, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.1513671875, |
| "rewards/margins": 0.0888671875, |
| "rewards/rejected": -0.2412109375, |
| "step": 68 |
| }, |
| { |
| "epoch": 0.25248713550600344, |
| "grad_norm": 350.5845642089844, |
| "learning_rate": 1.4945054945054945e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -188.0, |
| "logps/rejected": -182.0, |
| "loss": 0.6567, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1591796875, |
| "rewards/margins": 0.08056640625, |
| "rewards/rejected": -0.2392578125, |
| "step": 69 |
| }, |
| { |
| "epoch": 0.25614636935391655, |
| "grad_norm": 329.70184326171875, |
| "learning_rate": 1.487179487179487e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -162.0, |
| "logps/rejected": -145.0, |
| "loss": 0.6819, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.158203125, |
| "rewards/margins": 0.0281982421875, |
| "rewards/rejected": -0.1865234375, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.2598056032018296, |
| "grad_norm": 324.77587890625, |
| "learning_rate": 1.4798534798534798e-06, |
| "logits/chosen": -0.60546875, |
| "logits/rejected": -0.63671875, |
| "logps/chosen": -149.0, |
| "logps/rejected": -177.0, |
| "loss": 0.6274, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.119140625, |
| "rewards/margins": 0.15625, |
| "rewards/rejected": -0.275390625, |
| "step": 71 |
| }, |
| { |
| "epoch": 0.2634648370497427, |
| "grad_norm": 348.8105773925781, |
| "learning_rate": 1.4725274725274726e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -164.0, |
| "logps/rejected": -217.0, |
| "loss": 0.6568, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.1455078125, |
| "rewards/margins": 0.109375, |
| "rewards/rejected": -0.255859375, |
| "step": 72 |
| }, |
| { |
| "epoch": 0.26712407089765583, |
| "grad_norm": 293.3565368652344, |
| "learning_rate": 1.465201465201465e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -150.0, |
| "logps/rejected": -162.0, |
| "loss": 0.6266, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.15234375, |
| "rewards/margins": 0.1552734375, |
| "rewards/rejected": -0.30859375, |
| "step": 73 |
| }, |
| { |
| "epoch": 0.2707833047455689, |
| "grad_norm": 322.1363220214844, |
| "learning_rate": 1.4578754578754577e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -167.0, |
| "logps/rejected": -164.0, |
| "loss": 0.6566, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.150390625, |
| "rewards/margins": 0.08984375, |
| "rewards/rejected": -0.240234375, |
| "step": 74 |
| }, |
| { |
| "epoch": 0.274442538593482, |
| "grad_norm": 337.89581298828125, |
| "learning_rate": 1.4505494505494505e-06, |
| "logits/chosen": -0.625, |
| "logits/rejected": -0.62109375, |
| "logps/chosen": -166.0, |
| "logps/rejected": -174.0, |
| "loss": 0.6544, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1689453125, |
| "rewards/margins": 0.0869140625, |
| "rewards/rejected": -0.255859375, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.27810177244139506, |
| "grad_norm": 309.1400146484375, |
| "learning_rate": 1.443223443223443e-06, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -143.0, |
| "logps/rejected": -160.0, |
| "loss": 0.6221, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.10498046875, |
| "rewards/margins": 0.171875, |
| "rewards/rejected": -0.27734375, |
| "step": 76 |
| }, |
| { |
| "epoch": 0.28176100628930817, |
| "grad_norm": 338.5057373046875, |
| "learning_rate": 1.4358974358974359e-06, |
| "logits/chosen": -0.625, |
| "logits/rejected": -0.59765625, |
| "logps/chosen": -139.0, |
| "logps/rejected": -171.0, |
| "loss": 0.6537, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1572265625, |
| "rewards/margins": 0.0947265625, |
| "rewards/rejected": -0.251953125, |
| "step": 77 |
| }, |
| { |
| "epoch": 0.2854202401372213, |
| "grad_norm": 332.32794189453125, |
| "learning_rate": 1.4285714285714286e-06, |
| "logits/chosen": -0.609375, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -154.0, |
| "logps/rejected": -185.0, |
| "loss": 0.6379, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.142578125, |
| "rewards/margins": 0.126953125, |
| "rewards/rejected": -0.26953125, |
| "step": 78 |
| }, |
| { |
| "epoch": 0.28907947398513434, |
| "grad_norm": 333.6225280761719, |
| "learning_rate": 1.4212454212454212e-06, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -151.0, |
| "logps/rejected": -183.0, |
| "loss": 0.6387, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1298828125, |
| "rewards/margins": 0.1328125, |
| "rewards/rejected": -0.263671875, |
| "step": 79 |
| }, |
| { |
| "epoch": 0.29273870783304745, |
| "grad_norm": 302.85931396484375, |
| "learning_rate": 1.413919413919414e-06, |
| "logits/chosen": -0.6015625, |
| "logits/rejected": -0.6015625, |
| "logps/chosen": -153.0, |
| "logps/rejected": -169.0, |
| "loss": 0.6433, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.1708984375, |
| "rewards/margins": 0.111328125, |
| "rewards/rejected": -0.28125, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.29639794168096056, |
| "grad_norm": 329.51885986328125, |
| "learning_rate": 1.4065934065934067e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -158.0, |
| "logps/rejected": -162.0, |
| "loss": 0.6799, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.1708984375, |
| "rewards/margins": 0.034423828125, |
| "rewards/rejected": -0.205078125, |
| "step": 81 |
| }, |
| { |
| "epoch": 0.3000571755288736, |
| "grad_norm": 311.5542907714844, |
| "learning_rate": 1.399267399267399e-06, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -131.0, |
| "logps/rejected": -158.0, |
| "loss": 0.6389, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1611328125, |
| "rewards/margins": 0.140625, |
| "rewards/rejected": -0.302734375, |
| "step": 82 |
| }, |
| { |
| "epoch": 0.30371640937678673, |
| "grad_norm": 442.2575378417969, |
| "learning_rate": 1.3919413919413919e-06, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.625, |
| "logps/chosen": -158.0, |
| "logps/rejected": -195.0, |
| "loss": 0.6479, |
| "rewards/accuracies": 0.53125, |
| "rewards/chosen": -0.173828125, |
| "rewards/margins": 0.11279296875, |
| "rewards/rejected": -0.287109375, |
| "step": 83 |
| }, |
| { |
| "epoch": 0.30737564322469985, |
| "grad_norm": 341.3179016113281, |
| "learning_rate": 1.3846153846153844e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -144.0, |
| "logps/rejected": -176.0, |
| "loss": 0.6302, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.126953125, |
| "rewards/margins": 0.1474609375, |
| "rewards/rejected": -0.2734375, |
| "step": 84 |
| }, |
| { |
| "epoch": 0.3110348770726129, |
| "grad_norm": 305.62109375, |
| "learning_rate": 1.3772893772893772e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -140.0, |
| "logps/rejected": -159.0, |
| "loss": 0.6205, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.0751953125, |
| "rewards/margins": 0.158203125, |
| "rewards/rejected": -0.2333984375, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.314694110920526, |
| "grad_norm": 300.09014892578125, |
| "learning_rate": 1.36996336996337e-06, |
| "logits/chosen": -0.59765625, |
| "logits/rejected": -0.578125, |
| "logps/chosen": -142.0, |
| "logps/rejected": -157.0, |
| "loss": 0.5966, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.038330078125, |
| "rewards/margins": 0.2138671875, |
| "rewards/rejected": -0.251953125, |
| "step": 86 |
| }, |
| { |
| "epoch": 0.31835334476843913, |
| "grad_norm": 325.3841247558594, |
| "learning_rate": 1.3626373626373625e-06, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -164.0, |
| "logps/rejected": -170.0, |
| "loss": 0.6437, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.10546875, |
| "rewards/margins": 0.12060546875, |
| "rewards/rejected": -0.2255859375, |
| "step": 87 |
| }, |
| { |
| "epoch": 0.3220125786163522, |
| "grad_norm": 347.8658447265625, |
| "learning_rate": 1.3553113553113553e-06, |
| "logits/chosen": -0.703125, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -154.0, |
| "logps/rejected": -167.0, |
| "loss": 0.6111, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.109375, |
| "rewards/margins": 0.185546875, |
| "rewards/rejected": -0.294921875, |
| "step": 88 |
| }, |
| { |
| "epoch": 0.3256718124642653, |
| "grad_norm": 319.26513671875, |
| "learning_rate": 1.347985347985348e-06, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.66015625, |
| "logps/chosen": -140.0, |
| "logps/rejected": -184.0, |
| "loss": 0.6299, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.1279296875, |
| "rewards/margins": 0.1611328125, |
| "rewards/rejected": -0.2890625, |
| "step": 89 |
| }, |
| { |
| "epoch": 0.3293310463121784, |
| "grad_norm": 326.7582702636719, |
| "learning_rate": 1.3406593406593404e-06, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -159.0, |
| "logps/rejected": -178.0, |
| "loss": 0.6315, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1376953125, |
| "rewards/margins": 0.1640625, |
| "rewards/rejected": -0.302734375, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.33299028016009147, |
| "grad_norm": 296.3334655761719, |
| "learning_rate": 1.3333333333333332e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -127.5, |
| "logps/rejected": -174.0, |
| "loss": 0.6214, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.158203125, |
| "rewards/margins": 0.1767578125, |
| "rewards/rejected": -0.333984375, |
| "step": 91 |
| }, |
| { |
| "epoch": 0.3366495140080046, |
| "grad_norm": 352.0799865722656, |
| "learning_rate": 1.326007326007326e-06, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.62109375, |
| "logps/chosen": -165.0, |
| "logps/rejected": -151.0, |
| "loss": 0.6456, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.1826171875, |
| "rewards/margins": 0.111328125, |
| "rewards/rejected": -0.294921875, |
| "step": 92 |
| }, |
| { |
| "epoch": 0.3403087478559177, |
| "grad_norm": 342.6123962402344, |
| "learning_rate": 1.3186813186813185e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -153.0, |
| "logps/rejected": -198.0, |
| "loss": 0.6411, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.1748046875, |
| "rewards/margins": 0.1318359375, |
| "rewards/rejected": -0.306640625, |
| "step": 93 |
| }, |
| { |
| "epoch": 0.34396798170383075, |
| "grad_norm": 400.05133056640625, |
| "learning_rate": 1.3113553113553113e-06, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -154.0, |
| "logps/rejected": -185.0, |
| "loss": 0.6475, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.1455078125, |
| "rewards/margins": 0.10595703125, |
| "rewards/rejected": -0.251953125, |
| "step": 94 |
| }, |
| { |
| "epoch": 0.34762721555174386, |
| "grad_norm": 314.9571838378906, |
| "learning_rate": 1.304029304029304e-06, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -160.0, |
| "logps/rejected": -175.0, |
| "loss": 0.6207, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.169921875, |
| "rewards/margins": 0.1650390625, |
| "rewards/rejected": -0.333984375, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.3512864493996569, |
| "grad_norm": 333.59881591796875, |
| "learning_rate": 1.2967032967032966e-06, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.703125, |
| "logps/chosen": -143.0, |
| "logps/rejected": -218.0, |
| "loss": 0.617, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.126953125, |
| "rewards/margins": 0.1826171875, |
| "rewards/rejected": -0.310546875, |
| "step": 96 |
| }, |
| { |
| "epoch": 0.35494568324757003, |
| "grad_norm": 356.0664978027344, |
| "learning_rate": 1.2893772893772894e-06, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -164.0, |
| "logps/rejected": -185.0, |
| "loss": 0.6273, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1669921875, |
| "rewards/margins": 0.1865234375, |
| "rewards/rejected": -0.353515625, |
| "step": 97 |
| }, |
| { |
| "epoch": 0.35860491709548314, |
| "grad_norm": 354.0408630371094, |
| "learning_rate": 1.2820512820512822e-06, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -145.0, |
| "logps/rejected": -165.0, |
| "loss": 0.6429, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1494140625, |
| "rewards/margins": 0.119140625, |
| "rewards/rejected": -0.267578125, |
| "step": 98 |
| }, |
| { |
| "epoch": 0.3622641509433962, |
| "grad_norm": 350.7596130371094, |
| "learning_rate": 1.2747252747252745e-06, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.61328125, |
| "logps/chosen": -186.0, |
| "logps/rejected": -204.0, |
| "loss": 0.5828, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.140625, |
| "rewards/margins": 0.25, |
| "rewards/rejected": -0.390625, |
| "step": 99 |
| }, |
| { |
| "epoch": 0.3659233847913093, |
| "grad_norm": 314.3235778808594, |
| "learning_rate": 1.2673992673992673e-06, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -159.0, |
| "logps/rejected": -175.0, |
| "loss": 0.6156, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1953125, |
| "rewards/margins": 0.1845703125, |
| "rewards/rejected": -0.380859375, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.3695826186392224, |
| "grad_norm": 308.3187561035156, |
| "learning_rate": 1.2600732600732599e-06, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -152.0, |
| "logps/rejected": -206.0, |
| "loss": 0.6235, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1728515625, |
| "rewards/margins": 0.16015625, |
| "rewards/rejected": -0.333984375, |
| "step": 101 |
| }, |
| { |
| "epoch": 0.3732418524871355, |
| "grad_norm": 339.0077209472656, |
| "learning_rate": 1.2527472527472527e-06, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -150.0, |
| "logps/rejected": -198.0, |
| "loss": 0.6145, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1669921875, |
| "rewards/margins": 0.1796875, |
| "rewards/rejected": -0.34765625, |
| "step": 102 |
| }, |
| { |
| "epoch": 0.3769010863350486, |
| "grad_norm": 333.14642333984375, |
| "learning_rate": 1.2454212454212454e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.63671875, |
| "logps/chosen": -160.0, |
| "logps/rejected": -172.0, |
| "loss": 0.5947, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.125, |
| "rewards/margins": 0.2265625, |
| "rewards/rejected": -0.3515625, |
| "step": 103 |
| }, |
| { |
| "epoch": 0.3805603201829617, |
| "grad_norm": 327.3541564941406, |
| "learning_rate": 1.238095238095238e-06, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -173.0, |
| "logps/rejected": -195.0, |
| "loss": 0.6166, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.1904296875, |
| "rewards/margins": 0.1796875, |
| "rewards/rejected": -0.37109375, |
| "step": 104 |
| }, |
| { |
| "epoch": 0.38421955403087477, |
| "grad_norm": 353.1539611816406, |
| "learning_rate": 1.2307692307692308e-06, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.66015625, |
| "logps/chosen": -157.0, |
| "logps/rejected": -188.0, |
| "loss": 0.6166, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.2158203125, |
| "rewards/margins": 0.189453125, |
| "rewards/rejected": -0.40625, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.3878787878787879, |
| "grad_norm": 338.275390625, |
| "learning_rate": 1.2234432234432235e-06, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -150.0, |
| "logps/rejected": -162.0, |
| "loss": 0.5969, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1337890625, |
| "rewards/margins": 0.2177734375, |
| "rewards/rejected": -0.3515625, |
| "step": 106 |
| }, |
| { |
| "epoch": 0.391538021726701, |
| "grad_norm": 338.1829528808594, |
| "learning_rate": 1.2161172161172159e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -157.0, |
| "logps/rejected": -145.0, |
| "loss": 0.645, |
| "rewards/accuracies": 0.5, |
| "rewards/chosen": -0.1943359375, |
| "rewards/margins": 0.10546875, |
| "rewards/rejected": -0.30078125, |
| "step": 107 |
| }, |
| { |
| "epoch": 0.39519725557461405, |
| "grad_norm": 334.7667236328125, |
| "learning_rate": 1.2087912087912087e-06, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -132.0, |
| "logps/rejected": -213.0, |
| "loss": 0.5587, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.138671875, |
| "rewards/margins": 0.314453125, |
| "rewards/rejected": -0.453125, |
| "step": 108 |
| }, |
| { |
| "epoch": 0.39885648942252716, |
| "grad_norm": 328.79302978515625, |
| "learning_rate": 1.2014652014652014e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -180.0, |
| "logps/rejected": -173.0, |
| "loss": 0.6177, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.205078125, |
| "rewards/margins": 0.189453125, |
| "rewards/rejected": -0.39453125, |
| "step": 109 |
| }, |
| { |
| "epoch": 0.4025157232704403, |
| "grad_norm": 397.0133056640625, |
| "learning_rate": 1.194139194139194e-06, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -157.0, |
| "logps/rejected": -169.0, |
| "loss": 0.6269, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.203125, |
| "rewards/margins": 0.171875, |
| "rewards/rejected": -0.375, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.40617495711835333, |
| "grad_norm": 333.8108825683594, |
| "learning_rate": 1.1868131868131868e-06, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -176.0, |
| "logps/rejected": -204.0, |
| "loss": 0.6445, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.2177734375, |
| "rewards/margins": 0.1220703125, |
| "rewards/rejected": -0.33984375, |
| "step": 111 |
| }, |
| { |
| "epoch": 0.40983419096626644, |
| "grad_norm": 340.60955810546875, |
| "learning_rate": 1.1794871794871795e-06, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -152.0, |
| "logps/rejected": -188.0, |
| "loss": 0.6043, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.1669921875, |
| "rewards/margins": 0.212890625, |
| "rewards/rejected": -0.380859375, |
| "step": 112 |
| }, |
| { |
| "epoch": 0.41349342481417956, |
| "grad_norm": 338.4276123046875, |
| "learning_rate": 1.172161172161172e-06, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.65625, |
| "logps/chosen": -145.0, |
| "logps/rejected": -161.0, |
| "loss": 0.6066, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.1865234375, |
| "rewards/margins": 0.203125, |
| "rewards/rejected": -0.388671875, |
| "step": 113 |
| }, |
| { |
| "epoch": 0.4171526586620926, |
| "grad_norm": 357.9419250488281, |
| "learning_rate": 1.1648351648351649e-06, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -176.0, |
| "logps/rejected": -196.0, |
| "loss": 0.6153, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.291015625, |
| "rewards/margins": 0.189453125, |
| "rewards/rejected": -0.48046875, |
| "step": 114 |
| }, |
| { |
| "epoch": 0.4208118925100057, |
| "grad_norm": 319.607177734375, |
| "learning_rate": 1.1575091575091577e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.625, |
| "logps/chosen": -153.0, |
| "logps/rejected": -165.0, |
| "loss": 0.5956, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1865234375, |
| "rewards/margins": 0.2109375, |
| "rewards/rejected": -0.396484375, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.42447112635791884, |
| "grad_norm": 332.0836181640625, |
| "learning_rate": 1.15018315018315e-06, |
| "logits/chosen": -0.6953125, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -160.0, |
| "logps/rejected": -176.0, |
| "loss": 0.6059, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1767578125, |
| "rewards/margins": 0.2099609375, |
| "rewards/rejected": -0.38671875, |
| "step": 116 |
| }, |
| { |
| "epoch": 0.4281303602058319, |
| "grad_norm": 327.6258544921875, |
| "learning_rate": 1.1428571428571428e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.734375, |
| "logps/chosen": -152.0, |
| "logps/rejected": -161.0, |
| "loss": 0.639, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.2177734375, |
| "rewards/margins": 0.1357421875, |
| "rewards/rejected": -0.353515625, |
| "step": 117 |
| }, |
| { |
| "epoch": 0.431789594053745, |
| "grad_norm": 348.0113525390625, |
| "learning_rate": 1.1355311355311353e-06, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -180.0, |
| "logps/rejected": -178.0, |
| "loss": 0.6324, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.302734375, |
| "rewards/margins": 0.1611328125, |
| "rewards/rejected": -0.46484375, |
| "step": 118 |
| }, |
| { |
| "epoch": 0.43544882790165806, |
| "grad_norm": 323.5909729003906, |
| "learning_rate": 1.1282051282051281e-06, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -171.0, |
| "logps/rejected": -162.0, |
| "loss": 0.6286, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.205078125, |
| "rewards/margins": 0.1630859375, |
| "rewards/rejected": -0.3671875, |
| "step": 119 |
| }, |
| { |
| "epoch": 0.4391080617495712, |
| "grad_norm": 356.1521301269531, |
| "learning_rate": 1.1208791208791209e-06, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -141.0, |
| "logps/rejected": -170.0, |
| "loss": 0.6358, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.224609375, |
| "rewards/margins": 0.142578125, |
| "rewards/rejected": -0.3671875, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.4427672955974843, |
| "grad_norm": 323.0431213378906, |
| "learning_rate": 1.1135531135531134e-06, |
| "logits/chosen": -0.58984375, |
| "logits/rejected": -0.63671875, |
| "logps/chosen": -149.0, |
| "logps/rejected": -205.0, |
| "loss": 0.5646, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.1533203125, |
| "rewards/margins": 0.314453125, |
| "rewards/rejected": -0.466796875, |
| "step": 121 |
| }, |
| { |
| "epoch": 0.44642652944539735, |
| "grad_norm": 394.70880126953125, |
| "learning_rate": 1.1062271062271062e-06, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.734375, |
| "logps/chosen": -145.0, |
| "logps/rejected": -171.0, |
| "loss": 0.6443, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.2138671875, |
| "rewards/margins": 0.1171875, |
| "rewards/rejected": -0.33203125, |
| "step": 122 |
| }, |
| { |
| "epoch": 0.45008576329331046, |
| "grad_norm": 361.07220458984375, |
| "learning_rate": 1.098901098901099e-06, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -174.0, |
| "logps/rejected": -179.0, |
| "loss": 0.6489, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.279296875, |
| "rewards/margins": 0.1123046875, |
| "rewards/rejected": -0.390625, |
| "step": 123 |
| }, |
| { |
| "epoch": 0.45374499714122357, |
| "grad_norm": 378.9485778808594, |
| "learning_rate": 1.0915750915750913e-06, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -166.0, |
| "logps/rejected": -181.0, |
| "loss": 0.6277, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.2314453125, |
| "rewards/margins": 0.158203125, |
| "rewards/rejected": -0.388671875, |
| "step": 124 |
| }, |
| { |
| "epoch": 0.45740423098913663, |
| "grad_norm": 361.32470703125, |
| "learning_rate": 1.0842490842490841e-06, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.65625, |
| "logps/chosen": -162.0, |
| "logps/rejected": -201.0, |
| "loss": 0.6276, |
| "rewards/accuracies": 0.625, |
| "rewards/chosen": -0.2314453125, |
| "rewards/margins": 0.1826171875, |
| "rewards/rejected": -0.4140625, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.46106346483704974, |
| "grad_norm": 344.46905517578125, |
| "learning_rate": 1.0769230769230769e-06, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -175.0, |
| "logps/rejected": -169.0, |
| "loss": 0.6512, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.255859375, |
| "rewards/margins": 0.10400390625, |
| "rewards/rejected": -0.361328125, |
| "step": 126 |
| }, |
| { |
| "epoch": 0.46472269868496285, |
| "grad_norm": 350.5096130371094, |
| "learning_rate": 1.0695970695970695e-06, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.72265625, |
| "logps/chosen": -177.0, |
| "logps/rejected": -169.0, |
| "loss": 0.621, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.2265625, |
| "rewards/margins": 0.185546875, |
| "rewards/rejected": -0.412109375, |
| "step": 127 |
| }, |
| { |
| "epoch": 0.4683819325328759, |
| "grad_norm": 321.7780456542969, |
| "learning_rate": 1.0622710622710622e-06, |
| "logits/chosen": -0.6953125, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -144.0, |
| "logps/rejected": -200.0, |
| "loss": 0.5705, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.1669921875, |
| "rewards/margins": 0.3046875, |
| "rewards/rejected": -0.47265625, |
| "step": 128 |
| }, |
| { |
| "epoch": 0.472041166380789, |
| "grad_norm": 312.19964599609375, |
| "learning_rate": 1.054945054945055e-06, |
| "logits/chosen": -0.60546875, |
| "logits/rejected": -0.58203125, |
| "logps/chosen": -150.0, |
| "logps/rejected": -210.0, |
| "loss": 0.5723, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1396484375, |
| "rewards/margins": 0.3359375, |
| "rewards/rejected": -0.474609375, |
| "step": 129 |
| }, |
| { |
| "epoch": 0.47570040022870214, |
| "grad_norm": 354.0895080566406, |
| "learning_rate": 1.0476190476190476e-06, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -166.0, |
| "logps/rejected": -170.0, |
| "loss": 0.6128, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.2890625, |
| "rewards/margins": 0.1875, |
| "rewards/rejected": -0.4765625, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.4793596340766152, |
| "grad_norm": 359.5901184082031, |
| "learning_rate": 1.0402930402930403e-06, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -155.0, |
| "logps/rejected": -184.0, |
| "loss": 0.6275, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1689453125, |
| "rewards/margins": 0.1669921875, |
| "rewards/rejected": -0.3359375, |
| "step": 131 |
| }, |
| { |
| "epoch": 0.4830188679245283, |
| "grad_norm": 325.01580810546875, |
| "learning_rate": 1.0329670329670331e-06, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.61328125, |
| "logps/chosen": -155.0, |
| "logps/rejected": -172.0, |
| "loss": 0.6331, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.2021484375, |
| "rewards/margins": 0.169921875, |
| "rewards/rejected": -0.37109375, |
| "step": 132 |
| }, |
| { |
| "epoch": 0.4866781017724414, |
| "grad_norm": 306.410888671875, |
| "learning_rate": 1.0256410256410255e-06, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -146.0, |
| "logps/rejected": -181.0, |
| "loss": 0.5833, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.1416015625, |
| "rewards/margins": 0.248046875, |
| "rewards/rejected": -0.388671875, |
| "step": 133 |
| }, |
| { |
| "epoch": 0.4903373356203545, |
| "grad_norm": 330.1487121582031, |
| "learning_rate": 1.0183150183150182e-06, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -176.0, |
| "logps/rejected": -228.0, |
| "loss": 0.5701, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.2099609375, |
| "rewards/margins": 0.314453125, |
| "rewards/rejected": -0.5234375, |
| "step": 134 |
| }, |
| { |
| "epoch": 0.4939965694682676, |
| "grad_norm": 355.6802673339844, |
| "learning_rate": 1.010989010989011e-06, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -166.0, |
| "logps/rejected": -146.0, |
| "loss": 0.6357, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.150390625, |
| "rewards/margins": 0.150390625, |
| "rewards/rejected": -0.30078125, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.4976558033161807, |
| "grad_norm": 325.06951904296875, |
| "learning_rate": 1.0036630036630036e-06, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.609375, |
| "logps/chosen": -151.0, |
| "logps/rejected": -168.0, |
| "loss": 0.62, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1455078125, |
| "rewards/margins": 0.17578125, |
| "rewards/rejected": -0.322265625, |
| "step": 136 |
| }, |
| { |
| "epoch": 0.5013150371640938, |
| "grad_norm": 342.5461120605469, |
| "learning_rate": 9.963369963369963e-07, |
| "logits/chosen": -0.6328125, |
| "logits/rejected": -0.625, |
| "logps/chosen": -179.0, |
| "logps/rejected": -178.0, |
| "loss": 0.6149, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.1533203125, |
| "rewards/margins": 0.1826171875, |
| "rewards/rejected": -0.3359375, |
| "step": 137 |
| }, |
| { |
| "epoch": 0.5049742710120069, |
| "grad_norm": 294.18218994140625, |
| "learning_rate": 9.89010989010989e-07, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -158.0, |
| "logps/rejected": -189.0, |
| "loss": 0.5771, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.1455078125, |
| "rewards/margins": 0.287109375, |
| "rewards/rejected": -0.43359375, |
| "step": 138 |
| }, |
| { |
| "epoch": 0.50863350485992, |
| "grad_norm": 349.0784606933594, |
| "learning_rate": 9.816849816849817e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.6171875, |
| "logps/chosen": -141.0, |
| "logps/rejected": -147.0, |
| "loss": 0.6296, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.08154296875, |
| "rewards/margins": 0.15234375, |
| "rewards/rejected": -0.2333984375, |
| "step": 139 |
| }, |
| { |
| "epoch": 0.5122927387078331, |
| "grad_norm": 296.64013671875, |
| "learning_rate": 9.743589743589742e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -142.0, |
| "logps/rejected": -157.0, |
| "loss": 0.5943, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.10498046875, |
| "rewards/margins": 0.224609375, |
| "rewards/rejected": -0.330078125, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.5159519725557461, |
| "grad_norm": 366.21868896484375, |
| "learning_rate": 9.67032967032967e-07, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.61328125, |
| "logps/chosen": -158.0, |
| "logps/rejected": -163.0, |
| "loss": 0.6171, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.1376953125, |
| "rewards/margins": 0.1796875, |
| "rewards/rejected": -0.318359375, |
| "step": 141 |
| }, |
| { |
| "epoch": 0.5196112064036592, |
| "grad_norm": 356.66766357421875, |
| "learning_rate": 9.597069597069598e-07, |
| "logits/chosen": -0.625, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -137.0, |
| "logps/rejected": -187.0, |
| "loss": 0.5703, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.16015625, |
| "rewards/margins": 0.322265625, |
| "rewards/rejected": -0.482421875, |
| "step": 142 |
| }, |
| { |
| "epoch": 0.5232704402515723, |
| "grad_norm": 328.0158996582031, |
| "learning_rate": 9.523809523809522e-07, |
| "logits/chosen": -0.625, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -176.0, |
| "logps/rejected": -209.0, |
| "loss": 0.6207, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.244140625, |
| "rewards/margins": 0.1884765625, |
| "rewards/rejected": -0.43359375, |
| "step": 143 |
| }, |
| { |
| "epoch": 0.5269296740994854, |
| "grad_norm": 343.86138916015625, |
| "learning_rate": 9.45054945054945e-07, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.703125, |
| "logps/chosen": -152.0, |
| "logps/rejected": -203.0, |
| "loss": 0.5776, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.291015625, |
| "rewards/rejected": -0.3359375, |
| "step": 144 |
| }, |
| { |
| "epoch": 0.5305889079473985, |
| "grad_norm": 339.378662109375, |
| "learning_rate": 9.377289377289377e-07, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.61328125, |
| "logps/chosen": -133.0, |
| "logps/rejected": -168.0, |
| "loss": 0.5869, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.0908203125, |
| "rewards/margins": 0.2451171875, |
| "rewards/rejected": -0.3359375, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.5342481417953117, |
| "grad_norm": 357.1183166503906, |
| "learning_rate": 9.304029304029304e-07, |
| "logits/chosen": -0.72265625, |
| "logits/rejected": -0.71484375, |
| "logps/chosen": -152.0, |
| "logps/rejected": -175.0, |
| "loss": 0.6175, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.134765625, |
| "rewards/margins": 0.197265625, |
| "rewards/rejected": -0.33203125, |
| "step": 146 |
| }, |
| { |
| "epoch": 0.5379073756432247, |
| "grad_norm": 315.3673095703125, |
| "learning_rate": 9.230769230769231e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -151.0, |
| "logps/rejected": -154.0, |
| "loss": 0.5938, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.08837890625, |
| "rewards/margins": 0.228515625, |
| "rewards/rejected": -0.31640625, |
| "step": 147 |
| }, |
| { |
| "epoch": 0.5415666094911378, |
| "grad_norm": 378.76580810546875, |
| "learning_rate": 9.157509157509157e-07, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.66015625, |
| "logps/chosen": -163.0, |
| "logps/rejected": -184.0, |
| "loss": 0.581, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.087890625, |
| "rewards/margins": 0.263671875, |
| "rewards/rejected": -0.3515625, |
| "step": 148 |
| }, |
| { |
| "epoch": 0.5452258433390509, |
| "grad_norm": 315.32427978515625, |
| "learning_rate": 9.084249084249084e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -160.0, |
| "logps/rejected": -180.0, |
| "loss": 0.615, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.12060546875, |
| "rewards/margins": 0.1826171875, |
| "rewards/rejected": -0.302734375, |
| "step": 149 |
| }, |
| { |
| "epoch": 0.548885077186964, |
| "grad_norm": 302.259521484375, |
| "learning_rate": 9.010989010989011e-07, |
| "logits/chosen": -0.62890625, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -147.0, |
| "logps/rejected": -189.0, |
| "loss": 0.5753, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.0634765625, |
| "rewards/margins": 0.291015625, |
| "rewards/rejected": -0.35546875, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.5525443110348771, |
| "grad_norm": 327.5895690917969, |
| "learning_rate": 8.937728937728938e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -168.0, |
| "logps/rejected": -167.0, |
| "loss": 0.5899, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.0654296875, |
| "rewards/margins": 0.23828125, |
| "rewards/rejected": -0.302734375, |
| "step": 151 |
| }, |
| { |
| "epoch": 0.5562035448827901, |
| "grad_norm": 323.4364929199219, |
| "learning_rate": 8.864468864468864e-07, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -150.0, |
| "logps/rejected": -155.0, |
| "loss": 0.6307, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.07958984375, |
| "rewards/margins": 0.1435546875, |
| "rewards/rejected": -0.2236328125, |
| "step": 152 |
| }, |
| { |
| "epoch": 0.5598627787307032, |
| "grad_norm": 307.4286804199219, |
| "learning_rate": 8.79120879120879e-07, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.65625, |
| "logps/chosen": -136.0, |
| "logps/rejected": -162.0, |
| "loss": 0.5807, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.09375, |
| "rewards/margins": 0.26953125, |
| "rewards/rejected": -0.36328125, |
| "step": 153 |
| }, |
| { |
| "epoch": 0.5635220125786163, |
| "grad_norm": 307.41644287109375, |
| "learning_rate": 8.717948717948718e-07, |
| "logits/chosen": -0.71484375, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -136.0, |
| "logps/rejected": -173.0, |
| "loss": 0.5708, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.0546875, |
| "rewards/margins": 0.283203125, |
| "rewards/rejected": -0.337890625, |
| "step": 154 |
| }, |
| { |
| "epoch": 0.5671812464265295, |
| "grad_norm": 329.59661865234375, |
| "learning_rate": 8.644688644688645e-07, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -143.0, |
| "logps/rejected": -196.0, |
| "loss": 0.6014, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1669921875, |
| "rewards/margins": 0.2255859375, |
| "rewards/rejected": -0.392578125, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.5708404802744426, |
| "grad_norm": 330.7326354980469, |
| "learning_rate": 8.57142857142857e-07, |
| "logits/chosen": -0.5859375, |
| "logits/rejected": -0.609375, |
| "logps/chosen": -154.0, |
| "logps/rejected": -199.0, |
| "loss": 0.5914, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.09130859375, |
| "rewards/margins": 0.265625, |
| "rewards/rejected": -0.357421875, |
| "step": 156 |
| }, |
| { |
| "epoch": 0.5744997141223557, |
| "grad_norm": 335.531494140625, |
| "learning_rate": 8.498168498168498e-07, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -145.0, |
| "logps/rejected": -178.0, |
| "loss": 0.59, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.018798828125, |
| "rewards/margins": 0.259765625, |
| "rewards/rejected": -0.27734375, |
| "step": 157 |
| }, |
| { |
| "epoch": 0.5781589479702687, |
| "grad_norm": 324.5040588378906, |
| "learning_rate": 8.424908424908425e-07, |
| "logits/chosen": -0.71875, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -159.0, |
| "logps/rejected": -193.0, |
| "loss": 0.575, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.03125, |
| "rewards/margins": 0.279296875, |
| "rewards/rejected": -0.310546875, |
| "step": 158 |
| }, |
| { |
| "epoch": 0.5818181818181818, |
| "grad_norm": 370.49615478515625, |
| "learning_rate": 8.351648351648351e-07, |
| "logits/chosen": -0.71484375, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -158.0, |
| "logps/rejected": -142.0, |
| "loss": 0.6056, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.0830078125, |
| "rewards/margins": 0.2099609375, |
| "rewards/rejected": -0.29296875, |
| "step": 159 |
| }, |
| { |
| "epoch": 0.5854774156660949, |
| "grad_norm": 320.1568603515625, |
| "learning_rate": 8.278388278388277e-07, |
| "logits/chosen": -0.7109375, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -161.0, |
| "logps/rejected": -146.0, |
| "loss": 0.6121, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.053955078125, |
| "rewards/margins": 0.1904296875, |
| "rewards/rejected": -0.244140625, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.589136649514008, |
| "grad_norm": 313.9174499511719, |
| "learning_rate": 8.205128205128205e-07, |
| "logits/chosen": -0.62109375, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -153.0, |
| "logps/rejected": -178.0, |
| "loss": 0.6049, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.08984375, |
| "rewards/margins": 0.2197265625, |
| "rewards/rejected": -0.30859375, |
| "step": 161 |
| }, |
| { |
| "epoch": 0.5927958833619211, |
| "grad_norm": 373.2075500488281, |
| "learning_rate": 8.131868131868131e-07, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -159.0, |
| "logps/rejected": -190.0, |
| "loss": 0.5909, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.03759765625, |
| "rewards/margins": 0.2412109375, |
| "rewards/rejected": -0.279296875, |
| "step": 162 |
| }, |
| { |
| "epoch": 0.5964551172098342, |
| "grad_norm": 335.1282958984375, |
| "learning_rate": 8.058608058608058e-07, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -164.0, |
| "logps/rejected": -181.0, |
| "loss": 0.5697, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.296875, |
| "rewards/rejected": -0.341796875, |
| "step": 163 |
| }, |
| { |
| "epoch": 0.6001143510577472, |
| "grad_norm": 335.8233642578125, |
| "learning_rate": 7.985347985347986e-07, |
| "logits/chosen": -0.7421875, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -146.0, |
| "logps/rejected": -187.0, |
| "loss": 0.5668, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": 0.02197265625, |
| "rewards/margins": 0.306640625, |
| "rewards/rejected": -0.28515625, |
| "step": 164 |
| }, |
| { |
| "epoch": 0.6037735849056604, |
| "grad_norm": 321.95721435546875, |
| "learning_rate": 7.912087912087911e-07, |
| "logits/chosen": -0.64453125, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -156.0, |
| "logps/rejected": -180.0, |
| "loss": 0.5832, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.08984375, |
| "rewards/margins": 0.2890625, |
| "rewards/rejected": -0.380859375, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.6074328187535735, |
| "grad_norm": 311.0093078613281, |
| "learning_rate": 7.838827838827838e-07, |
| "logits/chosen": -0.71875, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -143.0, |
| "logps/rejected": -170.0, |
| "loss": 0.5898, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.07373046875, |
| "rewards/margins": 0.244140625, |
| "rewards/rejected": -0.318359375, |
| "step": 166 |
| }, |
| { |
| "epoch": 0.6110920526014866, |
| "grad_norm": 370.56341552734375, |
| "learning_rate": 7.765567765567766e-07, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -162.0, |
| "logps/rejected": -193.0, |
| "loss": 0.6026, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.12060546875, |
| "rewards/margins": 0.2216796875, |
| "rewards/rejected": -0.341796875, |
| "step": 167 |
| }, |
| { |
| "epoch": 0.6147512864493997, |
| "grad_norm": 338.77239990234375, |
| "learning_rate": 7.692307692307693e-07, |
| "logits/chosen": -0.7109375, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -149.0, |
| "logps/rejected": -163.0, |
| "loss": 0.5883, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.032958984375, |
| "rewards/margins": 0.2412109375, |
| "rewards/rejected": -0.2734375, |
| "step": 168 |
| }, |
| { |
| "epoch": 0.6184105202973128, |
| "grad_norm": 408.2104187011719, |
| "learning_rate": 7.619047619047618e-07, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.65234375, |
| "logps/chosen": -187.0, |
| "logps/rejected": -174.0, |
| "loss": 0.6561, |
| "rewards/accuracies": 0.5625, |
| "rewards/chosen": -0.1357421875, |
| "rewards/margins": 0.0947265625, |
| "rewards/rejected": -0.23046875, |
| "step": 169 |
| }, |
| { |
| "epoch": 0.6220697541452258, |
| "grad_norm": 321.3171691894531, |
| "learning_rate": 7.545787545787545e-07, |
| "logits/chosen": -0.7421875, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -164.0, |
| "logps/rejected": -182.0, |
| "loss": 0.5742, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.0281982421875, |
| "rewards/margins": 0.296875, |
| "rewards/rejected": -0.32421875, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.6257289879931389, |
| "grad_norm": 327.4810485839844, |
| "learning_rate": 7.472527472527473e-07, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -151.0, |
| "logps/rejected": -160.0, |
| "loss": 0.6061, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.07666015625, |
| "rewards/margins": 0.2021484375, |
| "rewards/rejected": -0.279296875, |
| "step": 171 |
| }, |
| { |
| "epoch": 0.629388221841052, |
| "grad_norm": 357.8804016113281, |
| "learning_rate": 7.399267399267399e-07, |
| "logits/chosen": -0.734375, |
| "logits/rejected": -0.703125, |
| "logps/chosen": -154.0, |
| "logps/rejected": -172.0, |
| "loss": 0.5742, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.050048828125, |
| "rewards/margins": 0.28515625, |
| "rewards/rejected": -0.333984375, |
| "step": 172 |
| }, |
| { |
| "epoch": 0.6330474556889651, |
| "grad_norm": 320.7285461425781, |
| "learning_rate": 7.326007326007325e-07, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -161.0, |
| "logps/rejected": -184.0, |
| "loss": 0.564, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0361328125, |
| "rewards/margins": 0.30859375, |
| "rewards/rejected": -0.345703125, |
| "step": 173 |
| }, |
| { |
| "epoch": 0.6367066895368783, |
| "grad_norm": 321.6654968261719, |
| "learning_rate": 7.252747252747253e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -145.0, |
| "logps/rejected": -153.0, |
| "loss": 0.6049, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.038330078125, |
| "rewards/margins": 0.22265625, |
| "rewards/rejected": -0.26171875, |
| "step": 174 |
| }, |
| { |
| "epoch": 0.6403659233847913, |
| "grad_norm": 354.5701599121094, |
| "learning_rate": 7.179487179487179e-07, |
| "logits/chosen": -0.6953125, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -180.0, |
| "logps/rejected": -174.0, |
| "loss": 0.5979, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.09521484375, |
| "rewards/margins": 0.2265625, |
| "rewards/rejected": -0.322265625, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.6440251572327044, |
| "grad_norm": 333.7059020996094, |
| "learning_rate": 7.106227106227106e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -157.0, |
| "logps/rejected": -198.0, |
| "loss": 0.5878, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.09228515625, |
| "rewards/margins": 0.24609375, |
| "rewards/rejected": -0.337890625, |
| "step": 176 |
| }, |
| { |
| "epoch": 0.6476843910806175, |
| "grad_norm": 325.05035400390625, |
| "learning_rate": 7.032967032967034e-07, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -146.0, |
| "logps/rejected": -186.0, |
| "loss": 0.5739, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.0008392333984375, |
| "rewards/margins": 0.3046875, |
| "rewards/rejected": -0.306640625, |
| "step": 177 |
| }, |
| { |
| "epoch": 0.6513436249285306, |
| "grad_norm": 352.8323974609375, |
| "learning_rate": 6.959706959706959e-07, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.640625, |
| "logps/chosen": -179.0, |
| "logps/rejected": -218.0, |
| "loss": 0.5719, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.10009765625, |
| "rewards/margins": 0.314453125, |
| "rewards/rejected": -0.4140625, |
| "step": 178 |
| }, |
| { |
| "epoch": 0.6550028587764437, |
| "grad_norm": 307.6293029785156, |
| "learning_rate": 6.886446886446886e-07, |
| "logits/chosen": -0.6328125, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -146.0, |
| "logps/rejected": -172.0, |
| "loss": 0.5661, |
| "rewards/accuracies": 0.9375, |
| "rewards/chosen": -0.046142578125, |
| "rewards/margins": 0.302734375, |
| "rewards/rejected": -0.349609375, |
| "step": 179 |
| }, |
| { |
| "epoch": 0.6586620926243568, |
| "grad_norm": 356.95587158203125, |
| "learning_rate": 6.813186813186813e-07, |
| "logits/chosen": -0.66015625, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -149.0, |
| "logps/rejected": -164.0, |
| "loss": 0.6138, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.10009765625, |
| "rewards/margins": 0.1865234375, |
| "rewards/rejected": -0.287109375, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.6623213264722698, |
| "grad_norm": 290.56317138671875, |
| "learning_rate": 6.73992673992674e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -136.0, |
| "logps/rejected": -171.0, |
| "loss": 0.5764, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.031982421875, |
| "rewards/margins": 0.29296875, |
| "rewards/rejected": -0.32421875, |
| "step": 181 |
| }, |
| { |
| "epoch": 0.6659805603201829, |
| "grad_norm": 313.93927001953125, |
| "learning_rate": 6.666666666666666e-07, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -145.0, |
| "logps/rejected": -164.0, |
| "loss": 0.5668, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.0015411376953125, |
| "rewards/margins": 0.287109375, |
| "rewards/rejected": -0.287109375, |
| "step": 182 |
| }, |
| { |
| "epoch": 0.669639794168096, |
| "grad_norm": 315.1253967285156, |
| "learning_rate": 6.593406593406593e-07, |
| "logits/chosen": -0.734375, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -125.5, |
| "logps/rejected": -232.0, |
| "loss": 0.5599, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.0186767578125, |
| "rewards/margins": 0.3515625, |
| "rewards/rejected": -0.333984375, |
| "step": 183 |
| }, |
| { |
| "epoch": 0.6732990280160092, |
| "grad_norm": 326.492431640625, |
| "learning_rate": 6.52014652014652e-07, |
| "logits/chosen": -0.62890625, |
| "logits/rejected": -0.61328125, |
| "logps/chosen": -162.0, |
| "logps/rejected": -222.0, |
| "loss": 0.5437, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.030517578125, |
| "rewards/margins": 0.36328125, |
| "rewards/rejected": -0.392578125, |
| "step": 184 |
| }, |
| { |
| "epoch": 0.6769582618639223, |
| "grad_norm": 348.7772521972656, |
| "learning_rate": 6.446886446886447e-07, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.6484375, |
| "logps/chosen": -167.0, |
| "logps/rejected": -177.0, |
| "loss": 0.5911, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.0947265625, |
| "rewards/margins": 0.2431640625, |
| "rewards/rejected": -0.337890625, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.6806174957118354, |
| "grad_norm": 336.41827392578125, |
| "learning_rate": 6.373626373626373e-07, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -150.0, |
| "logps/rejected": -175.0, |
| "loss": 0.5616, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.01165771484375, |
| "rewards/margins": 0.298828125, |
| "rewards/rejected": -0.310546875, |
| "step": 186 |
| }, |
| { |
| "epoch": 0.6842767295597484, |
| "grad_norm": 328.49530029296875, |
| "learning_rate": 6.300366300366299e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -146.0, |
| "logps/rejected": -151.0, |
| "loss": 0.5895, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.047607421875, |
| "rewards/margins": 0.232421875, |
| "rewards/rejected": -0.279296875, |
| "step": 187 |
| }, |
| { |
| "epoch": 0.6879359634076615, |
| "grad_norm": 351.6430969238281, |
| "learning_rate": 6.227106227106227e-07, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -148.0, |
| "logps/rejected": -218.0, |
| "loss": 0.5731, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1025390625, |
| "rewards/margins": 0.30859375, |
| "rewards/rejected": -0.41015625, |
| "step": 188 |
| }, |
| { |
| "epoch": 0.6915951972555746, |
| "grad_norm": 318.9908752441406, |
| "learning_rate": 6.153846153846154e-07, |
| "logits/chosen": -0.73828125, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -145.0, |
| "logps/rejected": -169.0, |
| "loss": 0.5258, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": -1.52587890625e-05, |
| "rewards/margins": 0.390625, |
| "rewards/rejected": -0.388671875, |
| "step": 189 |
| }, |
| { |
| "epoch": 0.6952544311034877, |
| "grad_norm": 342.9971618652344, |
| "learning_rate": 6.080586080586079e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -157.0, |
| "logps/rejected": -189.0, |
| "loss": 0.5966, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.07666015625, |
| "rewards/margins": 0.234375, |
| "rewards/rejected": -0.310546875, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.6989136649514008, |
| "grad_norm": 309.0221862792969, |
| "learning_rate": 6.007326007326007e-07, |
| "logits/chosen": -0.6484375, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -146.0, |
| "logps/rejected": -201.0, |
| "loss": 0.5492, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.04931640625, |
| "rewards/margins": 0.37109375, |
| "rewards/rejected": -0.419921875, |
| "step": 191 |
| }, |
| { |
| "epoch": 0.7025728987993138, |
| "grad_norm": 384.8427429199219, |
| "learning_rate": 5.934065934065934e-07, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -182.0, |
| "logps/rejected": -179.0, |
| "loss": 0.6272, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.12353515625, |
| "rewards/margins": 0.1796875, |
| "rewards/rejected": -0.302734375, |
| "step": 192 |
| }, |
| { |
| "epoch": 0.706232132647227, |
| "grad_norm": 345.20538330078125, |
| "learning_rate": 5.86080586080586e-07, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -152.0, |
| "logps/rejected": -150.0, |
| "loss": 0.6308, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.08447265625, |
| "rewards/margins": 0.1513671875, |
| "rewards/rejected": -0.236328125, |
| "step": 193 |
| }, |
| { |
| "epoch": 0.7098913664951401, |
| "grad_norm": 373.9530334472656, |
| "learning_rate": 5.787545787545788e-07, |
| "logits/chosen": -0.71484375, |
| "logits/rejected": -0.7265625, |
| "logps/chosen": -174.0, |
| "logps/rejected": -163.0, |
| "loss": 0.619, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.09228515625, |
| "rewards/margins": 0.1767578125, |
| "rewards/rejected": -0.26953125, |
| "step": 194 |
| }, |
| { |
| "epoch": 0.7135506003430532, |
| "grad_norm": 378.72406005859375, |
| "learning_rate": 5.714285714285714e-07, |
| "logits/chosen": -0.7109375, |
| "logits/rejected": -0.7109375, |
| "logps/chosen": -160.0, |
| "logps/rejected": -201.0, |
| "loss": 0.5354, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0311279296875, |
| "rewards/margins": 0.380859375, |
| "rewards/rejected": -0.412109375, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.7172098341909663, |
| "grad_norm": 326.7547912597656, |
| "learning_rate": 5.641025641025641e-07, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -152.0, |
| "logps/rejected": -164.0, |
| "loss": 0.594, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.2392578125, |
| "rewards/rejected": -0.28515625, |
| "step": 196 |
| }, |
| { |
| "epoch": 0.7208690680388794, |
| "grad_norm": 332.29193115234375, |
| "learning_rate": 5.567765567765567e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -152.0, |
| "logps/rejected": -191.0, |
| "loss": 0.5926, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.0703125, |
| "rewards/margins": 0.232421875, |
| "rewards/rejected": -0.302734375, |
| "step": 197 |
| }, |
| { |
| "epoch": 0.7245283018867924, |
| "grad_norm": 345.8672180175781, |
| "learning_rate": 5.494505494505495e-07, |
| "logits/chosen": -0.71484375, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -153.0, |
| "logps/rejected": -190.0, |
| "loss": 0.5678, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.07177734375, |
| "rewards/margins": 0.310546875, |
| "rewards/rejected": -0.3828125, |
| "step": 198 |
| }, |
| { |
| "epoch": 0.7281875357347055, |
| "grad_norm": 418.1891174316406, |
| "learning_rate": 5.421245421245421e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -167.0, |
| "logps/rejected": -182.0, |
| "loss": 0.5964, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.10302734375, |
| "rewards/margins": 0.2392578125, |
| "rewards/rejected": -0.341796875, |
| "step": 199 |
| }, |
| { |
| "epoch": 0.7318467695826186, |
| "grad_norm": 364.3031921386719, |
| "learning_rate": 5.347985347985347e-07, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -186.0, |
| "logps/rejected": -180.0, |
| "loss": 0.5733, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.058837890625, |
| "rewards/margins": 0.28515625, |
| "rewards/rejected": -0.34375, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.7355060034305317, |
| "grad_norm": 344.48663330078125, |
| "learning_rate": 5.274725274725275e-07, |
| "logits/chosen": -0.76171875, |
| "logits/rejected": -0.765625, |
| "logps/chosen": -141.0, |
| "logps/rejected": -156.0, |
| "loss": 0.571, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": -0.03125, |
| "rewards/margins": 0.275390625, |
| "rewards/rejected": -0.306640625, |
| "step": 201 |
| }, |
| { |
| "epoch": 0.7391652372784449, |
| "grad_norm": 341.8359069824219, |
| "learning_rate": 5.201465201465202e-07, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -152.0, |
| "logps/rejected": -169.0, |
| "loss": 0.5828, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.109375, |
| "rewards/margins": 0.27734375, |
| "rewards/rejected": -0.38671875, |
| "step": 202 |
| }, |
| { |
| "epoch": 0.742824471126358, |
| "grad_norm": 331.3113708496094, |
| "learning_rate": 5.128205128205127e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -160.0, |
| "logps/rejected": -174.0, |
| "loss": 0.5509, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.0211181640625, |
| "rewards/margins": 0.328125, |
| "rewards/rejected": -0.349609375, |
| "step": 203 |
| }, |
| { |
| "epoch": 0.746483704974271, |
| "grad_norm": 331.06451416015625, |
| "learning_rate": 5.054945054945055e-07, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.64453125, |
| "logps/chosen": -146.0, |
| "logps/rejected": -145.0, |
| "loss": 0.5928, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.050048828125, |
| "rewards/margins": 0.2333984375, |
| "rewards/rejected": -0.283203125, |
| "step": 204 |
| }, |
| { |
| "epoch": 0.7501429388221841, |
| "grad_norm": 386.7975769042969, |
| "learning_rate": 4.981684981684982e-07, |
| "logits/chosen": -0.640625, |
| "logits/rejected": -0.6328125, |
| "logps/chosen": -131.0, |
| "logps/rejected": -168.0, |
| "loss": 0.5466, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.050048828125, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.396484375, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.7538021726700972, |
| "grad_norm": 773.7296142578125, |
| "learning_rate": 4.908424908424908e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -143.0, |
| "logps/rejected": -178.0, |
| "loss": 0.6085, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1611328125, |
| "rewards/margins": 0.2138671875, |
| "rewards/rejected": -0.375, |
| "step": 206 |
| }, |
| { |
| "epoch": 0.7574614065180103, |
| "grad_norm": 327.77874755859375, |
| "learning_rate": 4.835164835164835e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.66015625, |
| "logps/chosen": -168.0, |
| "logps/rejected": -157.0, |
| "loss": 0.5521, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.040771484375, |
| "rewards/margins": 0.318359375, |
| "rewards/rejected": -0.359375, |
| "step": 207 |
| }, |
| { |
| "epoch": 0.7611206403659234, |
| "grad_norm": 380.7069396972656, |
| "learning_rate": 4.761904761904761e-07, |
| "logits/chosen": -0.7265625, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -148.0, |
| "logps/rejected": -183.0, |
| "loss": 0.5838, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.2578125, |
| "rewards/rejected": -0.302734375, |
| "step": 208 |
| }, |
| { |
| "epoch": 0.7647798742138365, |
| "grad_norm": 362.50927734375, |
| "learning_rate": 4.6886446886446884e-07, |
| "logits/chosen": -0.73828125, |
| "logits/rejected": -0.72265625, |
| "logps/chosen": -177.0, |
| "logps/rejected": -156.0, |
| "loss": 0.5818, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.0517578125, |
| "rewards/margins": 0.2490234375, |
| "rewards/rejected": -0.30078125, |
| "step": 209 |
| }, |
| { |
| "epoch": 0.7684391080617495, |
| "grad_norm": 374.5245666503906, |
| "learning_rate": 4.6153846153846156e-07, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -140.0, |
| "logps/rejected": -198.0, |
| "loss": 0.6289, |
| "rewards/accuracies": 0.59375, |
| "rewards/chosen": -0.14453125, |
| "rewards/margins": 0.193359375, |
| "rewards/rejected": -0.337890625, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.7720983419096626, |
| "grad_norm": 341.1881103515625, |
| "learning_rate": 4.542124542124542e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -143.0, |
| "logps/rejected": -178.0, |
| "loss": 0.576, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.0203857421875, |
| "rewards/margins": 0.279296875, |
| "rewards/rejected": -0.30078125, |
| "step": 211 |
| }, |
| { |
| "epoch": 0.7757575757575758, |
| "grad_norm": 337.7765808105469, |
| "learning_rate": 4.468864468864469e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -135.0, |
| "logps/rejected": -186.0, |
| "loss": 0.5106, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": 0.005462646484375, |
| "rewards/margins": 0.44140625, |
| "rewards/rejected": -0.435546875, |
| "step": 212 |
| }, |
| { |
| "epoch": 0.7794168096054889, |
| "grad_norm": 355.8410339355469, |
| "learning_rate": 4.395604395604395e-07, |
| "logits/chosen": -0.6796875, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -158.0, |
| "logps/rejected": -233.0, |
| "loss": 0.5145, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.44140625, |
| "rewards/rejected": -0.48828125, |
| "step": 213 |
| }, |
| { |
| "epoch": 0.783076043453402, |
| "grad_norm": 336.0704345703125, |
| "learning_rate": 4.3223443223443223e-07, |
| "logits/chosen": -0.71484375, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -152.0, |
| "logps/rejected": -166.0, |
| "loss": 0.5754, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.0517578125, |
| "rewards/margins": 0.2890625, |
| "rewards/rejected": -0.341796875, |
| "step": 214 |
| }, |
| { |
| "epoch": 0.786735277301315, |
| "grad_norm": 361.47052001953125, |
| "learning_rate": 4.249084249084249e-07, |
| "logits/chosen": -0.7265625, |
| "logits/rejected": -0.73046875, |
| "logps/chosen": -148.0, |
| "logps/rejected": -169.0, |
| "loss": 0.5704, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.043701171875, |
| "rewards/margins": 0.298828125, |
| "rewards/rejected": -0.341796875, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.7903945111492281, |
| "grad_norm": 323.4903259277344, |
| "learning_rate": 4.1758241758241757e-07, |
| "logits/chosen": -0.6953125, |
| "logits/rejected": -0.73046875, |
| "logps/chosen": -171.0, |
| "logps/rejected": -186.0, |
| "loss": 0.5572, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.095703125, |
| "rewards/margins": 0.3359375, |
| "rewards/rejected": -0.431640625, |
| "step": 216 |
| }, |
| { |
| "epoch": 0.7940537449971412, |
| "grad_norm": 334.9060363769531, |
| "learning_rate": 4.1025641025641024e-07, |
| "logits/chosen": -0.63671875, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -174.0, |
| "logps/rejected": -170.0, |
| "loss": 0.59, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.05078125, |
| "rewards/margins": 0.2421875, |
| "rewards/rejected": -0.29296875, |
| "step": 217 |
| }, |
| { |
| "epoch": 0.7977129788450543, |
| "grad_norm": 318.1783142089844, |
| "learning_rate": 4.029304029304029e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.72265625, |
| "logps/chosen": -146.0, |
| "logps/rejected": -157.0, |
| "loss": 0.5676, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.08056640625, |
| "rewards/margins": 0.2890625, |
| "rewards/rejected": -0.369140625, |
| "step": 218 |
| }, |
| { |
| "epoch": 0.8013722126929674, |
| "grad_norm": 361.3813171386719, |
| "learning_rate": 3.9560439560439557e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -168.0, |
| "logps/rejected": -171.0, |
| "loss": 0.5864, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1064453125, |
| "rewards/margins": 0.23828125, |
| "rewards/rejected": -0.345703125, |
| "step": 219 |
| }, |
| { |
| "epoch": 0.8050314465408805, |
| "grad_norm": 366.4966735839844, |
| "learning_rate": 3.882783882783883e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -167.0, |
| "logps/rejected": -196.0, |
| "loss": 0.5481, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.12451171875, |
| "rewards/margins": 0.416015625, |
| "rewards/rejected": -0.5390625, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.8086906803887935, |
| "grad_norm": 337.0361022949219, |
| "learning_rate": 3.809523809523809e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -155.0, |
| "logps/rejected": -188.0, |
| "loss": 0.5627, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.072265625, |
| "rewards/margins": 0.322265625, |
| "rewards/rejected": -0.39453125, |
| "step": 221 |
| }, |
| { |
| "epoch": 0.8123499142367067, |
| "grad_norm": 323.75201416015625, |
| "learning_rate": 3.7362637362637363e-07, |
| "logits/chosen": -0.7421875, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -150.0, |
| "logps/rejected": -199.0, |
| "loss": 0.5634, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.09521484375, |
| "rewards/margins": 0.37109375, |
| "rewards/rejected": -0.466796875, |
| "step": 222 |
| }, |
| { |
| "epoch": 0.8160091480846198, |
| "grad_norm": 367.0987243652344, |
| "learning_rate": 3.6630036630036624e-07, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -168.0, |
| "logps/rejected": -156.0, |
| "loss": 0.5681, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.04052734375, |
| "rewards/margins": 0.294921875, |
| "rewards/rejected": -0.3359375, |
| "step": 223 |
| }, |
| { |
| "epoch": 0.8196683819325329, |
| "grad_norm": 362.4571838378906, |
| "learning_rate": 3.5897435897435896e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.73046875, |
| "logps/chosen": -142.0, |
| "logps/rejected": -204.0, |
| "loss": 0.5521, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.07373046875, |
| "rewards/margins": 0.34765625, |
| "rewards/rejected": -0.421875, |
| "step": 224 |
| }, |
| { |
| "epoch": 0.823327615780446, |
| "grad_norm": 340.5346374511719, |
| "learning_rate": 3.516483516483517e-07, |
| "logits/chosen": -0.76953125, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -173.0, |
| "logps/rejected": -190.0, |
| "loss": 0.5436, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0250244140625, |
| "rewards/margins": 0.37890625, |
| "rewards/rejected": -0.40234375, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.8269868496283591, |
| "grad_norm": 369.2457580566406, |
| "learning_rate": 3.443223443223443e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -153.0, |
| "logps/rejected": -188.0, |
| "loss": 0.6034, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.1328125, |
| "rewards/margins": 0.2265625, |
| "rewards/rejected": -0.359375, |
| "step": 226 |
| }, |
| { |
| "epoch": 0.8306460834762721, |
| "grad_norm": 373.647216796875, |
| "learning_rate": 3.36996336996337e-07, |
| "logits/chosen": -0.734375, |
| "logits/rejected": -0.75390625, |
| "logps/chosen": -176.0, |
| "logps/rejected": -161.0, |
| "loss": 0.5767, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.0751953125, |
| "rewards/margins": 0.283203125, |
| "rewards/rejected": -0.357421875, |
| "step": 227 |
| }, |
| { |
| "epoch": 0.8343053173241852, |
| "grad_norm": 391.55511474609375, |
| "learning_rate": 3.2967032967032963e-07, |
| "logits/chosen": -0.7109375, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -167.0, |
| "logps/rejected": -149.0, |
| "loss": 0.5791, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0947265625, |
| "rewards/margins": 0.265625, |
| "rewards/rejected": -0.361328125, |
| "step": 228 |
| }, |
| { |
| "epoch": 0.8379645511720983, |
| "grad_norm": 340.9716796875, |
| "learning_rate": 3.2234432234432236e-07, |
| "logits/chosen": -0.74609375, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -140.0, |
| "logps/rejected": -163.0, |
| "loss": 0.5575, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.06103515625, |
| "rewards/margins": 0.322265625, |
| "rewards/rejected": -0.3828125, |
| "step": 229 |
| }, |
| { |
| "epoch": 0.8416237850200115, |
| "grad_norm": 334.7230529785156, |
| "learning_rate": 3.1501831501831497e-07, |
| "logits/chosen": -0.7265625, |
| "logits/rejected": -0.734375, |
| "logps/chosen": -148.0, |
| "logps/rejected": -168.0, |
| "loss": 0.6023, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.06640625, |
| "rewards/margins": 0.22265625, |
| "rewards/rejected": -0.2890625, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.8452830188679246, |
| "grad_norm": 351.0285339355469, |
| "learning_rate": 3.076923076923077e-07, |
| "logits/chosen": -0.7265625, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -151.0, |
| "logps/rejected": -160.0, |
| "loss": 0.5653, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.07666015625, |
| "rewards/margins": 0.3046875, |
| "rewards/rejected": -0.380859375, |
| "step": 231 |
| }, |
| { |
| "epoch": 0.8489422527158377, |
| "grad_norm": 340.0621337890625, |
| "learning_rate": 3.0036630036630036e-07, |
| "logits/chosen": -0.75390625, |
| "logits/rejected": -0.7421875, |
| "logps/chosen": -145.0, |
| "logps/rejected": -157.0, |
| "loss": 0.5784, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.015625, |
| "rewards/margins": 0.26953125, |
| "rewards/rejected": -0.28515625, |
| "step": 232 |
| }, |
| { |
| "epoch": 0.8526014865637507, |
| "grad_norm": 373.56488037109375, |
| "learning_rate": 2.93040293040293e-07, |
| "logits/chosen": -0.71875, |
| "logits/rejected": -0.703125, |
| "logps/chosen": -156.0, |
| "logps/rejected": -218.0, |
| "loss": 0.5762, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.12451171875, |
| "rewards/margins": 0.275390625, |
| "rewards/rejected": -0.400390625, |
| "step": 233 |
| }, |
| { |
| "epoch": 0.8562607204116638, |
| "grad_norm": 340.0142517089844, |
| "learning_rate": 2.857142857142857e-07, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -149.0, |
| "logps/rejected": -139.0, |
| "loss": 0.6108, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.09375, |
| "rewards/margins": 0.18359375, |
| "rewards/rejected": -0.27734375, |
| "step": 234 |
| }, |
| { |
| "epoch": 0.8599199542595769, |
| "grad_norm": 363.7709655761719, |
| "learning_rate": 2.7838827838827836e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.7109375, |
| "logps/chosen": -168.0, |
| "logps/rejected": -213.0, |
| "loss": 0.551, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0859375, |
| "rewards/margins": 0.353515625, |
| "rewards/rejected": -0.439453125, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.86357918810749, |
| "grad_norm": 354.7901916503906, |
| "learning_rate": 2.7106227106227103e-07, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -137.0, |
| "logps/rejected": -155.0, |
| "loss": 0.6004, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.126953125, |
| "rewards/margins": 0.2265625, |
| "rewards/rejected": -0.353515625, |
| "step": 236 |
| }, |
| { |
| "epoch": 0.8672384219554031, |
| "grad_norm": 331.5390930175781, |
| "learning_rate": 2.6373626373626375e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -164.0, |
| "logps/rejected": -212.0, |
| "loss": 0.5416, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.0791015625, |
| "rewards/margins": 0.4140625, |
| "rewards/rejected": -0.4921875, |
| "step": 237 |
| }, |
| { |
| "epoch": 0.8708976558033161, |
| "grad_norm": 316.5717468261719, |
| "learning_rate": 2.5641025641025636e-07, |
| "logits/chosen": -0.71875, |
| "logits/rejected": -0.74609375, |
| "logps/chosen": -136.0, |
| "logps/rejected": -169.0, |
| "loss": 0.5723, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": 3.0517578125e-05, |
| "rewards/margins": 0.294921875, |
| "rewards/rejected": -0.294921875, |
| "step": 238 |
| }, |
| { |
| "epoch": 0.8745568896512292, |
| "grad_norm": 317.3512268066406, |
| "learning_rate": 2.490842490842491e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.671875, |
| "logps/chosen": -180.0, |
| "logps/rejected": -165.0, |
| "loss": 0.5302, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.006256103515625, |
| "rewards/margins": 0.388671875, |
| "rewards/rejected": -0.39453125, |
| "step": 239 |
| }, |
| { |
| "epoch": 0.8782161234991424, |
| "grad_norm": 365.77197265625, |
| "learning_rate": 2.4175824175824175e-07, |
| "logits/chosen": -0.71875, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -191.0, |
| "logps/rejected": -222.0, |
| "loss": 0.5382, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.06884765625, |
| "rewards/margins": 0.390625, |
| "rewards/rejected": -0.458984375, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.8818753573470555, |
| "grad_norm": 344.7721252441406, |
| "learning_rate": 2.3443223443223442e-07, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.6640625, |
| "logps/chosen": -142.0, |
| "logps/rejected": -182.0, |
| "loss": 0.5768, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.10400390625, |
| "rewards/margins": 0.271484375, |
| "rewards/rejected": -0.375, |
| "step": 241 |
| }, |
| { |
| "epoch": 0.8855345911949686, |
| "grad_norm": 343.99896240234375, |
| "learning_rate": 2.271062271062271e-07, |
| "logits/chosen": -0.68359375, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -167.0, |
| "logps/rejected": -187.0, |
| "loss": 0.5477, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0908203125, |
| "rewards/margins": 0.345703125, |
| "rewards/rejected": -0.435546875, |
| "step": 242 |
| }, |
| { |
| "epoch": 0.8891938250428817, |
| "grad_norm": 351.04913330078125, |
| "learning_rate": 2.1978021978021976e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -154.0, |
| "logps/rejected": -162.0, |
| "loss": 0.5964, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.01806640625, |
| "rewards/margins": 0.2431640625, |
| "rewards/rejected": -0.26171875, |
| "step": 243 |
| }, |
| { |
| "epoch": 0.8928530588907947, |
| "grad_norm": 331.8519287109375, |
| "learning_rate": 2.1245421245421245e-07, |
| "logits/chosen": -0.734375, |
| "logits/rejected": -0.73046875, |
| "logps/chosen": -137.0, |
| "logps/rejected": -184.0, |
| "loss": 0.5541, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.0140380859375, |
| "rewards/margins": 0.33984375, |
| "rewards/rejected": -0.353515625, |
| "step": 244 |
| }, |
| { |
| "epoch": 0.8965122927387078, |
| "grad_norm": 343.590087890625, |
| "learning_rate": 2.0512820512820512e-07, |
| "logits/chosen": -0.66796875, |
| "logits/rejected": -0.65625, |
| "logps/chosen": -147.0, |
| "logps/rejected": -196.0, |
| "loss": 0.5594, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": 0.00628662109375, |
| "rewards/margins": 0.314453125, |
| "rewards/rejected": -0.30859375, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.9001715265866209, |
| "grad_norm": 400.9288635253906, |
| "learning_rate": 1.9780219780219779e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -150.0, |
| "logps/rejected": -184.0, |
| "loss": 0.5757, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.06103515625, |
| "rewards/margins": 0.291015625, |
| "rewards/rejected": -0.3515625, |
| "step": 246 |
| }, |
| { |
| "epoch": 0.903830760434534, |
| "grad_norm": 346.36065673828125, |
| "learning_rate": 1.9047619047619045e-07, |
| "logits/chosen": -0.69140625, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -153.0, |
| "logps/rejected": -198.0, |
| "loss": 0.5306, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.045166015625, |
| "rewards/margins": 0.380859375, |
| "rewards/rejected": -0.42578125, |
| "step": 247 |
| }, |
| { |
| "epoch": 0.9074899942824471, |
| "grad_norm": 312.3040466308594, |
| "learning_rate": 1.8315018315018312e-07, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.67578125, |
| "logps/chosen": -149.0, |
| "logps/rejected": -218.0, |
| "loss": 0.5116, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.048583984375, |
| "rewards/margins": 0.48828125, |
| "rewards/rejected": -0.53515625, |
| "step": 248 |
| }, |
| { |
| "epoch": 0.9111492281303603, |
| "grad_norm": 320.7344665527344, |
| "learning_rate": 1.7582417582417584e-07, |
| "logits/chosen": -0.67578125, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -138.0, |
| "logps/rejected": -164.0, |
| "loss": 0.5797, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.035888671875, |
| "rewards/margins": 0.267578125, |
| "rewards/rejected": -0.302734375, |
| "step": 249 |
| }, |
| { |
| "epoch": 0.9148084619782733, |
| "grad_norm": 331.97637939453125, |
| "learning_rate": 1.684981684981685e-07, |
| "logits/chosen": -0.74609375, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -158.0, |
| "logps/rejected": -200.0, |
| "loss": 0.5593, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.04541015625, |
| "rewards/margins": 0.333984375, |
| "rewards/rejected": -0.380859375, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.9184676958261864, |
| "grad_norm": 334.0833435058594, |
| "learning_rate": 1.6117216117216118e-07, |
| "logits/chosen": -0.73046875, |
| "logits/rejected": -0.70703125, |
| "logps/chosen": -153.0, |
| "logps/rejected": -223.0, |
| "loss": 0.5211, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.0133056640625, |
| "rewards/margins": 0.40625, |
| "rewards/rejected": -0.419921875, |
| "step": 251 |
| }, |
| { |
| "epoch": 0.9221269296740995, |
| "grad_norm": 315.7561950683594, |
| "learning_rate": 1.5384615384615385e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.69921875, |
| "logps/chosen": -163.0, |
| "logps/rejected": -171.0, |
| "loss": 0.5767, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.01483154296875, |
| "rewards/margins": 0.27734375, |
| "rewards/rejected": -0.291015625, |
| "step": 252 |
| }, |
| { |
| "epoch": 0.9257861635220126, |
| "grad_norm": 340.6386413574219, |
| "learning_rate": 1.465201465201465e-07, |
| "logits/chosen": -0.65234375, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -142.0, |
| "logps/rejected": -180.0, |
| "loss": 0.5369, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.051025390625, |
| "rewards/margins": 0.380859375, |
| "rewards/rejected": -0.431640625, |
| "step": 253 |
| }, |
| { |
| "epoch": 0.9294453973699257, |
| "grad_norm": 325.98187255859375, |
| "learning_rate": 1.3919413919413918e-07, |
| "logits/chosen": -0.6875, |
| "logits/rejected": -0.63671875, |
| "logps/chosen": -150.0, |
| "logps/rejected": -147.0, |
| "loss": 0.5456, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.0250244140625, |
| "rewards/margins": 0.34375, |
| "rewards/rejected": -0.369140625, |
| "step": 254 |
| }, |
| { |
| "epoch": 0.9331046312178388, |
| "grad_norm": 355.2104187011719, |
| "learning_rate": 1.3186813186813187e-07, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.6953125, |
| "logps/chosen": -154.0, |
| "logps/rejected": -208.0, |
| "loss": 0.5472, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.05078125, |
| "rewards/margins": 0.404296875, |
| "rewards/rejected": -0.455078125, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.9367638650657518, |
| "grad_norm": 366.7920227050781, |
| "learning_rate": 1.2454212454212454e-07, |
| "logits/chosen": -0.703125, |
| "logits/rejected": -0.7109375, |
| "logps/chosen": -161.0, |
| "logps/rejected": -159.0, |
| "loss": 0.5634, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.043701171875, |
| "rewards/margins": 0.318359375, |
| "rewards/rejected": -0.361328125, |
| "step": 256 |
| }, |
| { |
| "epoch": 0.9404230989136649, |
| "grad_norm": 345.1733093261719, |
| "learning_rate": 1.1721611721611721e-07, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.703125, |
| "logps/chosen": -182.0, |
| "logps/rejected": -167.0, |
| "loss": 0.5864, |
| "rewards/accuracies": 0.71875, |
| "rewards/chosen": -0.10498046875, |
| "rewards/margins": 0.255859375, |
| "rewards/rejected": -0.361328125, |
| "step": 257 |
| }, |
| { |
| "epoch": 0.944082332761578, |
| "grad_norm": 342.8348083496094, |
| "learning_rate": 1.0989010989010988e-07, |
| "logits/chosen": -0.69921875, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -165.0, |
| "logps/rejected": -150.0, |
| "loss": 0.5539, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.00311279296875, |
| "rewards/margins": 0.328125, |
| "rewards/rejected": -0.330078125, |
| "step": 258 |
| }, |
| { |
| "epoch": 0.9477415666094912, |
| "grad_norm": 317.3765563964844, |
| "learning_rate": 1.0256410256410256e-07, |
| "logits/chosen": -0.75, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -133.0, |
| "logps/rejected": -194.0, |
| "loss": 0.5248, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.00946044921875, |
| "rewards/margins": 0.41796875, |
| "rewards/rejected": -0.427734375, |
| "step": 259 |
| }, |
| { |
| "epoch": 0.9514008004574043, |
| "grad_norm": 334.01129150390625, |
| "learning_rate": 9.523809523809523e-08, |
| "logits/chosen": -0.7109375, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -142.0, |
| "logps/rejected": -180.0, |
| "loss": 0.5478, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": 0.003936767578125, |
| "rewards/margins": 0.353515625, |
| "rewards/rejected": -0.349609375, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.9550600343053173, |
| "grad_norm": 384.39617919921875, |
| "learning_rate": 8.791208791208792e-08, |
| "logits/chosen": -0.671875, |
| "logits/rejected": -0.62890625, |
| "logps/chosen": -160.0, |
| "logps/rejected": -151.0, |
| "loss": 0.5869, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.078125, |
| "rewards/margins": 0.2392578125, |
| "rewards/rejected": -0.318359375, |
| "step": 261 |
| }, |
| { |
| "epoch": 0.9587192681532304, |
| "grad_norm": 355.11517333984375, |
| "learning_rate": 8.058608058608059e-08, |
| "logits/chosen": -0.7421875, |
| "logits/rejected": -0.7734375, |
| "logps/chosen": -159.0, |
| "logps/rejected": -186.0, |
| "loss": 0.5905, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.1220703125, |
| "rewards/margins": 0.2314453125, |
| "rewards/rejected": -0.353515625, |
| "step": 262 |
| }, |
| { |
| "epoch": 0.9623785020011435, |
| "grad_norm": 361.0973815917969, |
| "learning_rate": 7.326007326007326e-08, |
| "logits/chosen": -0.70703125, |
| "logits/rejected": -0.6875, |
| "logps/chosen": -151.0, |
| "logps/rejected": -163.0, |
| "loss": 0.5654, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.0859375, |
| "rewards/margins": 0.30078125, |
| "rewards/rejected": -0.38671875, |
| "step": 263 |
| }, |
| { |
| "epoch": 0.9660377358490566, |
| "grad_norm": 341.8106994628906, |
| "learning_rate": 6.593406593406594e-08, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.6796875, |
| "logps/chosen": -155.0, |
| "logps/rejected": -181.0, |
| "loss": 0.567, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.078125, |
| "rewards/margins": 0.291015625, |
| "rewards/rejected": -0.369140625, |
| "step": 264 |
| }, |
| { |
| "epoch": 0.9696969696969697, |
| "grad_norm": 306.0340576171875, |
| "learning_rate": 5.8608058608058605e-08, |
| "logits/chosen": -0.765625, |
| "logits/rejected": -0.71484375, |
| "logps/chosen": -132.0, |
| "logps/rejected": -165.0, |
| "loss": 0.5439, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.052490234375, |
| "rewards/margins": 0.359375, |
| "rewards/rejected": -0.41015625, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.9733562035448828, |
| "grad_norm": 353.2430114746094, |
| "learning_rate": 5.128205128205128e-08, |
| "logits/chosen": -0.7109375, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -181.0, |
| "logps/rejected": -171.0, |
| "loss": 0.6016, |
| "rewards/accuracies": 0.65625, |
| "rewards/chosen": -0.033447265625, |
| "rewards/margins": 0.22265625, |
| "rewards/rejected": -0.255859375, |
| "step": 266 |
| }, |
| { |
| "epoch": 0.9770154373927958, |
| "grad_norm": 316.0684509277344, |
| "learning_rate": 4.395604395604396e-08, |
| "logits/chosen": -0.71484375, |
| "logits/rejected": -0.71875, |
| "logps/chosen": -137.0, |
| "logps/rejected": -163.0, |
| "loss": 0.551, |
| "rewards/accuracies": 0.875, |
| "rewards/chosen": -0.031982421875, |
| "rewards/margins": 0.330078125, |
| "rewards/rejected": -0.36328125, |
| "step": 267 |
| }, |
| { |
| "epoch": 0.980674671240709, |
| "grad_norm": 334.58050537109375, |
| "learning_rate": 3.663003663003663e-08, |
| "logits/chosen": -0.703125, |
| "logits/rejected": -0.66015625, |
| "logps/chosen": -165.0, |
| "logps/rejected": -170.0, |
| "loss": 0.6031, |
| "rewards/accuracies": 0.78125, |
| "rewards/chosen": -0.1416015625, |
| "rewards/margins": 0.2060546875, |
| "rewards/rejected": -0.34765625, |
| "step": 268 |
| }, |
| { |
| "epoch": 0.9843339050886221, |
| "grad_norm": 328.41644287109375, |
| "learning_rate": 2.9304029304029303e-08, |
| "logits/chosen": -0.6640625, |
| "logits/rejected": -0.66796875, |
| "logps/chosen": -143.0, |
| "logps/rejected": -156.0, |
| "loss": 0.579, |
| "rewards/accuracies": 0.6875, |
| "rewards/chosen": -0.06787109375, |
| "rewards/margins": 0.26171875, |
| "rewards/rejected": -0.330078125, |
| "step": 269 |
| }, |
| { |
| "epoch": 0.9879931389365352, |
| "grad_norm": 307.48974609375, |
| "learning_rate": 2.197802197802198e-08, |
| "logits/chosen": -0.65625, |
| "logits/rejected": -0.68359375, |
| "logps/chosen": -145.0, |
| "logps/rejected": -164.0, |
| "loss": 0.5715, |
| "rewards/accuracies": 0.8125, |
| "rewards/chosen": -0.0546875, |
| "rewards/margins": 0.283203125, |
| "rewards/rejected": -0.337890625, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.9916523727844483, |
| "grad_norm": 323.255126953125, |
| "learning_rate": 1.4652014652014651e-08, |
| "logits/chosen": -0.73828125, |
| "logits/rejected": -0.69140625, |
| "logps/chosen": -135.0, |
| "logps/rejected": -147.0, |
| "loss": 0.5377, |
| "rewards/accuracies": 0.90625, |
| "rewards/chosen": -0.01092529296875, |
| "rewards/margins": 0.36328125, |
| "rewards/rejected": -0.375, |
| "step": 271 |
| }, |
| { |
| "epoch": 0.9953116066323614, |
| "grad_norm": 365.11749267578125, |
| "learning_rate": 7.326007326007326e-09, |
| "logits/chosen": -0.74609375, |
| "logits/rejected": -0.734375, |
| "logps/chosen": -157.0, |
| "logps/rejected": -194.0, |
| "loss": 0.5576, |
| "rewards/accuracies": 0.84375, |
| "rewards/chosen": -0.07275390625, |
| "rewards/margins": 0.3203125, |
| "rewards/rejected": -0.39453125, |
| "step": 272 |
| }, |
| { |
| "epoch": 0.9989708404802744, |
| "grad_norm": 335.74774169921875, |
| "learning_rate": 0.0, |
| "logits/chosen": -0.71875, |
| "logits/rejected": -0.74609375, |
| "logps/chosen": -155.0, |
| "logps/rejected": -183.0, |
| "loss": 0.5493, |
| "rewards/accuracies": 0.75, |
| "rewards/chosen": -0.0296630859375, |
| "rewards/margins": 0.353515625, |
| "rewards/rejected": -0.3828125, |
| "step": 273 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 273, |
| "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 |
| } |
|
|
