| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.9011415272082455, | |
| "eval_steps": 500, | |
| "global_step": 4400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1e-08, | |
| "logits/chosen": -2.25, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -81.0, | |
| "loss": 0.6971, | |
| "rewards/accuracies": 0.453125, | |
| "rewards/chosen": -0.011858206242322922, | |
| "rewards/margins": -0.010589980520308018, | |
| "rewards/rejected": -0.0012682259548455477, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "logits/chosen": -2.25, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -248.0, | |
| "logps/rejected": -71.0, | |
| "loss": 0.6939, | |
| "rewards/accuracies": 0.49177631735801697, | |
| "rewards/chosen": -0.0004125148116145283, | |
| "rewards/margins": 0.0006769677856937051, | |
| "rewards/rejected": -0.0010894828010350466, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "logits/chosen": -2.25, | |
| "logits/rejected": -2.40625, | |
| "logps/chosen": -240.0, | |
| "logps/rejected": -71.5, | |
| "loss": 0.6942, | |
| "rewards/accuracies": 0.5140625238418579, | |
| "rewards/chosen": -0.001100730849429965, | |
| "rewards/margins": -0.00020681759633589536, | |
| "rewards/rejected": -0.0008939133840613067, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.000000000000001e-07, | |
| "logits/chosen": -2.25, | |
| "logits/rejected": -2.40625, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -73.0, | |
| "loss": 0.6913, | |
| "rewards/accuracies": 0.5179687738418579, | |
| "rewards/chosen": 0.005297874100506306, | |
| "rewards/margins": 0.007369426544755697, | |
| "rewards/rejected": -0.002071552677080035, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 8.000000000000001e-07, | |
| "logits/chosen": -2.234375, | |
| "logits/rejected": -2.40625, | |
| "logps/chosen": -252.0, | |
| "logps/rejected": -68.0, | |
| "loss": 0.6827, | |
| "rewards/accuracies": 0.5992187261581421, | |
| "rewards/chosen": 0.02091832458972931, | |
| "rewards/margins": 0.02193695865571499, | |
| "rewards/rejected": -0.001018635230138898, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "logits/chosen": -2.25, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -249.0, | |
| "logps/rejected": -69.5, | |
| "loss": 0.6689, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": 0.045813608914613724, | |
| "rewards/margins": 0.05108557268977165, | |
| "rewards/rejected": -0.005271967500448227, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "logits/chosen": -2.25, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -254.0, | |
| "logps/rejected": -69.5, | |
| "loss": 0.6495, | |
| "rewards/accuracies": 0.813281238079071, | |
| "rewards/chosen": 0.07879128307104111, | |
| "rewards/margins": 0.09384522587060928, | |
| "rewards/rejected": -0.01505393348634243, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "logits/chosen": -2.265625, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -239.0, | |
| "logps/rejected": -71.5, | |
| "loss": 0.6311, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": 0.10534026473760605, | |
| "rewards/margins": 0.13596875965595245, | |
| "rewards/rejected": -0.03062848187983036, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "logits/chosen": -2.265625, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -228.0, | |
| "logps/rejected": -69.5, | |
| "loss": 0.6082, | |
| "rewards/accuracies": 0.8656250238418579, | |
| "rewards/chosen": 0.13130317628383636, | |
| "rewards/margins": 0.19439265131950378, | |
| "rewards/rejected": -0.06308946758508682, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.8000000000000001e-06, | |
| "logits/chosen": -2.265625, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -238.0, | |
| "logps/rejected": -68.5, | |
| "loss": 0.5785, | |
| "rewards/accuracies": 0.874218761920929, | |
| "rewards/chosen": 0.17366917431354523, | |
| "rewards/margins": 0.2717178463935852, | |
| "rewards/rejected": -0.09804867208003998, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "logits/chosen": -2.28125, | |
| "logits/rejected": -2.4375, | |
| "logps/chosen": -234.0, | |
| "logps/rejected": -70.0, | |
| "loss": 0.548, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": 0.18854746222496033, | |
| "rewards/margins": 0.355171263217926, | |
| "rewards/rejected": -0.1666238009929657, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.2e-06, | |
| "logits/chosen": -2.296875, | |
| "logits/rejected": -2.453125, | |
| "logps/chosen": -247.0, | |
| "logps/rejected": -75.5, | |
| "loss": 0.5123, | |
| "rewards/accuracies": 0.885937511920929, | |
| "rewards/chosen": 0.20314271748065948, | |
| "rewards/margins": 0.4698442816734314, | |
| "rewards/rejected": -0.2667016088962555, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "logits/chosen": -2.3125, | |
| "logits/rejected": -2.46875, | |
| "logps/chosen": -243.0, | |
| "logps/rejected": -75.5, | |
| "loss": 0.4766, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": 0.18197833001613617, | |
| "rewards/margins": 0.5841859579086304, | |
| "rewards/rejected": -0.4022076725959778, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.6e-06, | |
| "logits/chosen": -2.3125, | |
| "logits/rejected": -2.46875, | |
| "logps/chosen": -243.0, | |
| "logps/rejected": -77.0, | |
| "loss": 0.4487, | |
| "rewards/accuracies": 0.890625, | |
| "rewards/chosen": 0.12376229465007782, | |
| "rewards/margins": 0.7004261612892151, | |
| "rewards/rejected": -0.5766638517379761, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "logits/chosen": -2.328125, | |
| "logits/rejected": -2.5, | |
| "logps/chosen": -238.0, | |
| "logps/rejected": -76.0, | |
| "loss": 0.4131, | |
| "rewards/accuracies": 0.910937488079071, | |
| "rewards/chosen": 0.08943221718072891, | |
| "rewards/margins": 0.8512886762619019, | |
| "rewards/rejected": -0.7618564963340759, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3e-06, | |
| "logits/chosen": -2.34375, | |
| "logits/rejected": -2.515625, | |
| "logps/chosen": -232.0, | |
| "logps/rejected": -77.5, | |
| "loss": 0.3867, | |
| "rewards/accuracies": 0.9234374761581421, | |
| "rewards/chosen": 3.435462713241577e-05, | |
| "rewards/margins": 0.9718645215034485, | |
| "rewards/rejected": -0.9718301892280579, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "logits/chosen": -2.359375, | |
| "logits/rejected": -2.53125, | |
| "logps/chosen": -246.0, | |
| "logps/rejected": -82.0, | |
| "loss": 0.3574, | |
| "rewards/accuracies": 0.91796875, | |
| "rewards/chosen": -0.028441106900572777, | |
| "rewards/margins": 1.1482356786727905, | |
| "rewards/rejected": -1.1766767501831055, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "logits/chosen": -2.359375, | |
| "logits/rejected": -2.53125, | |
| "logps/chosen": -247.0, | |
| "logps/rejected": -84.0, | |
| "loss": 0.3378, | |
| "rewards/accuracies": 0.9156249761581421, | |
| "rewards/chosen": -0.1456923484802246, | |
| "rewards/margins": 1.2620505094528198, | |
| "rewards/rejected": -1.4077427387237549, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "logits/chosen": -2.375, | |
| "logits/rejected": -2.546875, | |
| "logps/chosen": -245.0, | |
| "logps/rejected": -86.0, | |
| "loss": 0.313, | |
| "rewards/accuracies": 0.91796875, | |
| "rewards/chosen": -0.2751704156398773, | |
| "rewards/margins": 1.4280754327774048, | |
| "rewards/rejected": -1.703245759010315, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.8000000000000005e-06, | |
| "logits/chosen": -2.359375, | |
| "logits/rejected": -2.53125, | |
| "logps/chosen": -244.0, | |
| "logps/rejected": -91.0, | |
| "loss": 0.2764, | |
| "rewards/accuracies": 0.938281238079071, | |
| "rewards/chosen": -0.4051324427127838, | |
| "rewards/margins": 1.6381193399429321, | |
| "rewards/rejected": -2.0432517528533936, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.000000000000001e-06, | |
| "logits/chosen": -2.34375, | |
| "logits/rejected": -2.53125, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -95.0, | |
| "loss": 0.2592, | |
| "rewards/accuracies": 0.936718761920929, | |
| "rewards/chosen": -0.5756620168685913, | |
| "rewards/margins": 1.8360493183135986, | |
| "rewards/rejected": -2.4117112159729004, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "logits/chosen": -2.328125, | |
| "logits/rejected": -2.5, | |
| "logps/chosen": -249.0, | |
| "logps/rejected": -97.5, | |
| "loss": 0.2379, | |
| "rewards/accuracies": 0.928906261920929, | |
| "rewards/chosen": -0.6275719404220581, | |
| "rewards/margins": 2.029207706451416, | |
| "rewards/rejected": -2.6567800045013428, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.4e-06, | |
| "logits/chosen": -2.296875, | |
| "logits/rejected": -2.453125, | |
| "logps/chosen": -244.0, | |
| "logps/rejected": -95.0, | |
| "loss": 0.2296, | |
| "rewards/accuracies": 0.934374988079071, | |
| "rewards/chosen": -0.5478817820549011, | |
| "rewards/margins": 2.155457019805908, | |
| "rewards/rejected": -2.703339099884033, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.600000000000001e-06, | |
| "logits/chosen": -2.296875, | |
| "logits/rejected": -2.453125, | |
| "logps/chosen": -250.0, | |
| "logps/rejected": -101.0, | |
| "loss": 0.2055, | |
| "rewards/accuracies": 0.9359375238418579, | |
| "rewards/chosen": -0.6783720850944519, | |
| "rewards/margins": 2.4690537452697754, | |
| "rewards/rejected": -3.147425651550293, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.800000000000001e-06, | |
| "logits/chosen": -2.265625, | |
| "logits/rejected": -2.421875, | |
| "logps/chosen": -254.0, | |
| "logps/rejected": -108.5, | |
| "loss": 0.1813, | |
| "rewards/accuracies": 0.9476562738418579, | |
| "rewards/chosen": -0.891997218132019, | |
| "rewards/margins": 2.769514799118042, | |
| "rewards/rejected": -3.6615123748779297, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5e-06, | |
| "logits/chosen": -2.265625, | |
| "logits/rejected": -2.40625, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -111.5, | |
| "loss": 0.1712, | |
| "rewards/accuracies": 0.94140625, | |
| "rewards/chosen": -1.160636305809021, | |
| "rewards/margins": 2.947413682937622, | |
| "rewards/rejected": -4.1080498695373535, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.2e-06, | |
| "logits/chosen": -2.203125, | |
| "logits/rejected": -2.34375, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -115.5, | |
| "loss": 0.1571, | |
| "rewards/accuracies": 0.953906238079071, | |
| "rewards/chosen": -1.2190247774124146, | |
| "rewards/margins": 3.16599702835083, | |
| "rewards/rejected": -4.385021209716797, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5.400000000000001e-06, | |
| "logits/chosen": -2.203125, | |
| "logits/rejected": -2.296875, | |
| "logps/chosen": -247.0, | |
| "logps/rejected": -117.5, | |
| "loss": 0.1566, | |
| "rewards/accuracies": 0.9507812261581421, | |
| "rewards/chosen": -1.6336673498153687, | |
| "rewards/margins": 3.272843837738037, | |
| "rewards/rejected": -4.906511306762695, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 5.600000000000001e-06, | |
| "logits/chosen": -2.140625, | |
| "logits/rejected": -2.234375, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -120.5, | |
| "loss": 0.1302, | |
| "rewards/accuracies": 0.96484375, | |
| "rewards/chosen": -1.5824471712112427, | |
| "rewards/margins": 3.517754077911377, | |
| "rewards/rejected": -5.10020112991333, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 5.8e-06, | |
| "logits/chosen": -2.15625, | |
| "logits/rejected": -2.203125, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -127.5, | |
| "loss": 0.1328, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": -1.754668951034546, | |
| "rewards/margins": 3.8442912101745605, | |
| "rewards/rejected": -5.598960876464844, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 6e-06, | |
| "logits/chosen": -2.046875, | |
| "logits/rejected": -2.0625, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -127.5, | |
| "loss": 0.11, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": -1.844238519668579, | |
| "rewards/margins": 4.067028522491455, | |
| "rewards/rejected": -5.911267280578613, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 6.200000000000001e-06, | |
| "logits/chosen": -2.015625, | |
| "logits/rejected": -1.9921875, | |
| "logps/chosen": -262.0, | |
| "logps/rejected": -133.0, | |
| "loss": 0.1116, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": -2.0047552585601807, | |
| "rewards/margins": 4.232540130615234, | |
| "rewards/rejected": -6.237295150756836, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "logits/chosen": -1.9921875, | |
| "logits/rejected": -1.984375, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -134.0, | |
| "loss": 0.1099, | |
| "rewards/accuracies": 0.95703125, | |
| "rewards/chosen": -1.988957405090332, | |
| "rewards/margins": 4.430800437927246, | |
| "rewards/rejected": -6.4197587966918945, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 6.600000000000001e-06, | |
| "logits/chosen": -1.984375, | |
| "logits/rejected": -1.9765625, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -134.0, | |
| "loss": 0.1005, | |
| "rewards/accuracies": 0.9632812738418579, | |
| "rewards/chosen": -1.875688910484314, | |
| "rewards/margins": 4.4820122718811035, | |
| "rewards/rejected": -6.357701301574707, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 6.800000000000001e-06, | |
| "logits/chosen": -1.96875, | |
| "logits/rejected": -1.953125, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -143.0, | |
| "loss": 0.0932, | |
| "rewards/accuracies": 0.967968761920929, | |
| "rewards/chosen": -2.173671245574951, | |
| "rewards/margins": 4.973406791687012, | |
| "rewards/rejected": -7.1470770835876465, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7e-06, | |
| "logits/chosen": -1.9765625, | |
| "logits/rejected": -1.9609375, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -138.0, | |
| "loss": 0.0932, | |
| "rewards/accuracies": 0.9671875238418579, | |
| "rewards/chosen": -2.037585735321045, | |
| "rewards/margins": 4.935227394104004, | |
| "rewards/rejected": -6.972813606262207, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "logits/chosen": -1.8828125, | |
| "logits/rejected": -1.84375, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -142.0, | |
| "loss": 0.0866, | |
| "rewards/accuracies": 0.9750000238418579, | |
| "rewards/chosen": -2.1355109214782715, | |
| "rewards/margins": 5.11392879486084, | |
| "rewards/rejected": -7.2494401931762695, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 7.4e-06, | |
| "logits/chosen": -1.8125, | |
| "logits/rejected": -1.7578125, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -141.0, | |
| "loss": 0.0873, | |
| "rewards/accuracies": 0.9664062261581421, | |
| "rewards/chosen": -2.000317335128784, | |
| "rewards/margins": 5.19692325592041, | |
| "rewards/rejected": -7.197241306304932, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 7.600000000000001e-06, | |
| "logits/chosen": -1.765625, | |
| "logits/rejected": -1.6875, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -143.0, | |
| "loss": 0.0798, | |
| "rewards/accuracies": 0.97265625, | |
| "rewards/chosen": -1.756317138671875, | |
| "rewards/margins": 5.629761219024658, | |
| "rewards/rejected": -7.386077880859375, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 7.800000000000002e-06, | |
| "logits/chosen": -1.78125, | |
| "logits/rejected": -1.7265625, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -151.0, | |
| "loss": 0.0707, | |
| "rewards/accuracies": 0.9710937738418579, | |
| "rewards/chosen": -2.1809704303741455, | |
| "rewards/margins": 5.904406547546387, | |
| "rewards/rejected": -8.08537769317627, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 8.000000000000001e-06, | |
| "logits/chosen": -1.71875, | |
| "logits/rejected": -1.6640625, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -147.0, | |
| "loss": 0.0753, | |
| "rewards/accuracies": 0.967968761920929, | |
| "rewards/chosen": -1.8227026462554932, | |
| "rewards/margins": 5.907409191131592, | |
| "rewards/rejected": -7.730111598968506, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 8.2e-06, | |
| "logits/chosen": -1.7265625, | |
| "logits/rejected": -1.6953125, | |
| "logps/chosen": -255.0, | |
| "logps/rejected": -147.0, | |
| "loss": 0.0651, | |
| "rewards/accuracies": 0.9765625, | |
| "rewards/chosen": -1.8092275857925415, | |
| "rewards/margins": 6.15098237991333, | |
| "rewards/rejected": -7.960209846496582, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 8.400000000000001e-06, | |
| "logits/chosen": -1.75, | |
| "logits/rejected": -1.7421875, | |
| "logps/chosen": -255.0, | |
| "logps/rejected": -147.0, | |
| "loss": 0.1021, | |
| "rewards/accuracies": 0.969531238079071, | |
| "rewards/chosen": -1.6101219654083252, | |
| "rewards/margins": 6.002054691314697, | |
| "rewards/rejected": -7.612176418304443, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.6e-06, | |
| "logits/chosen": -1.734375, | |
| "logits/rejected": -1.6953125, | |
| "logps/chosen": -253.0, | |
| "logps/rejected": -146.0, | |
| "loss": 0.0796, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": -1.733062505722046, | |
| "rewards/margins": 5.910679817199707, | |
| "rewards/rejected": -7.643742561340332, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 8.8e-06, | |
| "logits/chosen": -1.703125, | |
| "logits/rejected": -1.6796875, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -148.0, | |
| "loss": 0.099, | |
| "rewards/accuracies": 0.9585937261581421, | |
| "rewards/chosen": -1.9996089935302734, | |
| "rewards/margins": 5.942723751068115, | |
| "rewards/rejected": -7.9423322677612305, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 9e-06, | |
| "logits/chosen": -1.6015625, | |
| "logits/rejected": -1.5703125, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -149.0, | |
| "loss": 0.0651, | |
| "rewards/accuracies": 0.9750000238418579, | |
| "rewards/chosen": -1.6655864715576172, | |
| "rewards/margins": 6.289559364318848, | |
| "rewards/rejected": -7.955146789550781, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.200000000000002e-06, | |
| "logits/chosen": -1.6953125, | |
| "logits/rejected": -1.703125, | |
| "logps/chosen": -252.0, | |
| "logps/rejected": -148.0, | |
| "loss": 0.0676, | |
| "rewards/accuracies": 0.973437488079071, | |
| "rewards/chosen": -1.2225464582443237, | |
| "rewards/margins": 6.570587158203125, | |
| "rewards/rejected": -7.7931342124938965, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 9.4e-06, | |
| "logits/chosen": -1.6484375, | |
| "logits/rejected": -1.6171875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -154.0, | |
| "loss": 0.0582, | |
| "rewards/accuracies": 0.984375, | |
| "rewards/chosen": -1.5915133953094482, | |
| "rewards/margins": 6.880612850189209, | |
| "rewards/rejected": -8.472126007080078, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 9.600000000000001e-06, | |
| "logits/chosen": -1.6796875, | |
| "logits/rejected": -1.6796875, | |
| "logps/chosen": -242.0, | |
| "logps/rejected": -148.0, | |
| "loss": 0.0659, | |
| "rewards/accuracies": 0.97265625, | |
| "rewards/chosen": -1.3238825798034668, | |
| "rewards/margins": 6.604360103607178, | |
| "rewards/rejected": -7.928241729736328, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.800000000000001e-06, | |
| "logits/chosen": -1.5859375, | |
| "logits/rejected": -1.546875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -159.0, | |
| "loss": 0.0621, | |
| "rewards/accuracies": 0.9742187261581421, | |
| "rewards/chosen": -1.903455376625061, | |
| "rewards/margins": 6.904341220855713, | |
| "rewards/rejected": -8.807796478271484, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1e-05, | |
| "logits/chosen": -1.53125, | |
| "logits/rejected": -1.4765625, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -161.0, | |
| "loss": 0.0598, | |
| "rewards/accuracies": 0.969531238079071, | |
| "rewards/chosen": -1.7653465270996094, | |
| "rewards/margins": 7.352142333984375, | |
| "rewards/rejected": -9.117487907409668, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 9.942987457240593e-06, | |
| "logits/chosen": -1.515625, | |
| "logits/rejected": -1.4765625, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -166.0, | |
| "loss": 0.0725, | |
| "rewards/accuracies": 0.9710937738418579, | |
| "rewards/chosen": -2.206141948699951, | |
| "rewards/margins": 7.433462619781494, | |
| "rewards/rejected": -9.639605522155762, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 9.885974914481187e-06, | |
| "logits/chosen": -1.4921875, | |
| "logits/rejected": -1.5, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -165.0, | |
| "loss": 0.0579, | |
| "rewards/accuracies": 0.973437488079071, | |
| "rewards/chosen": -1.83426034450531, | |
| "rewards/margins": 7.645236015319824, | |
| "rewards/rejected": -9.479497909545898, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 9.828962371721779e-06, | |
| "logits/chosen": -1.3359375, | |
| "logits/rejected": -1.28125, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -165.0, | |
| "loss": 0.0569, | |
| "rewards/accuracies": 0.9789062738418579, | |
| "rewards/chosen": -2.0447144508361816, | |
| "rewards/margins": 7.545592308044434, | |
| "rewards/rejected": -9.590306282043457, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 9.771949828962373e-06, | |
| "logits/chosen": -1.4296875, | |
| "logits/rejected": -1.390625, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -162.0, | |
| "loss": 0.071, | |
| "rewards/accuracies": 0.975781261920929, | |
| "rewards/chosen": -2.062221050262451, | |
| "rewards/margins": 7.198528289794922, | |
| "rewards/rejected": -9.260747909545898, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 9.714937286202965e-06, | |
| "logits/chosen": -1.4296875, | |
| "logits/rejected": -1.359375, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -156.0, | |
| "loss": 0.0608, | |
| "rewards/accuracies": 0.973437488079071, | |
| "rewards/chosen": -1.3428866863250732, | |
| "rewards/margins": 7.246352195739746, | |
| "rewards/rejected": -8.589239120483398, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.657924743443559e-06, | |
| "logits/chosen": -1.4140625, | |
| "logits/rejected": -1.34375, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -163.0, | |
| "loss": 0.0705, | |
| "rewards/accuracies": 0.971875011920929, | |
| "rewards/chosen": -2.211174488067627, | |
| "rewards/margins": 7.069491386413574, | |
| "rewards/rejected": -9.280667304992676, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.600912200684151e-06, | |
| "logits/chosen": -1.4453125, | |
| "logits/rejected": -1.3671875, | |
| "logps/chosen": -251.0, | |
| "logps/rejected": -162.0, | |
| "loss": 0.0424, | |
| "rewards/accuracies": 0.9859374761581421, | |
| "rewards/chosen": -1.804861307144165, | |
| "rewards/margins": 7.515401363372803, | |
| "rewards/rejected": -9.32026195526123, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.543899657924745e-06, | |
| "logits/chosen": -1.4375, | |
| "logits/rejected": -1.3828125, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -172.0, | |
| "loss": 0.0362, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -2.1582987308502197, | |
| "rewards/margins": 7.993732452392578, | |
| "rewards/rejected": -10.152030944824219, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.486887115165337e-06, | |
| "logits/chosen": -1.359375, | |
| "logits/rejected": -1.3203125, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -183.0, | |
| "loss": 0.0376, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -2.6995084285736084, | |
| "rewards/margins": 8.586763381958008, | |
| "rewards/rejected": -11.286273002624512, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.429874572405931e-06, | |
| "logits/chosen": -1.34375, | |
| "logits/rejected": -1.2734375, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -171.0, | |
| "loss": 0.0336, | |
| "rewards/accuracies": 0.984375, | |
| "rewards/chosen": -1.5670948028564453, | |
| "rewards/margins": 8.42174243927002, | |
| "rewards/rejected": -9.988837242126465, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 9.372862029646523e-06, | |
| "logits/chosen": -1.421875, | |
| "logits/rejected": -1.375, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -172.0, | |
| "loss": 0.029, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -1.5389244556427002, | |
| "rewards/margins": 8.682077407836914, | |
| "rewards/rejected": -10.221002578735352, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 9.315849486887117e-06, | |
| "logits/chosen": -1.4609375, | |
| "logits/rejected": -1.390625, | |
| "logps/chosen": -262.0, | |
| "logps/rejected": -177.0, | |
| "loss": 0.0381, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -2.027346134185791, | |
| "rewards/margins": 8.802923202514648, | |
| "rewards/rejected": -10.830268859863281, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 9.25883694412771e-06, | |
| "logits/chosen": -1.4296875, | |
| "logits/rejected": -1.3125, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -173.0, | |
| "loss": 0.0342, | |
| "rewards/accuracies": 0.9859374761581421, | |
| "rewards/chosen": -1.638588309288025, | |
| "rewards/margins": 8.736043930053711, | |
| "rewards/rejected": -10.374631881713867, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 9.201824401368301e-06, | |
| "logits/chosen": -1.34375, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -255.0, | |
| "logps/rejected": -176.0, | |
| "loss": 0.0338, | |
| "rewards/accuracies": 0.9898437261581421, | |
| "rewards/chosen": -1.8327977657318115, | |
| "rewards/margins": 9.06210708618164, | |
| "rewards/rejected": -10.894905090332031, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 9.144811858608895e-06, | |
| "logits/chosen": -1.375, | |
| "logits/rejected": -1.296875, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -178.0, | |
| "loss": 0.0301, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -1.7202352285385132, | |
| "rewards/margins": 9.098135948181152, | |
| "rewards/rejected": -10.81837272644043, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 9.087799315849487e-06, | |
| "logits/chosen": -1.4296875, | |
| "logits/rejected": -1.34375, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -178.0, | |
| "loss": 0.0312, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -1.8896671533584595, | |
| "rewards/margins": 8.800851821899414, | |
| "rewards/rejected": -10.69051742553711, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 9.03078677309008e-06, | |
| "logits/chosen": -1.46875, | |
| "logits/rejected": -1.4453125, | |
| "logps/chosen": -254.0, | |
| "logps/rejected": -177.0, | |
| "loss": 0.0318, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -1.7370885610580444, | |
| "rewards/margins": 9.022928237915039, | |
| "rewards/rejected": -10.760017395019531, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 8.973774230330673e-06, | |
| "logits/chosen": -1.46875, | |
| "logits/rejected": -1.4375, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -184.0, | |
| "loss": 0.0296, | |
| "rewards/accuracies": 0.9906250238418579, | |
| "rewards/chosen": -2.1952197551727295, | |
| "rewards/margins": 9.119722366333008, | |
| "rewards/rejected": -11.314942359924316, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 8.916761687571266e-06, | |
| "logits/chosen": -1.40625, | |
| "logits/rejected": -1.3125, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -179.0, | |
| "loss": 0.0341, | |
| "rewards/accuracies": 0.9859374761581421, | |
| "rewards/chosen": -1.6254527568817139, | |
| "rewards/margins": 9.138445854187012, | |
| "rewards/rejected": -10.763897895812988, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 8.85974914481186e-06, | |
| "logits/chosen": -1.3515625, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -255.0, | |
| "logps/rejected": -171.0, | |
| "loss": 0.0301, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -1.489253282546997, | |
| "rewards/margins": 8.813684463500977, | |
| "rewards/rejected": -10.302938461303711, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 8.802736602052452e-06, | |
| "logits/chosen": -1.3515625, | |
| "logits/rejected": -1.234375, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -183.0, | |
| "loss": 0.0293, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -1.8335425853729248, | |
| "rewards/margins": 9.455236434936523, | |
| "rewards/rejected": -11.288779258728027, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 8.745724059293046e-06, | |
| "logits/chosen": -1.3359375, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -185.0, | |
| "loss": 0.0321, | |
| "rewards/accuracies": 0.9898437261581421, | |
| "rewards/chosen": -2.0949649810791016, | |
| "rewards/margins": 9.466029167175293, | |
| "rewards/rejected": -11.560995101928711, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 8.688711516533638e-06, | |
| "logits/chosen": -1.40625, | |
| "logits/rejected": -1.2890625, | |
| "logps/chosen": -262.0, | |
| "logps/rejected": -184.0, | |
| "loss": 0.0364, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -2.0965704917907715, | |
| "rewards/margins": 9.323990821838379, | |
| "rewards/rejected": -11.420560836791992, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 8.63169897377423e-06, | |
| "logits/chosen": -1.453125, | |
| "logits/rejected": -1.3125, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -184.0, | |
| "loss": 0.0286, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -2.029646396636963, | |
| "rewards/margins": 9.428507804870605, | |
| "rewards/rejected": -11.458154678344727, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 8.574686431014824e-06, | |
| "logits/chosen": -1.4765625, | |
| "logits/rejected": -1.328125, | |
| "logps/chosen": -254.0, | |
| "logps/rejected": -179.0, | |
| "loss": 0.0286, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -1.9806649684906006, | |
| "rewards/margins": 9.070594787597656, | |
| "rewards/rejected": -11.051259994506836, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 8.517673888255416e-06, | |
| "logits/chosen": -1.4609375, | |
| "logits/rejected": -1.3671875, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -186.0, | |
| "loss": 0.0428, | |
| "rewards/accuracies": 0.985156238079071, | |
| "rewards/chosen": -2.330333709716797, | |
| "rewards/margins": 9.129961013793945, | |
| "rewards/rejected": -11.460294723510742, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 8.46066134549601e-06, | |
| "logits/chosen": -1.3984375, | |
| "logits/rejected": -1.265625, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -188.0, | |
| "loss": 0.045, | |
| "rewards/accuracies": 0.9820312261581421, | |
| "rewards/chosen": -2.5738627910614014, | |
| "rewards/margins": 9.508840560913086, | |
| "rewards/rejected": -12.082701683044434, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 8.403648802736602e-06, | |
| "logits/chosen": -1.4765625, | |
| "logits/rejected": -1.3515625, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -179.0, | |
| "loss": 0.0215, | |
| "rewards/accuracies": 0.9906250238418579, | |
| "rewards/chosen": -1.6458170413970947, | |
| "rewards/margins": 9.304633140563965, | |
| "rewards/rejected": -10.95044994354248, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 8.346636259977196e-06, | |
| "logits/chosen": -1.3984375, | |
| "logits/rejected": -1.234375, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -185.0, | |
| "loss": 0.034, | |
| "rewards/accuracies": 0.985156238079071, | |
| "rewards/chosen": -1.9806787967681885, | |
| "rewards/margins": 9.267131805419922, | |
| "rewards/rejected": -11.247811317443848, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 8.289623717217788e-06, | |
| "logits/chosen": -1.328125, | |
| "logits/rejected": -1.1328125, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -196.0, | |
| "loss": 0.0282, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -2.805903434753418, | |
| "rewards/margins": 9.905074119567871, | |
| "rewards/rejected": -12.710977554321289, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 8.232611174458382e-06, | |
| "logits/chosen": -1.3203125, | |
| "logits/rejected": -1.1015625, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -187.0, | |
| "loss": 0.0377, | |
| "rewards/accuracies": 0.984375, | |
| "rewards/chosen": -1.9745209217071533, | |
| "rewards/margins": 9.847769737243652, | |
| "rewards/rejected": -11.822290420532227, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 8.175598631698974e-06, | |
| "logits/chosen": -1.3203125, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -190.0, | |
| "loss": 0.0324, | |
| "rewards/accuracies": 0.989062488079071, | |
| "rewards/chosen": -1.9384052753448486, | |
| "rewards/margins": 10.039579391479492, | |
| "rewards/rejected": -11.977984428405762, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 8.118586088939568e-06, | |
| "logits/chosen": -1.3671875, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -191.0, | |
| "loss": 0.0222, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -2.1441657543182373, | |
| "rewards/margins": 9.81259822845459, | |
| "rewards/rejected": -11.956764221191406, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 8.06157354618016e-06, | |
| "logits/chosen": -1.3046875, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -189.0, | |
| "loss": 0.0236, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -2.252146005630493, | |
| "rewards/margins": 9.743573188781738, | |
| "rewards/rejected": -11.995719909667969, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 8.004561003420754e-06, | |
| "logits/chosen": -1.3203125, | |
| "logits/rejected": -1.1015625, | |
| "logps/chosen": -262.0, | |
| "logps/rejected": -190.0, | |
| "loss": 0.0292, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -1.8711124658584595, | |
| "rewards/margins": 10.123231887817383, | |
| "rewards/rejected": -11.994344711303711, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 7.947548460661346e-06, | |
| "logits/chosen": -1.3984375, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -255.0, | |
| "logps/rejected": -176.0, | |
| "loss": 0.0359, | |
| "rewards/accuracies": 0.9820312261581421, | |
| "rewards/chosen": -1.2544156312942505, | |
| "rewards/margins": 9.407190322875977, | |
| "rewards/rejected": -10.661605834960938, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 7.89053591790194e-06, | |
| "logits/chosen": -1.421875, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -189.0, | |
| "loss": 0.0313, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -1.9942920207977295, | |
| "rewards/margins": 9.837443351745605, | |
| "rewards/rejected": -11.831735610961914, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 7.833523375142532e-06, | |
| "logits/chosen": -1.390625, | |
| "logits/rejected": -1.1875, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -186.0, | |
| "loss": 0.0325, | |
| "rewards/accuracies": 0.9867187738418579, | |
| "rewards/chosen": -1.7988389730453491, | |
| "rewards/margins": 9.885848999023438, | |
| "rewards/rejected": -11.684687614440918, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 7.776510832383124e-06, | |
| "logits/chosen": -1.40625, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -182.0, | |
| "loss": 0.0261, | |
| "rewards/accuracies": 0.9906250238418579, | |
| "rewards/chosen": -1.5674057006835938, | |
| "rewards/margins": 9.868083000183105, | |
| "rewards/rejected": -11.435487747192383, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 7.719498289623718e-06, | |
| "logits/chosen": -1.390625, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -196.0, | |
| "loss": 0.0439, | |
| "rewards/accuracies": 0.985156238079071, | |
| "rewards/chosen": -2.2471280097961426, | |
| "rewards/margins": 10.478254318237305, | |
| "rewards/rejected": -12.725381851196289, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 7.66248574686431e-06, | |
| "logits/chosen": -1.3984375, | |
| "logits/rejected": -1.2265625, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -192.0, | |
| "loss": 0.0255, | |
| "rewards/accuracies": 0.9906250238418579, | |
| "rewards/chosen": -2.0418763160705566, | |
| "rewards/margins": 10.087868690490723, | |
| "rewards/rejected": -12.129744529724121, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 7.605473204104904e-06, | |
| "logits/chosen": -1.359375, | |
| "logits/rejected": -1.1875, | |
| "logps/chosen": -254.0, | |
| "logps/rejected": -187.0, | |
| "loss": 0.0326, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -1.7770217657089233, | |
| "rewards/margins": 9.914891242980957, | |
| "rewards/rejected": -11.691912651062012, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 7.548460661345497e-06, | |
| "logits/chosen": -1.375, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -190.0, | |
| "loss": 0.0368, | |
| "rewards/accuracies": 0.989062488079071, | |
| "rewards/chosen": -1.742910385131836, | |
| "rewards/margins": 9.918044090270996, | |
| "rewards/rejected": -11.660953521728516, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 7.49144811858609e-06, | |
| "logits/chosen": -1.3359375, | |
| "logits/rejected": -1.1328125, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -186.0, | |
| "loss": 0.0306, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -1.813442587852478, | |
| "rewards/margins": 9.687917709350586, | |
| "rewards/rejected": -11.501361846923828, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 7.4344355758266825e-06, | |
| "logits/chosen": -1.3515625, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -188.0, | |
| "loss": 0.0322, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -2.153613328933716, | |
| "rewards/margins": 9.764740943908691, | |
| "rewards/rejected": -11.918354034423828, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 7.3774230330672755e-06, | |
| "logits/chosen": -1.3125, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -201.0, | |
| "loss": 0.0276, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -2.7011101245880127, | |
| "rewards/margins": 10.471219062805176, | |
| "rewards/rejected": -13.172329902648926, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 7.3204104903078685e-06, | |
| "logits/chosen": -1.3828125, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -189.0, | |
| "loss": 0.0355, | |
| "rewards/accuracies": 0.985156238079071, | |
| "rewards/chosen": -1.7472997903823853, | |
| "rewards/margins": 10.209493637084961, | |
| "rewards/rejected": -11.956793785095215, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 7.2633979475484615e-06, | |
| "logits/chosen": -1.375, | |
| "logits/rejected": -1.25, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -209.0, | |
| "loss": 0.0304, | |
| "rewards/accuracies": 0.985156238079071, | |
| "rewards/chosen": -2.909478187561035, | |
| "rewards/margins": 10.954832077026367, | |
| "rewards/rejected": -13.864311218261719, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 7.2063854047890545e-06, | |
| "logits/chosen": -1.3125, | |
| "logits/rejected": -1.1640625, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -204.0, | |
| "loss": 0.0285, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -2.4843478202819824, | |
| "rewards/margins": 11.059233665466309, | |
| "rewards/rejected": -13.54358196258545, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 7.1493728620296476e-06, | |
| "logits/chosen": -1.3828125, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -194.0, | |
| "loss": 0.0225, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -1.8076410293579102, | |
| "rewards/margins": 10.52374267578125, | |
| "rewards/rejected": -12.331381797790527, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 7.09236031927024e-06, | |
| "logits/chosen": -1.375, | |
| "logits/rejected": -1.234375, | |
| "logps/chosen": -251.0, | |
| "logps/rejected": -189.0, | |
| "loss": 0.0268, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -1.8604707717895508, | |
| "rewards/margins": 10.113370895385742, | |
| "rewards/rejected": -11.973841667175293, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 7.035347776510833e-06, | |
| "logits/chosen": -1.359375, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -192.0, | |
| "loss": 0.0304, | |
| "rewards/accuracies": 0.987500011920929, | |
| "rewards/chosen": -1.831870436668396, | |
| "rewards/margins": 10.311990737915039, | |
| "rewards/rejected": -12.143861770629883, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 6.978335233751426e-06, | |
| "logits/chosen": -1.4375, | |
| "logits/rejected": -1.296875, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -183.0, | |
| "loss": 0.0308, | |
| "rewards/accuracies": 0.98828125, | |
| "rewards/chosen": -1.3896019458770752, | |
| "rewards/margins": 10.015677452087402, | |
| "rewards/rejected": -11.405279159545898, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 6.921322690992019e-06, | |
| "logits/chosen": -1.375, | |
| "logits/rejected": -1.2265625, | |
| "logps/chosen": -253.0, | |
| "logps/rejected": -186.0, | |
| "loss": 0.0253, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -1.5604383945465088, | |
| "rewards/margins": 10.14197063446045, | |
| "rewards/rejected": -11.702408790588379, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.864310148232611e-06, | |
| "logits/chosen": -1.3203125, | |
| "logits/rejected": -1.1796875, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -198.0, | |
| "loss": 0.0259, | |
| "rewards/accuracies": 0.9898437261581421, | |
| "rewards/chosen": -2.0974209308624268, | |
| "rewards/margins": 10.54310131072998, | |
| "rewards/rejected": -12.640522003173828, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 6.807297605473204e-06, | |
| "logits/chosen": -1.28125, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -208.0, | |
| "loss": 0.0348, | |
| "rewards/accuracies": 0.9859374761581421, | |
| "rewards/chosen": -2.662431240081787, | |
| "rewards/margins": 11.303977966308594, | |
| "rewards/rejected": -13.966409683227539, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 6.750285062713797e-06, | |
| "logits/chosen": -1.3203125, | |
| "logits/rejected": -1.1640625, | |
| "logps/chosen": -262.0, | |
| "logps/rejected": -198.0, | |
| "loss": 0.0248, | |
| "rewards/accuracies": 0.9937499761581421, | |
| "rewards/chosen": -2.2072205543518066, | |
| "rewards/margins": 10.692474365234375, | |
| "rewards/rejected": -12.899694442749023, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 6.69327251995439e-06, | |
| "logits/chosen": -1.2890625, | |
| "logits/rejected": -1.140625, | |
| "logps/chosen": -256.0, | |
| "logps/rejected": -196.0, | |
| "loss": 0.0264, | |
| "rewards/accuracies": 0.989062488079071, | |
| "rewards/chosen": -1.9567207098007202, | |
| "rewards/margins": 10.699260711669922, | |
| "rewards/rejected": -12.655980110168457, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 6.636259977194983e-06, | |
| "logits/chosen": -1.2578125, | |
| "logits/rejected": -1.1171875, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -195.0, | |
| "loss": 0.0269, | |
| "rewards/accuracies": 0.9906250238418579, | |
| "rewards/chosen": -2.2203476428985596, | |
| "rewards/margins": 10.40222454071045, | |
| "rewards/rejected": -12.622572898864746, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.579247434435576e-06, | |
| "logits/chosen": -1.25, | |
| "logits/rejected": -1.0546875, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -188.0, | |
| "loss": 0.0203, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -1.528204321861267, | |
| "rewards/margins": 10.256698608398438, | |
| "rewards/rejected": -11.784903526306152, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.522234891676169e-06, | |
| "logits/chosen": -1.2109375, | |
| "logits/rejected": -1.03125, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -199.0, | |
| "loss": 0.0235, | |
| "rewards/accuracies": 0.9921875, | |
| "rewards/chosen": -2.464181423187256, | |
| "rewards/margins": 10.21046257019043, | |
| "rewards/rejected": -12.674644470214844, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.465222348916762e-06, | |
| "logits/chosen": -1.2265625, | |
| "logits/rejected": -1.0546875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -194.0, | |
| "loss": 0.0304, | |
| "rewards/accuracies": 0.9867187738418579, | |
| "rewards/chosen": -2.2391433715820312, | |
| "rewards/margins": 10.337366104125977, | |
| "rewards/rejected": -12.576510429382324, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 6.408209806157355e-06, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -0.9453125, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -191.0, | |
| "loss": 0.023, | |
| "rewards/accuracies": 0.991406261920929, | |
| "rewards/chosen": -2.1470155715942383, | |
| "rewards/margins": 10.05528450012207, | |
| "rewards/rejected": -12.202299118041992, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.351197263397948e-06, | |
| "logits/chosen": -1.171875, | |
| "logits/rejected": -0.95703125, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -205.0, | |
| "loss": 0.0041, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.200409173965454, | |
| "rewards/margins": 11.429448127746582, | |
| "rewards/rejected": -13.629857063293457, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.294184720638541e-06, | |
| "logits/chosen": -1.203125, | |
| "logits/rejected": -0.984375, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -203.0, | |
| "loss": 0.0057, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -1.8475700616836548, | |
| "rewards/margins": 11.384028434753418, | |
| "rewards/rejected": -13.231597900390625, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 6.237172177879134e-06, | |
| "logits/chosen": -1.1328125, | |
| "logits/rejected": -0.8984375, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -211.0, | |
| "loss": 0.0067, | |
| "rewards/accuracies": 0.9976562261581421, | |
| "rewards/chosen": -2.3629112243652344, | |
| "rewards/margins": 11.951788902282715, | |
| "rewards/rejected": -14.314702033996582, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 6.180159635119727e-06, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -0.80078125, | |
| "logps/chosen": -258.0, | |
| "logps/rejected": -219.0, | |
| "loss": 0.0048, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.6320786476135254, | |
| "rewards/margins": 12.362083435058594, | |
| "rewards/rejected": -14.994163513183594, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.123147092360319e-06, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -0.8125, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -218.0, | |
| "loss": 0.0044, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.58856463432312, | |
| "rewards/margins": 12.54808521270752, | |
| "rewards/rejected": -15.136648178100586, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 6.066134549600912e-06, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.71875, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -226.0, | |
| "loss": 0.0046, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.7806572914123535, | |
| "rewards/margins": 13.015957832336426, | |
| "rewards/rejected": -15.796615600585938, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 6.009122006841505e-06, | |
| "logits/chosen": -0.9765625, | |
| "logits/rejected": -0.671875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -228.0, | |
| "loss": 0.0133, | |
| "rewards/accuracies": 0.99609375, | |
| "rewards/chosen": -3.3743209838867188, | |
| "rewards/margins": 12.621271133422852, | |
| "rewards/rejected": -15.995593070983887, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.9521094640820984e-06, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -0.671875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -219.0, | |
| "loss": 0.0023, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.076186418533325, | |
| "rewards/margins": 12.830595016479492, | |
| "rewards/rejected": -14.906782150268555, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.8950969213226914e-06, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -0.6484375, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -228.0, | |
| "loss": 0.0028, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.168813467025757, | |
| "rewards/margins": 13.737543106079102, | |
| "rewards/rejected": -15.906353950500488, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 5.8380843785632845e-06, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -0.6640625, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -227.0, | |
| "loss": 0.005, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.49792742729187, | |
| "rewards/margins": 13.276643753051758, | |
| "rewards/rejected": -15.774572372436523, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.7810718358038775e-06, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -0.7265625, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -224.0, | |
| "loss": 0.008, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.374281167984009, | |
| "rewards/margins": 13.046087265014648, | |
| "rewards/rejected": -15.420369148254395, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 5.7240592930444705e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.6171875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -226.0, | |
| "loss": 0.0036, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.243548631668091, | |
| "rewards/margins": 13.532651901245117, | |
| "rewards/rejected": -15.776202201843262, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 5.6670467502850635e-06, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -0.64453125, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -234.0, | |
| "loss": 0.004, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.683474063873291, | |
| "rewards/margins": 13.599886894226074, | |
| "rewards/rejected": -16.28335952758789, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 5.6100342075256565e-06, | |
| "logits/chosen": -1.1484375, | |
| "logits/rejected": -0.71875, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -234.0, | |
| "loss": 0.0042, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.5661137104034424, | |
| "rewards/margins": 13.591761589050293, | |
| "rewards/rejected": -16.157875061035156, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 5.5530216647662496e-06, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -0.63671875, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -241.0, | |
| "loss": 0.0035, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.7359392642974854, | |
| "rewards/margins": 14.275691032409668, | |
| "rewards/rejected": -17.01163101196289, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 5.496009122006843e-06, | |
| "logits/chosen": -1.15625, | |
| "logits/rejected": -0.625, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -241.0, | |
| "loss": 0.0057, | |
| "rewards/accuracies": 0.9976562261581421, | |
| "rewards/chosen": -2.646916389465332, | |
| "rewards/margins": 14.39399242401123, | |
| "rewards/rejected": -17.040910720825195, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 5.438996579247436e-06, | |
| "logits/chosen": -1.171875, | |
| "logits/rejected": -0.71484375, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -242.0, | |
| "loss": 0.0073, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.6745243072509766, | |
| "rewards/margins": 14.485051155090332, | |
| "rewards/rejected": -17.159576416015625, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 5.381984036488029e-06, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -0.58203125, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -234.0, | |
| "loss": 0.0016, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.587552547454834, | |
| "rewards/margins": 13.87269115447998, | |
| "rewards/rejected": -16.46024513244629, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 5.324971493728621e-06, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -0.53515625, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -241.0, | |
| "loss": 0.0027, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.578281879425049, | |
| "rewards/margins": 14.362253189086914, | |
| "rewards/rejected": -16.940534591674805, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 5.267958950969214e-06, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.44140625, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -262.0, | |
| "loss": 0.0024, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.1921732425689697, | |
| "rewards/margins": 15.826263427734375, | |
| "rewards/rejected": -19.018436431884766, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 5.210946408209807e-06, | |
| "logits/chosen": -1.078125, | |
| "logits/rejected": -0.455078125, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -266.0, | |
| "loss": 0.0031, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.5577616691589355, | |
| "rewards/margins": 15.753667831420898, | |
| "rewards/rejected": -19.311429977416992, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 5.153933865450399e-06, | |
| "logits/chosen": -1.140625, | |
| "logits/rejected": -0.58984375, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -242.0, | |
| "loss": 0.0051, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.8636865615844727, | |
| "rewards/margins": 14.327638626098633, | |
| "rewards/rejected": -17.191326141357422, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 5.096921322690992e-06, | |
| "logits/chosen": -1.0390625, | |
| "logits/rejected": -0.498046875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -246.0, | |
| "loss": 0.0028, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.9436042308807373, | |
| "rewards/margins": 14.590520858764648, | |
| "rewards/rejected": -17.53412437438965, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 5.039908779931585e-06, | |
| "logits/chosen": -1.03125, | |
| "logits/rejected": -0.4453125, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -260.0, | |
| "loss": 0.0021, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.5330169200897217, | |
| "rewards/margins": 15.42271614074707, | |
| "rewards/rejected": -18.955734252929688, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 4.982896237172178e-06, | |
| "logits/chosen": -1.0390625, | |
| "logits/rejected": -0.4921875, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -251.0, | |
| "loss": 0.0156, | |
| "rewards/accuracies": 0.9937499761581421, | |
| "rewards/chosen": -3.311683177947998, | |
| "rewards/margins": 14.74211597442627, | |
| "rewards/rejected": -18.05379867553711, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 4.925883694412771e-06, | |
| "logits/chosen": -1.0234375, | |
| "logits/rejected": -0.490234375, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -258.0, | |
| "loss": 0.0058, | |
| "rewards/accuracies": 0.9976562261581421, | |
| "rewards/chosen": -3.4664859771728516, | |
| "rewards/margins": 15.337985038757324, | |
| "rewards/rejected": -18.804471969604492, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 4.868871151653364e-06, | |
| "logits/chosen": -1.140625, | |
| "logits/rejected": -0.63671875, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -252.0, | |
| "loss": 0.0041, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -3.0301337242126465, | |
| "rewards/margins": 15.127180099487305, | |
| "rewards/rejected": -18.15731430053711, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.811858608893957e-06, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -0.6875, | |
| "logps/chosen": -264.0, | |
| "logps/rejected": -238.0, | |
| "loss": 0.0112, | |
| "rewards/accuracies": 0.9976562261581421, | |
| "rewards/chosen": -2.7115814685821533, | |
| "rewards/margins": 14.367854118347168, | |
| "rewards/rejected": -17.079435348510742, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 4.75484606613455e-06, | |
| "logits/chosen": -1.203125, | |
| "logits/rejected": -0.7265625, | |
| "logps/chosen": -260.0, | |
| "logps/rejected": -238.0, | |
| "loss": 0.0034, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.354907274246216, | |
| "rewards/margins": 14.6588773727417, | |
| "rewards/rejected": -17.013784408569336, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 4.697833523375142e-06, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -0.69921875, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -246.0, | |
| "loss": 0.0047, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.840439796447754, | |
| "rewards/margins": 14.848834037780762, | |
| "rewards/rejected": -17.68927574157715, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.640820980615735e-06, | |
| "logits/chosen": -1.1015625, | |
| "logits/rejected": -0.60546875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -262.0, | |
| "loss": 0.0029, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.4766571521759033, | |
| "rewards/margins": 15.726774215698242, | |
| "rewards/rejected": -19.203433990478516, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.583808437856328e-06, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -0.58984375, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -262.0, | |
| "loss": 0.0043, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -3.400480270385742, | |
| "rewards/margins": 15.654681205749512, | |
| "rewards/rejected": -19.05516242980957, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.526795895096921e-06, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.5625, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -253.0, | |
| "loss": 0.003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4995570182800293, | |
| "rewards/margins": 14.704421997070312, | |
| "rewards/rejected": -18.2039794921875, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.469783352337514e-06, | |
| "logits/chosen": -1.046875, | |
| "logits/rejected": -0.5234375, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -252.0, | |
| "loss": 0.0025, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.058525323867798, | |
| "rewards/margins": 15.031721115112305, | |
| "rewards/rejected": -18.090246200561523, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.412770809578107e-06, | |
| "logits/chosen": -1.03125, | |
| "logits/rejected": -0.46875, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -254.0, | |
| "loss": 0.0035, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.96889066696167, | |
| "rewards/margins": 15.424901962280273, | |
| "rewards/rejected": -18.393795013427734, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 4.3557582668187004e-06, | |
| "logits/chosen": -1.0, | |
| "logits/rejected": -0.47265625, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -254.0, | |
| "loss": 0.0104, | |
| "rewards/accuracies": 0.996874988079071, | |
| "rewards/chosen": -3.0877561569213867, | |
| "rewards/margins": 15.393818855285645, | |
| "rewards/rejected": -18.4815731048584, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 4.2987457240592934e-06, | |
| "logits/chosen": -0.99609375, | |
| "logits/rejected": -0.490234375, | |
| "logps/chosen": -266.0, | |
| "logps/rejected": -251.0, | |
| "loss": 0.0097, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -2.89310884475708, | |
| "rewards/margins": 15.272462844848633, | |
| "rewards/rejected": -18.165569305419922, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 4.2417331812998865e-06, | |
| "logits/chosen": -1.0234375, | |
| "logits/rejected": -0.45703125, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -253.0, | |
| "loss": 0.0044, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.787309169769287, | |
| "rewards/margins": 15.575048446655273, | |
| "rewards/rejected": -18.362356185913086, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 4.1847206385404795e-06, | |
| "logits/chosen": -1.0078125, | |
| "logits/rejected": -0.458984375, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -244.0, | |
| "loss": 0.0127, | |
| "rewards/accuracies": 0.99609375, | |
| "rewards/chosen": -2.9177849292755127, | |
| "rewards/margins": 14.414236068725586, | |
| "rewards/rejected": -17.33202362060547, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 4.1277080957810725e-06, | |
| "logits/chosen": -1.0234375, | |
| "logits/rejected": -0.462890625, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -247.0, | |
| "loss": 0.0024, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.717505931854248, | |
| "rewards/margins": 14.872530937194824, | |
| "rewards/rejected": -17.590036392211914, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 4.0706955530216655e-06, | |
| "logits/chosen": -0.9921875, | |
| "logits/rejected": -0.4140625, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -248.0, | |
| "loss": 0.0038, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -2.782890796661377, | |
| "rewards/margins": 15.000226020812988, | |
| "rewards/rejected": -17.78311538696289, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.0136830102622585e-06, | |
| "logits/chosen": -0.9296875, | |
| "logits/rejected": -0.357421875, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -262.0, | |
| "loss": 0.0033, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -3.5591626167297363, | |
| "rewards/margins": 15.839828491210938, | |
| "rewards/rejected": -19.398990631103516, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.956670467502851e-06, | |
| "logits/chosen": -0.98046875, | |
| "logits/rejected": -0.384765625, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -266.0, | |
| "loss": 0.0036, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.5741126537323, | |
| "rewards/margins": 15.891105651855469, | |
| "rewards/rejected": -19.465219497680664, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 3.899657924743444e-06, | |
| "logits/chosen": -1.046875, | |
| "logits/rejected": -0.462890625, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -260.0, | |
| "loss": 0.0023, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.375121593475342, | |
| "rewards/margins": 15.77069091796875, | |
| "rewards/rejected": -19.145811080932617, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.842645381984037e-06, | |
| "logits/chosen": -1.0390625, | |
| "logits/rejected": -0.474609375, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -258.0, | |
| "loss": 0.0031, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.2375900745391846, | |
| "rewards/margins": 15.49463176727295, | |
| "rewards/rejected": -18.732221603393555, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.7856328392246298e-06, | |
| "logits/chosen": -1.046875, | |
| "logits/rejected": -0.46484375, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -247.0, | |
| "loss": 0.0109, | |
| "rewards/accuracies": 0.9976562261581421, | |
| "rewards/chosen": -2.936633348464966, | |
| "rewards/margins": 15.029220581054688, | |
| "rewards/rejected": -17.96585464477539, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 3.7286202964652224e-06, | |
| "logits/chosen": -0.984375, | |
| "logits/rejected": -0.44921875, | |
| "logps/chosen": -288.0, | |
| "logps/rejected": -262.0, | |
| "loss": 0.0026, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.5291266441345215, | |
| "rewards/margins": 15.453519821166992, | |
| "rewards/rejected": -18.982646942138672, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 3.6716077537058154e-06, | |
| "logits/chosen": -0.921875, | |
| "logits/rejected": -0.375, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -270.0, | |
| "loss": 0.0024, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.0102667808532715, | |
| "rewards/margins": 16.090139389038086, | |
| "rewards/rejected": -20.100406646728516, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 3.6145952109464084e-06, | |
| "logits/chosen": -1.015625, | |
| "logits/rejected": -0.419921875, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -264.0, | |
| "loss": 0.0039, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.3913803100585938, | |
| "rewards/margins": 15.949442863464355, | |
| "rewards/rejected": -19.340822219848633, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.5575826681870014e-06, | |
| "logits/chosen": -0.9921875, | |
| "logits/rejected": -0.3671875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -264.0, | |
| "loss": 0.0062, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.4078826904296875, | |
| "rewards/margins": 16.163570404052734, | |
| "rewards/rejected": -19.571453094482422, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 3.500570125427594e-06, | |
| "logits/chosen": -0.91796875, | |
| "logits/rejected": -0.318359375, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -270.0, | |
| "loss": 0.0049, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.7832558155059814, | |
| "rewards/margins": 16.340469360351562, | |
| "rewards/rejected": -20.12372398376465, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 3.443557582668187e-06, | |
| "logits/chosen": -0.99609375, | |
| "logits/rejected": -0.404296875, | |
| "logps/chosen": -272.0, | |
| "logps/rejected": -264.0, | |
| "loss": 0.0009, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.330648422241211, | |
| "rewards/margins": 15.921122550964355, | |
| "rewards/rejected": -19.25177001953125, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 3.38654503990878e-06, | |
| "logits/chosen": -0.97265625, | |
| "logits/rejected": -0.34375, | |
| "logps/chosen": -292.0, | |
| "logps/rejected": -270.0, | |
| "loss": 0.0027, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -3.5318961143493652, | |
| "rewards/margins": 16.236053466796875, | |
| "rewards/rejected": -19.767948150634766, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 3.329532497149373e-06, | |
| "logits/chosen": -1.015625, | |
| "logits/rejected": -0.396484375, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -264.0, | |
| "loss": 0.0035, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.347155809402466, | |
| "rewards/margins": 16.07892608642578, | |
| "rewards/rejected": -19.426082611083984, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 3.272519954389966e-06, | |
| "logits/chosen": -0.99609375, | |
| "logits/rejected": -0.359375, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -268.0, | |
| "loss": 0.0029, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.578090190887451, | |
| "rewards/margins": 16.344289779663086, | |
| "rewards/rejected": -19.922382354736328, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.215507411630559e-06, | |
| "logits/chosen": -0.94921875, | |
| "logits/rejected": -0.349609375, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -274.0, | |
| "loss": 0.0065, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.7228190898895264, | |
| "rewards/margins": 16.553964614868164, | |
| "rewards/rejected": -20.276782989501953, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.158494868871152e-06, | |
| "logits/chosen": -0.99609375, | |
| "logits/rejected": -0.375, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -264.0, | |
| "loss": 0.0023, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.2026283740997314, | |
| "rewards/margins": 16.301708221435547, | |
| "rewards/rejected": -19.504337310791016, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 3.1014823261117447e-06, | |
| "logits/chosen": -0.96484375, | |
| "logits/rejected": -0.318359375, | |
| "logps/chosen": -286.0, | |
| "logps/rejected": -280.0, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.813472270965576, | |
| "rewards/margins": 17.21391487121582, | |
| "rewards/rejected": -21.027387619018555, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 3.0444697833523378e-06, | |
| "logits/chosen": -0.97265625, | |
| "logits/rejected": -0.29296875, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.002, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -3.973036527633667, | |
| "rewards/margins": 17.74779510498047, | |
| "rewards/rejected": -21.7208309173584, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 2.9874572405929308e-06, | |
| "logits/chosen": -0.96875, | |
| "logits/rejected": -0.29296875, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -282.0, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.1294169425964355, | |
| "rewards/margins": 17.12251853942871, | |
| "rewards/rejected": -21.251937866210938, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 2.930444697833524e-06, | |
| "logits/chosen": -1.03125, | |
| "logits/rejected": -0.34375, | |
| "logps/chosen": -290.0, | |
| "logps/rejected": -294.0, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.9173831939697266, | |
| "rewards/margins": 18.43880844116211, | |
| "rewards/rejected": -22.356189727783203, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 2.873432155074117e-06, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.361328125, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7644851207733154, | |
| "rewards/margins": 17.8516845703125, | |
| "rewards/rejected": -21.61617088317871, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 2.81641961231471e-06, | |
| "logits/chosen": -0.9921875, | |
| "logits/rejected": -0.32421875, | |
| "logps/chosen": -286.0, | |
| "logps/rejected": -282.0, | |
| "loss": 0.0013, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.8479247093200684, | |
| "rewards/margins": 17.440242767333984, | |
| "rewards/rejected": -21.28816795349121, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 2.759407069555302e-06, | |
| "logits/chosen": -1.0390625, | |
| "logits/rejected": -0.404296875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -282.0, | |
| "loss": 0.0009, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.8636913299560547, | |
| "rewards/margins": 17.501983642578125, | |
| "rewards/rejected": -21.365676879882812, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 2.702394526795895e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.4375, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -280.0, | |
| "loss": 0.0005, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7761924266815186, | |
| "rewards/margins": 17.286312103271484, | |
| "rewards/rejected": -21.062503814697266, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 2.645381984036488e-06, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.408203125, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -280.0, | |
| "loss": 0.0005, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7179946899414062, | |
| "rewards/margins": 17.326318740844727, | |
| "rewards/rejected": -21.0443115234375, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 2.588369441277081e-06, | |
| "logits/chosen": -1.1015625, | |
| "logits/rejected": -0.453125, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -278.0, | |
| "loss": 0.0009, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4939072132110596, | |
| "rewards/margins": 17.304866790771484, | |
| "rewards/rejected": -20.79877281188965, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 2.531356898517674e-06, | |
| "logits/chosen": -1.078125, | |
| "logits/rejected": -0.41015625, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -284.0, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.338261842727661, | |
| "rewards/margins": 17.951480865478516, | |
| "rewards/rejected": -21.289743423461914, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 2.474344355758267e-06, | |
| "logits/chosen": -1.0234375, | |
| "logits/rejected": -0.3671875, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.018970012664795, | |
| "rewards/margins": 17.885696411132812, | |
| "rewards/rejected": -21.904666900634766, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 2.4173318129988597e-06, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -0.37890625, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -284.0, | |
| "loss": 0.0012, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.494691848754883, | |
| "rewards/margins": 17.86090850830078, | |
| "rewards/rejected": -21.355600357055664, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 2.3603192702394527e-06, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.34375, | |
| "logps/chosen": -288.0, | |
| "logps/rejected": -288.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.877596378326416, | |
| "rewards/margins": 17.996723175048828, | |
| "rewards/rejected": -21.874319076538086, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 2.3033067274800457e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.419921875, | |
| "logps/chosen": -270.0, | |
| "logps/rejected": -282.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.740429401397705, | |
| "rewards/margins": 17.418020248413086, | |
| "rewards/rejected": -21.158451080322266, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 2.2462941847206388e-06, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.38671875, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -298.0, | |
| "loss": 0.0005, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.9601192474365234, | |
| "rewards/margins": 18.503660202026367, | |
| "rewards/rejected": -22.46377944946289, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 2.1892816419612318e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.3828125, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7733867168426514, | |
| "rewards/margins": 18.210254669189453, | |
| "rewards/rejected": -21.983644485473633, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 2.132269099201825e-06, | |
| "logits/chosen": -1.1171875, | |
| "logits/rejected": -0.4375, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0015, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.738826036453247, | |
| "rewards/margins": 17.777729034423828, | |
| "rewards/rejected": -21.51655387878418, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 2.075256556442418e-06, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -0.4140625, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -284.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.708034038543701, | |
| "rewards/margins": 17.657176971435547, | |
| "rewards/rejected": -21.365211486816406, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 2.0182440136830104e-06, | |
| "logits/chosen": -1.03125, | |
| "logits/rejected": -0.3359375, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.185555934906006, | |
| "rewards/margins": 18.033720016479492, | |
| "rewards/rejected": -22.219276428222656, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 1.9612314709236034e-06, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.361328125, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -288.0, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.9414119720458984, | |
| "rewards/margins": 18.10892677307129, | |
| "rewards/rejected": -22.050336837768555, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 1.9042189281641962e-06, | |
| "logits/chosen": -1.0703125, | |
| "logits/rejected": -0.37109375, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0011, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.8034446239471436, | |
| "rewards/margins": 18.498294830322266, | |
| "rewards/rejected": -22.301740646362305, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1.8472063854047893e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.39453125, | |
| "logps/chosen": -286.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.825915813446045, | |
| "rewards/margins": 18.246068954467773, | |
| "rewards/rejected": -22.07198715209961, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 1.790193842645382e-06, | |
| "logits/chosen": -1.0703125, | |
| "logits/rejected": -0.369140625, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7209835052490234, | |
| "rewards/margins": 18.366897583007812, | |
| "rewards/rejected": -22.087881088256836, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.733181299885975e-06, | |
| "logits/chosen": -1.078125, | |
| "logits/rejected": -0.392578125, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.002, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.494877576828003, | |
| "rewards/margins": 18.009967803955078, | |
| "rewards/rejected": -21.504846572875977, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 1.6761687571265681e-06, | |
| "logits/chosen": -1.0703125, | |
| "logits/rejected": -0.349609375, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0005, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.914118528366089, | |
| "rewards/margins": 18.189693450927734, | |
| "rewards/rejected": -22.10381317138672, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 1.6191562143671611e-06, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -0.37890625, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.6217734813690186, | |
| "rewards/margins": 18.476469039916992, | |
| "rewards/rejected": -22.098241806030273, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 1.5621436716077537e-06, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -0.392578125, | |
| "logps/chosen": -292.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.6812424659729004, | |
| "rewards/margins": 18.31012725830078, | |
| "rewards/rejected": -21.991369247436523, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 1.5051311288483467e-06, | |
| "logits/chosen": -1.0703125, | |
| "logits/rejected": -0.35546875, | |
| "logps/chosen": -290.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0019, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.9753355979919434, | |
| "rewards/margins": 18.298744201660156, | |
| "rewards/rejected": -22.27408218383789, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 1.4481185860889396e-06, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -0.384765625, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -284.0, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4559783935546875, | |
| "rewards/margins": 18.017427444458008, | |
| "rewards/rejected": -21.473405838012695, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.3911060433295326e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.37109375, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4795451164245605, | |
| "rewards/margins": 18.1534423828125, | |
| "rewards/rejected": -21.63298797607422, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 1.3340935005701256e-06, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.345703125, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0009, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.496410369873047, | |
| "rewards/margins": 18.135766983032227, | |
| "rewards/rejected": -21.63217544555664, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 1.2770809578107184e-06, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.3515625, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -288.0, | |
| "loss": 0.001, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.894739866256714, | |
| "rewards/margins": 17.927284240722656, | |
| "rewards/rejected": -21.822025299072266, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 1.2200684150513114e-06, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.3359375, | |
| "logps/chosen": -292.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0042, | |
| "rewards/accuracies": 0.9984375238418579, | |
| "rewards/chosen": -3.9973883628845215, | |
| "rewards/margins": 18.207622528076172, | |
| "rewards/rejected": -22.205013275146484, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.1630558722919042e-06, | |
| "logits/chosen": -1.0703125, | |
| "logits/rejected": -0.37109375, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -284.0, | |
| "loss": 0.0034, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.779940128326416, | |
| "rewards/margins": 17.809322357177734, | |
| "rewards/rejected": -21.589258193969727, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.1060433295324972e-06, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.369140625, | |
| "logps/chosen": -286.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.665940761566162, | |
| "rewards/margins": 18.151546478271484, | |
| "rewards/rejected": -21.817487716674805, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.0490307867730903e-06, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.337890625, | |
| "logps/chosen": -288.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.9725680351257324, | |
| "rewards/margins": 18.010974884033203, | |
| "rewards/rejected": -21.983543395996094, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 9.92018244013683e-07, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -0.36328125, | |
| "logps/chosen": -294.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.550889253616333, | |
| "rewards/margins": 18.127422332763672, | |
| "rewards/rejected": -21.67831039428711, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 9.35005701254276e-07, | |
| "logits/chosen": -1.0234375, | |
| "logits/rejected": -0.30859375, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -282.0, | |
| "loss": 0.0015, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.5743343830108643, | |
| "rewards/margins": 17.971210479736328, | |
| "rewards/rejected": -21.545543670654297, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 8.779931584948689e-07, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.3515625, | |
| "logps/chosen": -276.0, | |
| "logps/rejected": -292.0, | |
| "loss": 0.0018, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -3.8023428916931152, | |
| "rewards/margins": 18.363157272338867, | |
| "rewards/rejected": -22.16550064086914, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 8.209806157354619e-07, | |
| "logits/chosen": -1.015625, | |
| "logits/rejected": -0.353515625, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -296.0, | |
| "loss": 0.0007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.0028395652771, | |
| "rewards/margins": 18.4693603515625, | |
| "rewards/rejected": -22.47220230102539, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 7.639680729760548e-07, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.337890625, | |
| "logps/chosen": -286.0, | |
| "logps/rejected": -288.0, | |
| "loss": 0.0008, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7589478492736816, | |
| "rewards/margins": 18.027183532714844, | |
| "rewards/rejected": -21.786130905151367, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 7.069555302166476e-07, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.373046875, | |
| "logps/chosen": -274.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.6975696086883545, | |
| "rewards/margins": 18.218469619750977, | |
| "rewards/rejected": -21.916040420532227, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 6.499429874572407e-07, | |
| "logits/chosen": -1.046875, | |
| "logits/rejected": -0.34375, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -290.0, | |
| "loss": 0.0013, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.9434094429016113, | |
| "rewards/margins": 18.002216339111328, | |
| "rewards/rejected": -21.94562339782715, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 5.929304446978336e-07, | |
| "logits/chosen": -1.046875, | |
| "logits/rejected": -0.330078125, | |
| "logps/chosen": -278.0, | |
| "logps/rejected": -288.0, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.605856418609619, | |
| "rewards/margins": 18.189517974853516, | |
| "rewards/rejected": -21.79537582397461, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 5.359179019384265e-07, | |
| "logits/chosen": -1.0625, | |
| "logits/rejected": -0.34375, | |
| "logps/chosen": -284.0, | |
| "logps/rejected": -298.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.889907121658325, | |
| "rewards/margins": 18.669677734375, | |
| "rewards/rejected": -22.55958366394043, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 4.789053591790194e-07, | |
| "logits/chosen": -1.0546875, | |
| "logits/rejected": -0.33984375, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -294.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.840752124786377, | |
| "rewards/margins": 18.512104034423828, | |
| "rewards/rejected": -22.352855682373047, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.218928164196123e-07, | |
| "logits/chosen": -1.0390625, | |
| "logits/rejected": -0.341796875, | |
| "logps/chosen": -280.0, | |
| "logps/rejected": -294.0, | |
| "loss": 0.0015, | |
| "rewards/accuracies": 0.999218761920929, | |
| "rewards/chosen": -4.100388526916504, | |
| "rewards/margins": 18.11039161682129, | |
| "rewards/rejected": -22.21078109741211, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 3.648802736602053e-07, | |
| "logits/chosen": -1.0234375, | |
| "logits/rejected": -0.341796875, | |
| "logps/chosen": -268.0, | |
| "logps/rejected": -286.0, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7678654193878174, | |
| "rewards/margins": 17.92337417602539, | |
| "rewards/rejected": -21.691240310668945, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 3.078677309007982e-07, | |
| "logits/chosen": -1.0390625, | |
| "logits/rejected": -0.33984375, | |
| "logps/chosen": -282.0, | |
| "logps/rejected": -294.0, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.8660149574279785, | |
| "rewards/margins": 18.40440559387207, | |
| "rewards/rejected": -22.27042007446289, | |
| "step": 4400 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 4508, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 200, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |