| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.93687707641196, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0664451827242525, | |
| "grad_norm": 40.747467041015625, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "logits/chosen": -0.12511932849884033, | |
| "logits/rejected": -0.18846073746681213, | |
| "logps/chosen": -1117.1700439453125, | |
| "logps/rejected": -1311.204345703125, | |
| "loss": 0.7232, | |
| "rewards/accuracies": 0.375, | |
| "rewards/chosen": 0.004250399302691221, | |
| "rewards/margins": -0.0051729860715568066, | |
| "rewards/rejected": 0.00942337978631258, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.132890365448505, | |
| "grad_norm": 63.1617546081543, | |
| "learning_rate": 6.666666666666667e-07, | |
| "logits/chosen": -0.1253906637430191, | |
| "logits/rejected": -0.1845361292362213, | |
| "logps/chosen": -1105.6373291015625, | |
| "logps/rejected": -1318.8868408203125, | |
| "loss": 0.7263, | |
| "rewards/accuracies": 0.5031250715255737, | |
| "rewards/chosen": -0.03344225138425827, | |
| "rewards/margins": 0.015075692906975746, | |
| "rewards/rejected": -0.04851795360445976, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19933554817275748, | |
| "grad_norm": 35.549259185791016, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "logits/chosen": -0.13364776968955994, | |
| "logits/rejected": -0.1953742951154709, | |
| "logps/chosen": -1098.7236328125, | |
| "logps/rejected": -1286.4495849609375, | |
| "loss": 0.7018, | |
| "rewards/accuracies": 0.515625, | |
| "rewards/chosen": -0.17741113901138306, | |
| "rewards/margins": 0.07103082537651062, | |
| "rewards/rejected": -0.24844194948673248, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.26578073089701, | |
| "grad_norm": 38.198516845703125, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "logits/chosen": -0.13651709258556366, | |
| "logits/rejected": -0.19466470181941986, | |
| "logps/chosen": -1057.7530517578125, | |
| "logps/rejected": -1252.78173828125, | |
| "loss": 0.6744, | |
| "rewards/accuracies": 0.515625, | |
| "rewards/chosen": -0.45364782214164734, | |
| "rewards/margins": 0.16406498849391937, | |
| "rewards/rejected": -0.6177127957344055, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.33222591362126247, | |
| "grad_norm": 27.63871955871582, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "logits/chosen": -0.13217487931251526, | |
| "logits/rejected": -0.19238920509815216, | |
| "logps/chosen": -1069.2860107421875, | |
| "logps/rejected": -1261.974853515625, | |
| "loss": 0.6492, | |
| "rewards/accuracies": 0.589062511920929, | |
| "rewards/chosen": -0.8812439441680908, | |
| "rewards/margins": 0.36069372296333313, | |
| "rewards/rejected": -1.2419377565383911, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.39867109634551495, | |
| "grad_norm": 21.243654251098633, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "logits/chosen": -0.13583233952522278, | |
| "logits/rejected": -0.1979021281003952, | |
| "logps/chosen": -1059.2923583984375, | |
| "logps/rejected": -1271.649169921875, | |
| "loss": 0.6293, | |
| "rewards/accuracies": 0.6343749761581421, | |
| "rewards/chosen": -0.9827756881713867, | |
| "rewards/margins": 0.5085424780845642, | |
| "rewards/rejected": -1.4913183450698853, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.46511627906976744, | |
| "grad_norm": 19.875577926635742, | |
| "learning_rate": 2.3333333333333336e-06, | |
| "logits/chosen": -0.13559813797473907, | |
| "logits/rejected": -0.19648894667625427, | |
| "logps/chosen": -1103.181396484375, | |
| "logps/rejected": -1314.6522216796875, | |
| "loss": 0.5424, | |
| "rewards/accuracies": 0.707812488079071, | |
| "rewards/chosen": -0.8136237859725952, | |
| "rewards/margins": 0.6972610950469971, | |
| "rewards/rejected": -1.5108850002288818, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.53156146179402, | |
| "grad_norm": 20.229171752929688, | |
| "learning_rate": 2.666666666666667e-06, | |
| "logits/chosen": -0.11205008625984192, | |
| "logits/rejected": -0.19006861746311188, | |
| "logps/chosen": -1065.726806640625, | |
| "logps/rejected": -1319.034423828125, | |
| "loss": 0.4856, | |
| "rewards/accuracies": 0.7828124761581421, | |
| "rewards/chosen": -0.6954863667488098, | |
| "rewards/margins": 1.0426087379455566, | |
| "rewards/rejected": -1.7380950450897217, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5980066445182725, | |
| "grad_norm": 17.880022048950195, | |
| "learning_rate": 3e-06, | |
| "logits/chosen": -0.12407304346561432, | |
| "logits/rejected": -0.18850643932819366, | |
| "logps/chosen": -1076.2564697265625, | |
| "logps/rejected": -1331.6634521484375, | |
| "loss": 0.4188, | |
| "rewards/accuracies": 0.840624988079071, | |
| "rewards/chosen": -0.3792995810508728, | |
| "rewards/margins": 1.265782117843628, | |
| "rewards/rejected": -1.645081639289856, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6644518272425249, | |
| "grad_norm": 12.63503646850586, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "logits/chosen": -0.121916763484478, | |
| "logits/rejected": -0.19661013782024384, | |
| "logps/chosen": -1058.5997314453125, | |
| "logps/rejected": -1334.9495849609375, | |
| "loss": 0.3597, | |
| "rewards/accuracies": 0.8609374761581421, | |
| "rewards/chosen": -0.5200275182723999, | |
| "rewards/margins": 1.9380321502685547, | |
| "rewards/rejected": -2.458059549331665, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7308970099667774, | |
| "grad_norm": 14.828380584716797, | |
| "learning_rate": 3.6666666666666666e-06, | |
| "logits/chosen": -0.13187702000141144, | |
| "logits/rejected": -0.19965648651123047, | |
| "logps/chosen": -1074.6199951171875, | |
| "logps/rejected": -1312.4927978515625, | |
| "loss": 0.2984, | |
| "rewards/accuracies": 0.90625, | |
| "rewards/chosen": -0.4199390709400177, | |
| "rewards/margins": 2.6119236946105957, | |
| "rewards/rejected": -3.031863212585449, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7973421926910299, | |
| "grad_norm": 10.107036590576172, | |
| "learning_rate": 4.000000000000001e-06, | |
| "logits/chosen": -0.12684178352355957, | |
| "logits/rejected": -0.1979806274175644, | |
| "logps/chosen": -1091.7188720703125, | |
| "logps/rejected": -1349.757568359375, | |
| "loss": 0.2236, | |
| "rewards/accuracies": 0.9312500357627869, | |
| "rewards/chosen": -0.07451216131448746, | |
| "rewards/margins": 4.150519847869873, | |
| "rewards/rejected": -4.225031852722168, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8637873754152824, | |
| "grad_norm": 4.532182693481445, | |
| "learning_rate": 4.333333333333334e-06, | |
| "logits/chosen": -0.13108162581920624, | |
| "logits/rejected": -0.19549141824245453, | |
| "logps/chosen": -1100.24951171875, | |
| "logps/rejected": -1363.146728515625, | |
| "loss": 0.1608, | |
| "rewards/accuracies": 0.9468750357627869, | |
| "rewards/chosen": 0.3449128270149231, | |
| "rewards/margins": 5.669872760772705, | |
| "rewards/rejected": -5.324959754943848, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9302325581395349, | |
| "grad_norm": 2.4136996269226074, | |
| "learning_rate": 4.666666666666667e-06, | |
| "logits/chosen": -0.13807465136051178, | |
| "logits/rejected": -0.20656287670135498, | |
| "logps/chosen": -1032.328369140625, | |
| "logps/rejected": -1357.612548828125, | |
| "loss": 0.1282, | |
| "rewards/accuracies": 0.9437499642372131, | |
| "rewards/chosen": 1.2508221864700317, | |
| "rewards/margins": 7.064514636993408, | |
| "rewards/rejected": -5.813692092895508, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9966777408637874, | |
| "grad_norm": 22.84921646118164, | |
| "learning_rate": 5e-06, | |
| "logits/chosen": -0.1460917443037033, | |
| "logits/rejected": -0.21644163131713867, | |
| "logps/chosen": -1055.5206298828125, | |
| "logps/rejected": -1391.220947265625, | |
| "loss": 0.1053, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 0.7099586725234985, | |
| "rewards/margins": 9.3050537109375, | |
| "rewards/rejected": -8.59509563446045, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0598006644518272, | |
| "grad_norm": 5.813467502593994, | |
| "learning_rate": 4.999323102948655e-06, | |
| "logits/chosen": -0.14507660269737244, | |
| "logits/rejected": -0.21615180373191833, | |
| "logps/chosen": -1054.1131591796875, | |
| "logps/rejected": -1392.970947265625, | |
| "loss": 0.0871, | |
| "rewards/accuracies": 0.9407894611358643, | |
| "rewards/chosen": 1.4173532724380493, | |
| "rewards/margins": 11.686441421508789, | |
| "rewards/rejected": -10.269088745117188, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.1262458471760797, | |
| "grad_norm": 25.732624053955078, | |
| "learning_rate": 4.997292778346312e-06, | |
| "logits/chosen": -0.1539006382226944, | |
| "logits/rejected": -0.21280081570148468, | |
| "logps/chosen": -1064.2578125, | |
| "logps/rejected": -1432.212646484375, | |
| "loss": 0.0596, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": 1.7107740640640259, | |
| "rewards/margins": 15.027220726013184, | |
| "rewards/rejected": -13.316445350646973, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1926910299003322, | |
| "grad_norm": 2.911512851715088, | |
| "learning_rate": 4.993910125649561e-06, | |
| "logits/chosen": -0.13852202892303467, | |
| "logits/rejected": -0.2171792984008789, | |
| "logps/chosen": -992.4456787109375, | |
| "logps/rejected": -1381.4122314453125, | |
| "loss": 0.0456, | |
| "rewards/accuracies": 0.964062511920929, | |
| "rewards/chosen": 3.4778242111206055, | |
| "rewards/margins": 15.223333358764648, | |
| "rewards/rejected": -11.745508193969727, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2591362126245846, | |
| "grad_norm": 0.5809551477432251, | |
| "learning_rate": 4.989176976624511e-06, | |
| "logits/chosen": -0.16654856503009796, | |
| "logits/rejected": -0.23170927166938782, | |
| "logps/chosen": -1053.9530029296875, | |
| "logps/rejected": -1434.646484375, | |
| "loss": 0.0562, | |
| "rewards/accuracies": 0.940625011920929, | |
| "rewards/chosen": 3.07829213142395, | |
| "rewards/margins": 18.499500274658203, | |
| "rewards/rejected": -15.421209335327148, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3255813953488373, | |
| "grad_norm": 4.461428165435791, | |
| "learning_rate": 4.983095894354858e-06, | |
| "logits/chosen": -0.17254643142223358, | |
| "logits/rejected": -0.23220513761043549, | |
| "logps/chosen": -1038.1956787109375, | |
| "logps/rejected": -1425.747802734375, | |
| "loss": 0.0551, | |
| "rewards/accuracies": 0.9453125, | |
| "rewards/chosen": 4.759212493896484, | |
| "rewards/margins": 18.334705352783203, | |
| "rewards/rejected": -13.575494766235352, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3920265780730898, | |
| "grad_norm": 0.13902103900909424, | |
| "learning_rate": 4.975670171853926e-06, | |
| "logits/chosen": -0.1616649478673935, | |
| "logits/rejected": -0.22982370853424072, | |
| "logps/chosen": -1068.1690673828125, | |
| "logps/rejected": -1542.0587158203125, | |
| "loss": 0.0766, | |
| "rewards/accuracies": 0.9515624642372131, | |
| "rewards/chosen": 1.5794833898544312, | |
| "rewards/margins": 24.19274139404297, | |
| "rewards/rejected": -22.613258361816406, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4584717607973423, | |
| "grad_norm": 111.07405090332031, | |
| "learning_rate": 4.966903830281449e-06, | |
| "logits/chosen": -0.1702309548854828, | |
| "logits/rejected": -0.23228994011878967, | |
| "logps/chosen": -1009.6397705078125, | |
| "logps/rejected": -1450.9083251953125, | |
| "loss": 0.0614, | |
| "rewards/accuracies": 0.9468750357627869, | |
| "rewards/chosen": 4.363029479980469, | |
| "rewards/margins": 21.61128044128418, | |
| "rewards/rejected": -17.248252868652344, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.5249169435215948, | |
| "grad_norm": 1.088158369064331, | |
| "learning_rate": 4.956801616766033e-06, | |
| "logits/chosen": -0.17264895141124725, | |
| "logits/rejected": -0.22918781638145447, | |
| "logps/chosen": -1067.2901611328125, | |
| "logps/rejected": -1552.646728515625, | |
| "loss": 0.0404, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 3.915598154067993, | |
| "rewards/margins": 26.750593185424805, | |
| "rewards/rejected": -22.83499526977539, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5913621262458473, | |
| "grad_norm": 1.4660481214523315, | |
| "learning_rate": 4.9453690018345144e-06, | |
| "logits/chosen": -0.17638970911502838, | |
| "logits/rejected": -0.23336604237556458, | |
| "logps/chosen": -1033.65771484375, | |
| "logps/rejected": -1489.400146484375, | |
| "loss": 0.0358, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 4.591525554656982, | |
| "rewards/margins": 24.92582893371582, | |
| "rewards/rejected": -20.334304809570312, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6578073089700998, | |
| "grad_norm": 0.6518616080284119, | |
| "learning_rate": 4.93261217644956e-06, | |
| "logits/chosen": -0.16551312804222107, | |
| "logits/rejected": -0.2351105809211731, | |
| "logps/chosen": -1069.6429443359375, | |
| "logps/rejected": -1581.255126953125, | |
| "loss": 0.0554, | |
| "rewards/accuracies": 0.9656250476837158, | |
| "rewards/chosen": 2.4628138542175293, | |
| "rewards/margins": 27.925844192504883, | |
| "rewards/rejected": -25.46302604675293, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7242524916943522, | |
| "grad_norm": 0.4813426434993744, | |
| "learning_rate": 4.91853804865716e-06, | |
| "logits/chosen": -0.18384608626365662, | |
| "logits/rejected": -0.24281182885169983, | |
| "logps/chosen": -968.432861328125, | |
| "logps/rejected": -1347.4447021484375, | |
| "loss": 0.0351, | |
| "rewards/accuracies": 0.9593750238418579, | |
| "rewards/chosen": 7.137730121612549, | |
| "rewards/margins": 21.095685958862305, | |
| "rewards/rejected": -13.957955360412598, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.7906976744186047, | |
| "grad_norm": 11.40811824798584, | |
| "learning_rate": 4.903154239845798e-06, | |
| "logits/chosen": -0.1812049150466919, | |
| "logits/rejected": -0.25064709782600403, | |
| "logps/chosen": -1056.5556640625, | |
| "logps/rejected": -1586.718017578125, | |
| "loss": 0.0393, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 3.66015625, | |
| "rewards/margins": 30.29279899597168, | |
| "rewards/rejected": -26.632644653320312, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 0.7297212481498718, | |
| "learning_rate": 4.88646908061933e-06, | |
| "logits/chosen": -0.18657241761684418, | |
| "logits/rejected": -0.2551916539669037, | |
| "logps/chosen": -968.1968383789062, | |
| "logps/rejected": -1500.6644287109375, | |
| "loss": 0.0391, | |
| "rewards/accuracies": 0.948437511920929, | |
| "rewards/chosen": 6.839397430419922, | |
| "rewards/margins": 27.143301010131836, | |
| "rewards/rejected": -20.303905487060547, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9235880398671097, | |
| "grad_norm": 0.17506621778011322, | |
| "learning_rate": 4.868491606285823e-06, | |
| "logits/chosen": -0.19025684893131256, | |
| "logits/rejected": -0.25329911708831787, | |
| "logps/chosen": -1026.042236328125, | |
| "logps/rejected": -1535.4869384765625, | |
| "loss": 0.0436, | |
| "rewards/accuracies": 0.9531250596046448, | |
| "rewards/chosen": 6.215095043182373, | |
| "rewards/margins": 29.239473342895508, | |
| "rewards/rejected": -23.024375915527344, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.9900332225913622, | |
| "grad_norm": 0.1333008110523224, | |
| "learning_rate": 4.849231551964771e-06, | |
| "logits/chosen": -0.1948496401309967, | |
| "logits/rejected": -0.25613006949424744, | |
| "logps/chosen": -1025.02783203125, | |
| "logps/rejected": -1517.8956298828125, | |
| "loss": 0.0381, | |
| "rewards/accuracies": 0.9546875953674316, | |
| "rewards/chosen": 7.619020938873291, | |
| "rewards/margins": 27.76497459411621, | |
| "rewards/rejected": -20.145952224731445, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.053156146179402, | |
| "grad_norm": 1.3418188095092773, | |
| "learning_rate": 4.828699347315357e-06, | |
| "logits/chosen": -0.17470988631248474, | |
| "logits/rejected": -0.2475568950176239, | |
| "logps/chosen": -994.5497436523438, | |
| "logps/rejected": -1511.7833251953125, | |
| "loss": 0.0276, | |
| "rewards/accuracies": 0.9621711373329163, | |
| "rewards/chosen": 7.3509602546691895, | |
| "rewards/margins": 28.877328872680664, | |
| "rewards/rejected": -21.526369094848633, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1196013289036544, | |
| "grad_norm": 0.7398647665977478, | |
| "learning_rate": 4.806906110888606e-06, | |
| "logits/chosen": -0.19838787615299225, | |
| "logits/rejected": -0.253930926322937, | |
| "logps/chosen": -1008.02734375, | |
| "logps/rejected": -1496.8922119140625, | |
| "loss": 0.0344, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 6.126870155334473, | |
| "rewards/margins": 30.307435989379883, | |
| "rewards/rejected": -24.180564880371094, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.186046511627907, | |
| "grad_norm": 0.2915615439414978, | |
| "learning_rate": 4.783863644106502e-06, | |
| "logits/chosen": -0.18888217210769653, | |
| "logits/rejected": -0.2543439567089081, | |
| "logps/chosen": -1001.6741333007812, | |
| "logps/rejected": -1547.251953125, | |
| "loss": 0.0294, | |
| "rewards/accuracies": 0.9593749642372131, | |
| "rewards/chosen": 6.256338119506836, | |
| "rewards/margins": 32.92168426513672, | |
| "rewards/rejected": -26.66534423828125, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.2524916943521593, | |
| "grad_norm": 0.44617852568626404, | |
| "learning_rate": 4.759584424871302e-06, | |
| "logits/chosen": -0.19481198489665985, | |
| "logits/rejected": -0.2515893876552582, | |
| "logps/chosen": -1063.011474609375, | |
| "logps/rejected": -1607.3277587890625, | |
| "loss": 0.0264, | |
| "rewards/accuracies": 0.964062511920929, | |
| "rewards/chosen": 7.461002826690674, | |
| "rewards/margins": 33.48613357543945, | |
| "rewards/rejected": -26.025131225585938, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.318936877076412, | |
| "grad_norm": 0.5236634016036987, | |
| "learning_rate": 4.734081600808531e-06, | |
| "logits/chosen": -0.2034038007259369, | |
| "logits/rejected": -0.2618187367916107, | |
| "logps/chosen": -1072.1470947265625, | |
| "logps/rejected": -1632.1077880859375, | |
| "loss": 0.0389, | |
| "rewards/accuracies": 0.9468750357627869, | |
| "rewards/chosen": 6.423084259033203, | |
| "rewards/margins": 35.269798278808594, | |
| "rewards/rejected": -28.846710205078125, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3853820598006643, | |
| "grad_norm": 0.05544855445623398, | |
| "learning_rate": 4.707368982147318e-06, | |
| "logits/chosen": -0.18389806151390076, | |
| "logits/rejected": -0.2554208040237427, | |
| "logps/chosen": -976.6298828125, | |
| "logps/rejected": -1534.483154296875, | |
| "loss": 0.0313, | |
| "rewards/accuracies": 0.9578125476837158, | |
| "rewards/chosen": 6.575656414031982, | |
| "rewards/margins": 32.54267120361328, | |
| "rewards/rejected": -25.967016220092773, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.451827242524917, | |
| "grad_norm": 0.2025989443063736, | |
| "learning_rate": 4.679461034241906e-06, | |
| "logits/chosen": -0.1815650463104248, | |
| "logits/rejected": -0.2540404796600342, | |
| "logps/chosen": -1047.0966796875, | |
| "logps/rejected": -1622.110595703125, | |
| "loss": 0.0276, | |
| "rewards/accuracies": 0.9625000953674316, | |
| "rewards/chosen": 6.719258785247803, | |
| "rewards/margins": 35.81674575805664, | |
| "rewards/rejected": -29.097484588623047, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5182724252491693, | |
| "grad_norm": 0.7222418189048767, | |
| "learning_rate": 4.650372869738415e-06, | |
| "logits/chosen": -0.18268240988254547, | |
| "logits/rejected": -0.2617727220058441, | |
| "logps/chosen": -965.8853759765625, | |
| "logps/rejected": -1582.7860107421875, | |
| "loss": 0.0332, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 7.1367573738098145, | |
| "rewards/margins": 33.75447082519531, | |
| "rewards/rejected": -26.617713928222656, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.584717607973422, | |
| "grad_norm": 0.1559152901172638, | |
| "learning_rate": 4.620120240391065e-06, | |
| "logits/chosen": -0.19453352689743042, | |
| "logits/rejected": -0.2535034418106079, | |
| "logps/chosen": -1019.660888671875, | |
| "logps/rejected": -1637.6041259765625, | |
| "loss": 0.032, | |
| "rewards/accuracies": 0.9640625715255737, | |
| "rewards/chosen": 6.416092395782471, | |
| "rewards/margins": 36.01222610473633, | |
| "rewards/rejected": -29.596132278442383, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.6511627906976747, | |
| "grad_norm": 0.05074378848075867, | |
| "learning_rate": 4.588719528532342e-06, | |
| "logits/chosen": -0.1765459030866623, | |
| "logits/rejected": -0.259499192237854, | |
| "logps/chosen": -942.0123291015625, | |
| "logps/rejected": -1461.3004150390625, | |
| "loss": 0.0334, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 8.615708351135254, | |
| "rewards/margins": 29.26686668395996, | |
| "rewards/rejected": -20.651159286499023, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.717607973421927, | |
| "grad_norm": 0.027149733155965805, | |
| "learning_rate": 4.556187738201656e-06, | |
| "logits/chosen": -0.19329792261123657, | |
| "logits/rejected": -0.2614816427230835, | |
| "logps/chosen": -974.2481689453125, | |
| "logps/rejected": -1543.07177734375, | |
| "loss": 0.0366, | |
| "rewards/accuracies": 0.949999988079071, | |
| "rewards/chosen": 5.765076637268066, | |
| "rewards/margins": 35.13932418823242, | |
| "rewards/rejected": -29.37424087524414, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.7840531561461797, | |
| "grad_norm": 0.07503657788038254, | |
| "learning_rate": 4.522542485937369e-06, | |
| "logits/chosen": -0.19822394847869873, | |
| "logits/rejected": -0.2589871287345886, | |
| "logps/chosen": -1010.1796264648438, | |
| "logps/rejected": -1573.482421875, | |
| "loss": 0.0347, | |
| "rewards/accuracies": 0.9562500715255737, | |
| "rewards/chosen": 6.699889183044434, | |
| "rewards/margins": 35.83893966674805, | |
| "rewards/rejected": -29.13905143737793, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.850498338870432, | |
| "grad_norm": 0.08216100186109543, | |
| "learning_rate": 4.48780199123712e-06, | |
| "logits/chosen": -0.19484762847423553, | |
| "logits/rejected": -0.2583960294723511, | |
| "logps/chosen": -1018.3121337890625, | |
| "logps/rejected": -1550.55615234375, | |
| "loss": 0.0327, | |
| "rewards/accuracies": 0.9546875357627869, | |
| "rewards/chosen": 8.480841636657715, | |
| "rewards/margins": 33.863243103027344, | |
| "rewards/rejected": -25.382402420043945, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.9169435215946846, | |
| "grad_norm": 0.04343642294406891, | |
| "learning_rate": 4.451985066691649e-06, | |
| "logits/chosen": -0.20001523196697235, | |
| "logits/rejected": -0.26353001594543457, | |
| "logps/chosen": -979.586669921875, | |
| "logps/rejected": -1509.11328125, | |
| "loss": 0.0313, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 8.662511825561523, | |
| "rewards/margins": 32.48591995239258, | |
| "rewards/rejected": -23.823410034179688, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.983388704318937, | |
| "grad_norm": 0.056994736194610596, | |
| "learning_rate": 4.415111107797445e-06, | |
| "logits/chosen": -0.1883145421743393, | |
| "logits/rejected": -0.2559402883052826, | |
| "logps/chosen": -1012.8984375, | |
| "logps/rejected": -1519.9129638671875, | |
| "loss": 0.0264, | |
| "rewards/accuracies": 0.9656250476837158, | |
| "rewards/chosen": 8.099644660949707, | |
| "rewards/margins": 34.30991744995117, | |
| "rewards/rejected": -26.21027183532715, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.046511627906977, | |
| "grad_norm": 0.027808276936411858, | |
| "learning_rate": 4.377200082453748e-06, | |
| "logits/chosen": -0.18382969498634338, | |
| "logits/rejected": -0.266806423664093, | |
| "logps/chosen": -933.7220458984375, | |
| "logps/rejected": -1490.690673828125, | |
| "loss": 0.0311, | |
| "rewards/accuracies": 0.9539474248886108, | |
| "rewards/chosen": 7.497554779052734, | |
| "rewards/margins": 33.65736770629883, | |
| "rewards/rejected": -26.159814834594727, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.1129568106312293, | |
| "grad_norm": 0.026268087327480316, | |
| "learning_rate": 4.338272520149572e-06, | |
| "logits/chosen": -0.20292654633522034, | |
| "logits/rejected": -0.27201807498931885, | |
| "logps/chosen": -1006.4375, | |
| "logps/rejected": -1589.052734375, | |
| "loss": 0.033, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 8.268844604492188, | |
| "rewards/margins": 35.16604232788086, | |
| "rewards/rejected": -26.89719581604004, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.179401993355482, | |
| "grad_norm": 0.014455691911280155, | |
| "learning_rate": 4.2983495008466285e-06, | |
| "logits/chosen": -0.17968863248825073, | |
| "logits/rejected": -0.26512834429740906, | |
| "logps/chosen": -972.7686157226562, | |
| "logps/rejected": -1628.4017333984375, | |
| "loss": 0.027, | |
| "rewards/accuracies": 0.9625000357627869, | |
| "rewards/chosen": 7.220202445983887, | |
| "rewards/margins": 38.235836029052734, | |
| "rewards/rejected": -31.01563262939453, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.2458471760797343, | |
| "grad_norm": 1.3661887645721436, | |
| "learning_rate": 4.257452643564155e-06, | |
| "logits/chosen": -0.1976873278617859, | |
| "logits/rejected": -0.26860570907592773, | |
| "logps/chosen": -957.9434814453125, | |
| "logps/rejected": -1552.618408203125, | |
| "loss": 0.0349, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 7.590810775756836, | |
| "rewards/margins": 35.01844787597656, | |
| "rewards/rejected": -27.427637100219727, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.3122923588039868, | |
| "grad_norm": 0.034927744418382645, | |
| "learning_rate": 4.215604094671835e-06, | |
| "logits/chosen": -0.2104714959859848, | |
| "logits/rejected": -0.26536011695861816, | |
| "logps/chosen": -1035.5135498046875, | |
| "logps/rejected": -1573.6461181640625, | |
| "loss": 0.0265, | |
| "rewards/accuracies": 0.9625000357627869, | |
| "rewards/chosen": 9.14853286743164, | |
| "rewards/margins": 35.578651428222656, | |
| "rewards/rejected": -26.43012046813965, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.3787375415282392, | |
| "grad_norm": 0.05187850445508957, | |
| "learning_rate": 4.172826515897146e-06, | |
| "logits/chosen": -0.19994229078292847, | |
| "logits/rejected": -0.26754456758499146, | |
| "logps/chosen": -997.74853515625, | |
| "logps/rejected": -1574.661865234375, | |
| "loss": 0.0233, | |
| "rewards/accuracies": 0.9671874642372131, | |
| "rewards/chosen": 8.397761344909668, | |
| "rewards/margins": 34.895263671875, | |
| "rewards/rejected": -26.49749755859375, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.4451827242524917, | |
| "grad_norm": 0.2731926739215851, | |
| "learning_rate": 4.129143072053639e-06, | |
| "logits/chosen": -0.19184930622577667, | |
| "logits/rejected": -0.26313984394073486, | |
| "logps/chosen": -1033.8189697265625, | |
| "logps/rejected": -1616.736572265625, | |
| "loss": 0.0222, | |
| "rewards/accuracies": 0.96875, | |
| "rewards/chosen": 8.199491500854492, | |
| "rewards/margins": 37.13923645019531, | |
| "rewards/rejected": -28.939746856689453, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.511627906976744, | |
| "grad_norm": 0.0523432120680809, | |
| "learning_rate": 4.084577418496775e-06, | |
| "logits/chosen": -0.20889827609062195, | |
| "logits/rejected": -0.2736192047595978, | |
| "logps/chosen": -1008.3775024414062, | |
| "logps/rejected": -1610.5330810546875, | |
| "loss": 0.0287, | |
| "rewards/accuracies": 0.9593750238418579, | |
| "rewards/chosen": 8.270615577697754, | |
| "rewards/margins": 36.194480895996094, | |
| "rewards/rejected": -27.923864364624023, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.5780730897009967, | |
| "grad_norm": 0.014995710924267769, | |
| "learning_rate": 4.039153688314146e-06, | |
| "logits/chosen": -0.20946355164051056, | |
| "logits/rejected": -0.26986002922058105, | |
| "logps/chosen": -999.2109375, | |
| "logps/rejected": -1588.408447265625, | |
| "loss": 0.0329, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 8.453977584838867, | |
| "rewards/margins": 36.9298095703125, | |
| "rewards/rejected": -28.475831985473633, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.644518272425249, | |
| "grad_norm": 0.02130577713251114, | |
| "learning_rate": 3.992896479256966e-06, | |
| "logits/chosen": -0.21112115681171417, | |
| "logits/rejected": -0.27220192551612854, | |
| "logps/chosen": -984.7587890625, | |
| "logps/rejected": -1543.271240234375, | |
| "loss": 0.0361, | |
| "rewards/accuracies": 0.9484374523162842, | |
| "rewards/chosen": 8.06696891784668, | |
| "rewards/margins": 37.377586364746094, | |
| "rewards/rejected": -29.31061553955078, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.7109634551495017, | |
| "grad_norm": 0.022987911477684975, | |
| "learning_rate": 3.945830840419966e-06, | |
| "logits/chosen": -0.18797051906585693, | |
| "logits/rejected": -0.2677266299724579, | |
| "logps/chosen": -1031.5941162109375, | |
| "logps/rejected": -1641.813720703125, | |
| "loss": 0.0263, | |
| "rewards/accuracies": 0.9625000357627869, | |
| "rewards/chosen": 8.485164642333984, | |
| "rewards/margins": 39.16246795654297, | |
| "rewards/rejected": -30.67730140686035, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.777408637873754, | |
| "grad_norm": 0.010663657449185848, | |
| "learning_rate": 3.897982258676867e-06, | |
| "logits/chosen": -0.19637425243854523, | |
| "logits/rejected": -0.2760464549064636, | |
| "logps/chosen": -961.6597900390625, | |
| "logps/rejected": -1573.847412109375, | |
| "loss": 0.0383, | |
| "rewards/accuracies": 0.9453125, | |
| "rewards/chosen": 8.17518424987793, | |
| "rewards/margins": 37.773780822753906, | |
| "rewards/rejected": -29.59859275817871, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.8438538205980066, | |
| "grad_norm": 0.05362631008028984, | |
| "learning_rate": 3.849376644878783e-06, | |
| "logits/chosen": -0.2064908742904663, | |
| "logits/rejected": -0.26836133003234863, | |
| "logps/chosen": -1019.5757446289062, | |
| "logps/rejected": -1615.7822265625, | |
| "loss": 0.028, | |
| "rewards/accuracies": 0.9609374403953552, | |
| "rewards/chosen": 7.7137041091918945, | |
| "rewards/margins": 40.03011703491211, | |
| "rewards/rejected": -32.31641387939453, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.910299003322259, | |
| "grad_norm": 0.0628666952252388, | |
| "learning_rate": 3.8000403198230385e-06, | |
| "logits/chosen": -0.20015954971313477, | |
| "logits/rejected": -0.2674901783466339, | |
| "logps/chosen": -986.7999267578125, | |
| "logps/rejected": -1530.0762939453125, | |
| "loss": 0.0223, | |
| "rewards/accuracies": 0.9703125357627869, | |
| "rewards/chosen": 8.094964981079102, | |
| "rewards/margins": 36.99173355102539, | |
| "rewards/rejected": -28.896766662597656, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.9767441860465116, | |
| "grad_norm": 0.27723661065101624, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "logits/chosen": -0.21333180367946625, | |
| "logits/rejected": -0.2772447466850281, | |
| "logps/chosen": -982.5689086914062, | |
| "logps/rejected": -1552.4580078125, | |
| "loss": 0.0357, | |
| "rewards/accuracies": 0.949999988079071, | |
| "rewards/chosen": 8.328544616699219, | |
| "rewards/margins": 38.28776550292969, | |
| "rewards/rejected": -29.959217071533203, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.039867109634551, | |
| "grad_norm": 0.08139760792255402, | |
| "learning_rate": 3.699282783125616e-06, | |
| "logits/chosen": -0.20954497158527374, | |
| "logits/rejected": -0.2789776921272278, | |
| "logps/chosen": -946.962646484375, | |
| "logps/rejected": -1561.8204345703125, | |
| "loss": 0.034, | |
| "rewards/accuracies": 0.9490132331848145, | |
| "rewards/chosen": 7.781158447265625, | |
| "rewards/margins": 38.58467102050781, | |
| "rewards/rejected": -30.80351448059082, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.106312292358804, | |
| "grad_norm": 0.005049778148531914, | |
| "learning_rate": 3.6479161334675294e-06, | |
| "logits/chosen": -0.18214014172554016, | |
| "logits/rejected": -0.27096524834632874, | |
| "logps/chosen": -959.0311279296875, | |
| "logps/rejected": -1593.2392578125, | |
| "loss": 0.0199, | |
| "rewards/accuracies": 0.971875011920929, | |
| "rewards/chosen": 8.438522338867188, | |
| "rewards/margins": 38.53790283203125, | |
| "rewards/rejected": -30.099374771118164, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.172757475083056, | |
| "grad_norm": 0.0067527154460549355, | |
| "learning_rate": 3.595927866972694e-06, | |
| "logits/chosen": -0.20282122492790222, | |
| "logits/rejected": -0.2759528160095215, | |
| "logps/chosen": -988.6021728515625, | |
| "logps/rejected": -1591.4195556640625, | |
| "loss": 0.0251, | |
| "rewards/accuracies": 0.964062511920929, | |
| "rewards/chosen": 8.654806137084961, | |
| "rewards/margins": 39.041419982910156, | |
| "rewards/rejected": -30.386611938476562, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.239202657807309, | |
| "grad_norm": 0.023877006024122238, | |
| "learning_rate": 3.543346136204545e-06, | |
| "logits/chosen": -0.20506156980991364, | |
| "logits/rejected": -0.27975520491600037, | |
| "logps/chosen": -1013.73681640625, | |
| "logps/rejected": -1679.869384765625, | |
| "loss": 0.0339, | |
| "rewards/accuracies": 0.9515625238418579, | |
| "rewards/chosen": 8.956099510192871, | |
| "rewards/margins": 41.54960250854492, | |
| "rewards/rejected": -32.593502044677734, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.305647840531561, | |
| "grad_norm": 0.16767418384552002, | |
| "learning_rate": 3.4901994150978926e-06, | |
| "logits/chosen": -0.2239498496055603, | |
| "logits/rejected": -0.27226123213768005, | |
| "logps/chosen": -1039.27685546875, | |
| "logps/rejected": -1601.485107421875, | |
| "loss": 0.0264, | |
| "rewards/accuracies": 0.9624999761581421, | |
| "rewards/chosen": 9.188974380493164, | |
| "rewards/margins": 39.43864440917969, | |
| "rewards/rejected": -30.249670028686523, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.372093023255814, | |
| "grad_norm": 0.023540904745459557, | |
| "learning_rate": 3.436516483539781e-06, | |
| "logits/chosen": -0.21794407069683075, | |
| "logits/rejected": -0.282469779253006, | |
| "logps/chosen": -966.4031372070312, | |
| "logps/rejected": -1527.18310546875, | |
| "loss": 0.0416, | |
| "rewards/accuracies": 0.940625011920929, | |
| "rewards/chosen": 7.827475070953369, | |
| "rewards/margins": 37.115562438964844, | |
| "rewards/rejected": -29.2880859375, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.438538205980066, | |
| "grad_norm": 0.0667799860239029, | |
| "learning_rate": 3.3823264117846722e-06, | |
| "logits/chosen": -0.200710266828537, | |
| "logits/rejected": -0.2715788185596466, | |
| "logps/chosen": -1030.8763427734375, | |
| "logps/rejected": -1679.390869140625, | |
| "loss": 0.0243, | |
| "rewards/accuracies": 0.965624988079071, | |
| "rewards/chosen": 7.525571823120117, | |
| "rewards/margins": 43.3552360534668, | |
| "rewards/rejected": -35.82966232299805, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.504983388704319, | |
| "grad_norm": 0.01825464330613613, | |
| "learning_rate": 3.3276585447123957e-06, | |
| "logits/chosen": -0.19761598110198975, | |
| "logits/rejected": -0.2707269787788391, | |
| "logps/chosen": -1000.3179931640625, | |
| "logps/rejected": -1630.790283203125, | |
| "loss": 0.0233, | |
| "rewards/accuracies": 0.9671875238418579, | |
| "rewards/chosen": 8.050663948059082, | |
| "rewards/margins": 41.542236328125, | |
| "rewards/rejected": -33.49156951904297, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.010783619247376919, | |
| "learning_rate": 3.272542485937369e-06, | |
| "logits/chosen": -0.21159793436527252, | |
| "logits/rejected": -0.2814417779445648, | |
| "logps/chosen": -988.6871337890625, | |
| "logps/rejected": -1556.5059814453125, | |
| "loss": 0.0295, | |
| "rewards/accuracies": 0.9578125476837158, | |
| "rewards/chosen": 8.39299201965332, | |
| "rewards/margins": 38.407005310058594, | |
| "rewards/rejected": -30.01401710510254, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.637873754152824, | |
| "grad_norm": 0.040845226496458054, | |
| "learning_rate": 3.217008081777726e-06, | |
| "logits/chosen": -0.2169944941997528, | |
| "logits/rejected": -0.28610751032829285, | |
| "logps/chosen": -930.8607788085938, | |
| "logps/rejected": -1484.1146240234375, | |
| "loss": 0.0346, | |
| "rewards/accuracies": 0.9515625238418579, | |
| "rewards/chosen": 9.505171775817871, | |
| "rewards/margins": 35.94363784790039, | |
| "rewards/rejected": -26.43846321105957, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.704318936877076, | |
| "grad_norm": 0.005064092110842466, | |
| "learning_rate": 3.1610854050930063e-06, | |
| "logits/chosen": -0.1968638151884079, | |
| "logits/rejected": -0.274067759513855, | |
| "logps/chosen": -988.1194458007812, | |
| "logps/rejected": -1650.0118408203125, | |
| "loss": 0.0294, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 9.581026077270508, | |
| "rewards/margins": 39.75960159301758, | |
| "rewards/rejected": -30.17857551574707, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.770764119601329, | |
| "grad_norm": 0.0075860158540308475, | |
| "learning_rate": 3.1048047389991693e-06, | |
| "logits/chosen": -0.20798590779304504, | |
| "logits/rejected": -0.2726728320121765, | |
| "logps/chosen": -1006.69873046875, | |
| "logps/rejected": -1619.1817626953125, | |
| "loss": 0.0327, | |
| "rewards/accuracies": 0.9546875357627869, | |
| "rewards/chosen": 8.933056831359863, | |
| "rewards/margins": 39.96248245239258, | |
| "rewards/rejected": -31.0294246673584, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.837209302325581, | |
| "grad_norm": 0.018649380654096603, | |
| "learning_rate": 3.0481965604697582e-06, | |
| "logits/chosen": -0.19381622970104218, | |
| "logits/rejected": -0.27504754066467285, | |
| "logps/chosen": -964.4519653320312, | |
| "logps/rejected": -1575.1983642578125, | |
| "loss": 0.024, | |
| "rewards/accuracies": 0.9656250476837158, | |
| "rewards/chosen": 8.419488906860352, | |
| "rewards/margins": 39.957523345947266, | |
| "rewards/rejected": -31.53803825378418, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.903654485049834, | |
| "grad_norm": 0.009096500463783741, | |
| "learning_rate": 2.9912915238320755e-06, | |
| "logits/chosen": -0.20088475942611694, | |
| "logits/rejected": -0.2820194959640503, | |
| "logps/chosen": -966.4456787109375, | |
| "logps/rejected": -1618.930908203125, | |
| "loss": 0.0306, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 9.212898254394531, | |
| "rewards/margins": 41.09564208984375, | |
| "rewards/rejected": -31.882747650146484, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.970099667774086, | |
| "grad_norm": 0.01880677044391632, | |
| "learning_rate": 2.9341204441673267e-06, | |
| "logits/chosen": -0.2191881388425827, | |
| "logits/rejected": -0.2813561260700226, | |
| "logps/chosen": -1037.1162109375, | |
| "logps/rejected": -1705.19873046875, | |
| "loss": 0.0305, | |
| "rewards/accuracies": 0.9562500715255737, | |
| "rewards/chosen": 8.659435272216797, | |
| "rewards/margins": 43.9373664855957, | |
| "rewards/rejected": -35.27792739868164, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.033222591362127, | |
| "grad_norm": 0.015276752412319183, | |
| "learning_rate": 2.876714280623708e-06, | |
| "logits/chosen": -0.1968104988336563, | |
| "logits/rejected": -0.27359306812286377, | |
| "logps/chosen": -971.7843017578125, | |
| "logps/rejected": -1637.6700439453125, | |
| "loss": 0.023, | |
| "rewards/accuracies": 0.9654605388641357, | |
| "rewards/chosen": 8.207253456115723, | |
| "rewards/margins": 42.495601654052734, | |
| "rewards/rejected": -34.28834533691406, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.099667774086379, | |
| "grad_norm": 0.005275533068925142, | |
| "learning_rate": 2.8191041196514874e-06, | |
| "logits/chosen": -0.198197603225708, | |
| "logits/rejected": -0.28718599677085876, | |
| "logps/chosen": -1012.2830200195312, | |
| "logps/rejected": -1695.5068359375, | |
| "loss": 0.0359, | |
| "rewards/accuracies": 0.9484374523162842, | |
| "rewards/chosen": 8.050834655761719, | |
| "rewards/margins": 43.739540100097656, | |
| "rewards/rejected": -35.688697814941406, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.166112956810632, | |
| "grad_norm": 0.03785371407866478, | |
| "learning_rate": 2.761321158169134e-06, | |
| "logits/chosen": -0.21472084522247314, | |
| "logits/rejected": -0.2798621654510498, | |
| "logps/chosen": -1062.145263671875, | |
| "logps/rejected": -1711.7852783203125, | |
| "loss": 0.0338, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 8.760993003845215, | |
| "rewards/margins": 44.73903274536133, | |
| "rewards/rejected": -35.9780387878418, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.232558139534884, | |
| "grad_norm": 0.05422116443514824, | |
| "learning_rate": 2.703396686669646e-06, | |
| "logits/chosen": -0.2090332806110382, | |
| "logits/rejected": -0.2846227288246155, | |
| "logps/chosen": -984.9669189453125, | |
| "logps/rejected": -1620.4434814453125, | |
| "loss": 0.0338, | |
| "rewards/accuracies": 0.9515625238418579, | |
| "rewards/chosen": 8.526132583618164, | |
| "rewards/margins": 41.130706787109375, | |
| "rewards/rejected": -32.60457229614258, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.299003322259137, | |
| "grad_norm": 0.021388115361332893, | |
| "learning_rate": 2.6453620722761897e-06, | |
| "logits/chosen": -0.22062420845031738, | |
| "logits/rejected": -0.2795643210411072, | |
| "logps/chosen": -1018.5145263671875, | |
| "logps/rejected": -1654.4093017578125, | |
| "loss": 0.0305, | |
| "rewards/accuracies": 0.9562499523162842, | |
| "rewards/chosen": 9.038843154907227, | |
| "rewards/margins": 43.468109130859375, | |
| "rewards/rejected": -34.42927169799805, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.365448504983389, | |
| "grad_norm": 0.02929343283176422, | |
| "learning_rate": 2.587248741756253e-06, | |
| "logits/chosen": -0.19668123126029968, | |
| "logits/rejected": -0.2809968888759613, | |
| "logps/chosen": -1002.3245239257812, | |
| "logps/rejected": -1654.1307373046875, | |
| "loss": 0.0203, | |
| "rewards/accuracies": 0.9718750715255737, | |
| "rewards/chosen": 7.798498630523682, | |
| "rewards/margins": 43.116241455078125, | |
| "rewards/rejected": -35.3177490234375, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.431893687707642, | |
| "grad_norm": 0.007884162478148937, | |
| "learning_rate": 2.5290881645034932e-06, | |
| "logits/chosen": -0.2037358433008194, | |
| "logits/rejected": -0.2698831558227539, | |
| "logps/chosen": -986.9371337890625, | |
| "logps/rejected": -1607.3692626953125, | |
| "loss": 0.0197, | |
| "rewards/accuracies": 0.9734375476837158, | |
| "rewards/chosen": 7.649607181549072, | |
| "rewards/margins": 42.175437927246094, | |
| "rewards/rejected": -34.52583312988281, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.498338870431894, | |
| "grad_norm": 0.01341085135936737, | |
| "learning_rate": 2.470911835496508e-06, | |
| "logits/chosen": -0.19817571341991425, | |
| "logits/rejected": -0.27480000257492065, | |
| "logps/chosen": -1008.7352905273438, | |
| "logps/rejected": -1691.955322265625, | |
| "loss": 0.024, | |
| "rewards/accuracies": 0.9656250476837158, | |
| "rewards/chosen": 7.666060447692871, | |
| "rewards/margins": 44.92629623413086, | |
| "rewards/rejected": -37.26023864746094, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.5647840531561465, | |
| "grad_norm": 0.19932714104652405, | |
| "learning_rate": 2.4127512582437486e-06, | |
| "logits/chosen": -0.21822643280029297, | |
| "logits/rejected": -0.2849328815937042, | |
| "logps/chosen": -952.5464477539062, | |
| "logps/rejected": -1569.07470703125, | |
| "loss": 0.0351, | |
| "rewards/accuracies": 0.9500000476837158, | |
| "rewards/chosen": 8.396932601928711, | |
| "rewards/margins": 40.544769287109375, | |
| "rewards/rejected": -32.14783477783203, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.631229235880399, | |
| "grad_norm": 0.024993257597088814, | |
| "learning_rate": 2.3546379277238107e-06, | |
| "logits/chosen": -0.2073173075914383, | |
| "logits/rejected": -0.2831364870071411, | |
| "logps/chosen": -982.529052734375, | |
| "logps/rejected": -1603.887939453125, | |
| "loss": 0.023, | |
| "rewards/accuracies": 0.9671875238418579, | |
| "rewards/chosen": 9.977124214172363, | |
| "rewards/margins": 41.838199615478516, | |
| "rewards/rejected": -31.8610782623291, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.6976744186046515, | |
| "grad_norm": 0.025327615439891815, | |
| "learning_rate": 2.296603313330355e-06, | |
| "logits/chosen": -0.22604137659072876, | |
| "logits/rejected": -0.2886336147785187, | |
| "logps/chosen": -949.3585205078125, | |
| "logps/rejected": -1545.9788818359375, | |
| "loss": 0.0393, | |
| "rewards/accuracies": 0.9437499642372131, | |
| "rewards/chosen": 9.380352973937988, | |
| "rewards/margins": 39.31431198120117, | |
| "rewards/rejected": -29.9339599609375, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.764119601328904, | |
| "grad_norm": 0.09348896890878677, | |
| "learning_rate": 2.238678841830867e-06, | |
| "logits/chosen": -0.21441370248794556, | |
| "logits/rejected": -0.28145626187324524, | |
| "logps/chosen": -969.75927734375, | |
| "logps/rejected": -1589.4075927734375, | |
| "loss": 0.0326, | |
| "rewards/accuracies": 0.9531249403953552, | |
| "rewards/chosen": 8.257079124450684, | |
| "rewards/margins": 40.917396545410156, | |
| "rewards/rejected": -32.660316467285156, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.8305647840531565, | |
| "grad_norm": 0.04749366641044617, | |
| "learning_rate": 2.1808958803485134e-06, | |
| "logits/chosen": -0.20293860137462616, | |
| "logits/rejected": -0.2881726026535034, | |
| "logps/chosen": -956.5830078125, | |
| "logps/rejected": -1648.4208984375, | |
| "loss": 0.0294, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 8.458630561828613, | |
| "rewards/margins": 43.482696533203125, | |
| "rewards/rejected": -35.02406311035156, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 5.897009966777409, | |
| "grad_norm": 0.016179274767637253, | |
| "learning_rate": 2.1232857193762923e-06, | |
| "logits/chosen": -0.2071429193019867, | |
| "logits/rejected": -0.2786925435066223, | |
| "logps/chosen": -974.3377685546875, | |
| "logps/rejected": -1581.3140869140625, | |
| "loss": 0.0284, | |
| "rewards/accuracies": 0.9593750238418579, | |
| "rewards/chosen": 8.533307075500488, | |
| "rewards/margins": 41.709964752197266, | |
| "rewards/rejected": -33.176658630371094, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 5.9634551495016614, | |
| "grad_norm": 0.0812649205327034, | |
| "learning_rate": 2.0658795558326745e-06, | |
| "logits/chosen": -0.21706193685531616, | |
| "logits/rejected": -0.2818005681037903, | |
| "logps/chosen": -990.1275634765625, | |
| "logps/rejected": -1619.1646728515625, | |
| "loss": 0.0316, | |
| "rewards/accuracies": 0.9546875357627869, | |
| "rewards/chosen": 8.88737678527832, | |
| "rewards/margins": 41.76374053955078, | |
| "rewards/rejected": -32.876365661621094, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.026578073089701, | |
| "grad_norm": 0.009111125022172928, | |
| "learning_rate": 2.0087084761679245e-06, | |
| "logits/chosen": -0.2077750265598297, | |
| "logits/rejected": -0.2753692865371704, | |
| "logps/chosen": -1017.2333984375, | |
| "logps/rejected": -1682.0185546875, | |
| "loss": 0.0176, | |
| "rewards/accuracies": 0.9736842513084412, | |
| "rewards/chosen": 8.799712181091309, | |
| "rewards/margins": 43.73455810546875, | |
| "rewards/rejected": -34.934844970703125, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.093023255813954, | |
| "grad_norm": 0.009804153814911842, | |
| "learning_rate": 1.9518034395302413e-06, | |
| "logits/chosen": -0.19864043593406677, | |
| "logits/rejected": -0.2817736566066742, | |
| "logps/chosen": -986.0798950195312, | |
| "logps/rejected": -1684.548095703125, | |
| "loss": 0.0274, | |
| "rewards/accuracies": 0.9609375596046448, | |
| "rewards/chosen": 9.566544532775879, | |
| "rewards/margins": 43.709625244140625, | |
| "rewards/rejected": -34.14308547973633, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.159468438538206, | |
| "grad_norm": 0.0946379005908966, | |
| "learning_rate": 1.895195261000831e-06, | |
| "logits/chosen": -0.20698292553424835, | |
| "logits/rejected": -0.2797500193119049, | |
| "logps/chosen": -999.882568359375, | |
| "logps/rejected": -1666.398193359375, | |
| "loss": 0.0261, | |
| "rewards/accuracies": 0.9625000357627869, | |
| "rewards/chosen": 9.238912582397461, | |
| "rewards/margins": 43.197425842285156, | |
| "rewards/rejected": -33.9585075378418, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.225913621262459, | |
| "grad_norm": 0.03615960851311684, | |
| "learning_rate": 1.8389145949069953e-06, | |
| "logits/chosen": -0.20727947354316711, | |
| "logits/rejected": -0.2856772541999817, | |
| "logps/chosen": -1012.0263061523438, | |
| "logps/rejected": -1656.6219482421875, | |
| "loss": 0.0299, | |
| "rewards/accuracies": 0.9578125476837158, | |
| "rewards/chosen": 9.405717849731445, | |
| "rewards/margins": 43.6558952331543, | |
| "rewards/rejected": -34.25017547607422, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.292358803986711, | |
| "grad_norm": 0.09421185404062271, | |
| "learning_rate": 1.7829919182222752e-06, | |
| "logits/chosen": -0.21414652466773987, | |
| "logits/rejected": -0.2791929244995117, | |
| "logps/chosen": -991.490966796875, | |
| "logps/rejected": -1588.97412109375, | |
| "loss": 0.0284, | |
| "rewards/accuracies": 0.9593750238418579, | |
| "rewards/chosen": 8.861186981201172, | |
| "rewards/margins": 42.22901153564453, | |
| "rewards/rejected": -33.367828369140625, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.358803986710964, | |
| "grad_norm": 0.1068965345621109, | |
| "learning_rate": 1.7274575140626318e-06, | |
| "logits/chosen": -0.20373263955116272, | |
| "logits/rejected": -0.2882729768753052, | |
| "logps/chosen": -975.765625, | |
| "logps/rejected": -1631.05810546875, | |
| "loss": 0.0327, | |
| "rewards/accuracies": 0.9546874761581421, | |
| "rewards/chosen": 9.140608787536621, | |
| "rewards/margins": 42.56184387207031, | |
| "rewards/rejected": -33.421234130859375, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.425249169435216, | |
| "grad_norm": 0.20864491164684296, | |
| "learning_rate": 1.6723414552876052e-06, | |
| "logits/chosen": -0.2219492346048355, | |
| "logits/rejected": -0.2909078299999237, | |
| "logps/chosen": -998.4899291992188, | |
| "logps/rejected": -1629.822998046875, | |
| "loss": 0.0316, | |
| "rewards/accuracies": 0.9546874761581421, | |
| "rewards/chosen": 10.093676567077637, | |
| "rewards/margins": 41.07781982421875, | |
| "rewards/rejected": -30.98414421081543, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.4916943521594686, | |
| "grad_norm": 0.0074838376604020596, | |
| "learning_rate": 1.6176735882153284e-06, | |
| "logits/chosen": -0.21261698007583618, | |
| "logits/rejected": -0.2835195064544678, | |
| "logps/chosen": -982.4485473632812, | |
| "logps/rejected": -1603.0489501953125, | |
| "loss": 0.0274, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": 9.805879592895508, | |
| "rewards/margins": 41.88581466674805, | |
| "rewards/rejected": -32.079933166503906, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.558139534883721, | |
| "grad_norm": 0.009234702214598656, | |
| "learning_rate": 1.56348351646022e-06, | |
| "logits/chosen": -0.21804842352867126, | |
| "logits/rejected": -0.2872644066810608, | |
| "logps/chosen": -992.9108276367188, | |
| "logps/rejected": -1682.7227783203125, | |
| "loss": 0.0326, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": 9.05055046081543, | |
| "rewards/margins": 44.20970916748047, | |
| "rewards/rejected": -35.15916442871094, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.6245847176079735, | |
| "grad_norm": 0.005751634947955608, | |
| "learning_rate": 1.509800584902108e-06, | |
| "logits/chosen": -0.21074561774730682, | |
| "logits/rejected": -0.28109174966812134, | |
| "logps/chosen": -962.9151611328125, | |
| "logps/rejected": -1613.1900634765625, | |
| "loss": 0.025, | |
| "rewards/accuracies": 0.9640624523162842, | |
| "rewards/chosen": 8.966195106506348, | |
| "rewards/margins": 42.510337829589844, | |
| "rewards/rejected": -33.54414367675781, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.691029900332226, | |
| "grad_norm": 0.004595512058585882, | |
| "learning_rate": 1.4566538637954556e-06, | |
| "logits/chosen": -0.206181600689888, | |
| "logits/rejected": -0.28721800446510315, | |
| "logps/chosen": -916.551025390625, | |
| "logps/rejected": -1526.8843994140625, | |
| "loss": 0.023, | |
| "rewards/accuracies": 0.9671874642372131, | |
| "rewards/chosen": 9.025623321533203, | |
| "rewards/margins": 39.80775833129883, | |
| "rewards/rejected": -30.782133102416992, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 6.7574750830564785, | |
| "grad_norm": 0.003542843274772167, | |
| "learning_rate": 1.4040721330273063e-06, | |
| "logits/chosen": -0.2257394641637802, | |
| "logits/rejected": -0.2908530533313751, | |
| "logps/chosen": -960.90234375, | |
| "logps/rejected": -1578.2750244140625, | |
| "loss": 0.0434, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 8.649526596069336, | |
| "rewards/margins": 42.02993392944336, | |
| "rewards/rejected": -33.380401611328125, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 6.823920265780731, | |
| "grad_norm": 0.06329812854528427, | |
| "learning_rate": 1.3520838665324704e-06, | |
| "logits/chosen": -0.2071303427219391, | |
| "logits/rejected": -0.2824835181236267, | |
| "logps/chosen": -973.2369995117188, | |
| "logps/rejected": -1608.9039306640625, | |
| "loss": 0.0295, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 8.745044708251953, | |
| "rewards/margins": 42.52543258666992, | |
| "rewards/rejected": -33.78038787841797, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 6.8903654485049834, | |
| "grad_norm": 0.014581708237528801, | |
| "learning_rate": 1.3007172168743854e-06, | |
| "logits/chosen": -0.21656513214111328, | |
| "logits/rejected": -0.2857995927333832, | |
| "logps/chosen": -993.361328125, | |
| "logps/rejected": -1631.5338134765625, | |
| "loss": 0.0273, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": 9.786979675292969, | |
| "rewards/margins": 42.11618423461914, | |
| "rewards/rejected": -32.32920455932617, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 6.956810631229236, | |
| "grad_norm": 0.042626578360795975, | |
| "learning_rate": 1.2500000000000007e-06, | |
| "logits/chosen": -0.20040905475616455, | |
| "logits/rejected": -0.2804938852787018, | |
| "logps/chosen": -1012.1871337890625, | |
| "logps/rejected": -1693.8763427734375, | |
| "loss": 0.024, | |
| "rewards/accuracies": 0.965624988079071, | |
| "rewards/chosen": 9.336542129516602, | |
| "rewards/margins": 44.35811996459961, | |
| "rewards/rejected": -35.02157211303711, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.019933554817276, | |
| "grad_norm": 0.08022702485322952, | |
| "learning_rate": 1.1999596801769617e-06, | |
| "logits/chosen": -0.237129807472229, | |
| "logits/rejected": -0.28911346197128296, | |
| "logps/chosen": -1007.3953857421875, | |
| "logps/rejected": -1635.5213623046875, | |
| "loss": 0.0337, | |
| "rewards/accuracies": 0.9490132331848145, | |
| "rewards/chosen": 9.52566909790039, | |
| "rewards/margins": 42.855995178222656, | |
| "rewards/rejected": -33.33032989501953, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.086378737541528, | |
| "grad_norm": 0.022050941362977028, | |
| "learning_rate": 1.1506233551212186e-06, | |
| "logits/chosen": -0.21040520071983337, | |
| "logits/rejected": -0.28766316175460815, | |
| "logps/chosen": -1012.5867919921875, | |
| "logps/rejected": -1669.2762451171875, | |
| "loss": 0.024, | |
| "rewards/accuracies": 0.965624988079071, | |
| "rewards/chosen": 9.755768775939941, | |
| "rewards/margins": 43.82841110229492, | |
| "rewards/rejected": -34.07263946533203, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.152823920265781, | |
| "grad_norm": 0.006165655329823494, | |
| "learning_rate": 1.1020177413231334e-06, | |
| "logits/chosen": -0.20854339003562927, | |
| "logits/rejected": -0.28692665696144104, | |
| "logps/chosen": -970.487060546875, | |
| "logps/rejected": -1616.68212890625, | |
| "loss": 0.0304, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 9.218266487121582, | |
| "rewards/margins": 42.4344367980957, | |
| "rewards/rejected": -33.21616744995117, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.219269102990033, | |
| "grad_norm": 0.02192852459847927, | |
| "learning_rate": 1.0541691595800338e-06, | |
| "logits/chosen": -0.20376423001289368, | |
| "logits/rejected": -0.2804113030433655, | |
| "logps/chosen": -1000.4962158203125, | |
| "logps/rejected": -1672.725341796875, | |
| "loss": 0.0349, | |
| "rewards/accuracies": 0.949999988079071, | |
| "rewards/chosen": 8.690274238586426, | |
| "rewards/margins": 44.81369400024414, | |
| "rewards/rejected": -36.1234130859375, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.285714285714286, | |
| "grad_norm": 0.07914319634437561, | |
| "learning_rate": 1.0071035207430352e-06, | |
| "logits/chosen": -0.19936969876289368, | |
| "logits/rejected": -0.28211313486099243, | |
| "logps/chosen": -978.279052734375, | |
| "logps/rejected": -1641.731201171875, | |
| "loss": 0.0262, | |
| "rewards/accuracies": 0.9624999761581421, | |
| "rewards/chosen": 8.959029197692871, | |
| "rewards/margins": 43.952213287353516, | |
| "rewards/rejected": -34.993186950683594, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.352159468438538, | |
| "grad_norm": 0.014706037007272243, | |
| "learning_rate": 9.608463116858544e-07, | |
| "logits/chosen": -0.21040727198123932, | |
| "logits/rejected": -0.2795957922935486, | |
| "logps/chosen": -1030.148681640625, | |
| "logps/rejected": -1709.89892578125, | |
| "loss": 0.0217, | |
| "rewards/accuracies": 0.96875, | |
| "rewards/chosen": 9.775686264038086, | |
| "rewards/margins": 45.375675201416016, | |
| "rewards/rejected": -35.59999084472656, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.4186046511627906, | |
| "grad_norm": 0.006406854372471571, | |
| "learning_rate": 9.154225815032242e-07, | |
| "logits/chosen": -0.20644372701644897, | |
| "logits/rejected": -0.2845055162906647, | |
| "logps/chosen": -941.191162109375, | |
| "logps/rejected": -1556.8076171875, | |
| "loss": 0.0284, | |
| "rewards/accuracies": 0.9593749642372131, | |
| "rewards/chosen": 9.463266372680664, | |
| "rewards/margins": 41.32258224487305, | |
| "rewards/rejected": -31.859315872192383, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.485049833887043, | |
| "grad_norm": 0.023300213739275932, | |
| "learning_rate": 8.708569279463622e-07, | |
| "logits/chosen": -0.2077961415052414, | |
| "logits/rejected": -0.27984070777893066, | |
| "logps/chosen": -962.038330078125, | |
| "logps/rejected": -1615.148193359375, | |
| "loss": 0.0273, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": 9.014985084533691, | |
| "rewards/margins": 42.5955924987793, | |
| "rewards/rejected": -33.580604553222656, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.5514950166112955, | |
| "grad_norm": 0.0013064617523923516, | |
| "learning_rate": 8.271734841028553e-07, | |
| "logits/chosen": -0.21521812677383423, | |
| "logits/rejected": -0.29194578528404236, | |
| "logps/chosen": -969.0303955078125, | |
| "logps/rejected": -1659.9932861328125, | |
| "loss": 0.0305, | |
| "rewards/accuracies": 0.9562500715255737, | |
| "rewards/chosen": 9.51773452758789, | |
| "rewards/margins": 43.94035720825195, | |
| "rewards/rejected": -34.42262268066406, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.617940199335548, | |
| "grad_norm": 0.027837639674544334, | |
| "learning_rate": 7.843959053281663e-07, | |
| "logits/chosen": -0.2188502699136734, | |
| "logits/rejected": -0.28792399168014526, | |
| "logps/chosen": -983.6058959960938, | |
| "logps/rejected": -1592.7557373046875, | |
| "loss": 0.024, | |
| "rewards/accuracies": 0.965624988079071, | |
| "rewards/chosen": 9.258206367492676, | |
| "rewards/margins": 42.35002136230469, | |
| "rewards/rejected": -33.09181213378906, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.6843853820598005, | |
| "grad_norm": 0.021290810778737068, | |
| "learning_rate": 7.425473564358457e-07, | |
| "logits/chosen": -0.21639959514141083, | |
| "logits/rejected": -0.2931935489177704, | |
| "logps/chosen": -947.8458251953125, | |
| "logps/rejected": -1650.803466796875, | |
| "loss": 0.0348, | |
| "rewards/accuracies": 0.9499999284744263, | |
| "rewards/chosen": 9.148832321166992, | |
| "rewards/margins": 43.69053268432617, | |
| "rewards/rejected": -34.54170227050781, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 7.750830564784053, | |
| "grad_norm": 0.0003092914412263781, | |
| "learning_rate": 7.016504991533727e-07, | |
| "logits/chosen": -0.21218986809253693, | |
| "logits/rejected": -0.2906990051269531, | |
| "logps/chosen": -1019.6378173828125, | |
| "logps/rejected": -1668.9599609375, | |
| "loss": 0.0273, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": 9.230569839477539, | |
| "rewards/margins": 43.63245391845703, | |
| "rewards/rejected": -34.40188217163086, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 7.8172757475083055, | |
| "grad_norm": 0.06043156608939171, | |
| "learning_rate": 6.617274798504286e-07, | |
| "logits/chosen": -0.22491909563541412, | |
| "logits/rejected": -0.287073016166687, | |
| "logps/chosen": -969.3558349609375, | |
| "logps/rejected": -1639.527099609375, | |
| "loss": 0.038, | |
| "rewards/accuracies": 0.9468749761581421, | |
| "rewards/chosen": 9.15910530090332, | |
| "rewards/margins": 43.614559173583984, | |
| "rewards/rejected": -34.45545196533203, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 7.883720930232558, | |
| "grad_norm": 0.07719282060861588, | |
| "learning_rate": 6.227999175462521e-07, | |
| "logits/chosen": -0.21083196997642517, | |
| "logits/rejected": -0.2891797125339508, | |
| "logps/chosen": -949.5618286132812, | |
| "logps/rejected": -1556.614990234375, | |
| "loss": 0.03, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 9.331879615783691, | |
| "rewards/margins": 40.77275466918945, | |
| "rewards/rejected": -31.440876007080078, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 7.95016611295681, | |
| "grad_norm": 0.0071638901717960835, | |
| "learning_rate": 5.848888922025553e-07, | |
| "logits/chosen": -0.21286137402057648, | |
| "logits/rejected": -0.2848753035068512, | |
| "logps/chosen": -994.6892700195312, | |
| "logps/rejected": -1621.3260498046875, | |
| "loss": 0.0282, | |
| "rewards/accuracies": 0.9593750238418579, | |
| "rewards/chosen": 9.580074310302734, | |
| "rewards/margins": 42.589351654052734, | |
| "rewards/rejected": -33.00927734375, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.013289036544851, | |
| "grad_norm": 0.09425640851259232, | |
| "learning_rate": 5.48014933308352e-07, | |
| "logits/chosen": -0.23732267320156097, | |
| "logits/rejected": -0.28549158573150635, | |
| "logps/chosen": -977.196044921875, | |
| "logps/rejected": -1548.5743408203125, | |
| "loss": 0.0337, | |
| "rewards/accuracies": 0.9490132331848145, | |
| "rewards/chosen": 9.362371444702148, | |
| "rewards/margins": 41.201656341552734, | |
| "rewards/rejected": -31.839282989501953, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.079734219269103, | |
| "grad_norm": 0.02250111848115921, | |
| "learning_rate": 5.121980087628802e-07, | |
| "logits/chosen": -0.21824708580970764, | |
| "logits/rejected": -0.2829313278198242, | |
| "logps/chosen": -1000.27001953125, | |
| "logps/rejected": -1619.780029296875, | |
| "loss": 0.0304, | |
| "rewards/accuracies": 0.9562499523162842, | |
| "rewards/chosen": 9.380255699157715, | |
| "rewards/margins": 43.14335632324219, | |
| "rewards/rejected": -33.763099670410156, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.146179401993356, | |
| "grad_norm": 0.0924224779009819, | |
| "learning_rate": 4.774575140626317e-07, | |
| "logits/chosen": -0.19590233266353607, | |
| "logits/rejected": -0.279987633228302, | |
| "logps/chosen": -965.008056640625, | |
| "logps/rejected": -1618.52197265625, | |
| "loss": 0.0229, | |
| "rewards/accuracies": 0.9671875238418579, | |
| "rewards/chosen": 9.053982734680176, | |
| "rewards/margins": 42.93439865112305, | |
| "rewards/rejected": -33.88041687011719, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.212624584717608, | |
| "grad_norm": 0.030095938593149185, | |
| "learning_rate": 4.438122617983442e-07, | |
| "logits/chosen": -0.21942079067230225, | |
| "logits/rejected": -0.2864038348197937, | |
| "logps/chosen": -996.6010131835938, | |
| "logps/rejected": -1615.1669921875, | |
| "loss": 0.0283, | |
| "rewards/accuracies": 0.9593749642372131, | |
| "rewards/chosen": 9.895916938781738, | |
| "rewards/margins": 42.90495681762695, | |
| "rewards/rejected": -33.00904083251953, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.279069767441861, | |
| "grad_norm": 0.00967055931687355, | |
| "learning_rate": 4.1128047146765936e-07, | |
| "logits/chosen": -0.21973438560962677, | |
| "logits/rejected": -0.2826644480228424, | |
| "logps/chosen": -1001.6805419921875, | |
| "logps/rejected": -1661.8216552734375, | |
| "loss": 0.0263, | |
| "rewards/accuracies": 0.9624999761581421, | |
| "rewards/chosen": 9.920458793640137, | |
| "rewards/margins": 44.114959716796875, | |
| "rewards/rejected": -34.19449996948242, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.345514950166113, | |
| "grad_norm": 0.06534740328788757, | |
| "learning_rate": 3.798797596089351e-07, | |
| "logits/chosen": -0.21031054854393005, | |
| "logits/rejected": -0.29491880536079407, | |
| "logps/chosen": -985.116455078125, | |
| "logps/rejected": -1668.977783203125, | |
| "loss": 0.0294, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 9.297616958618164, | |
| "rewards/margins": 44.02253723144531, | |
| "rewards/rejected": -34.72492218017578, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.411960132890366, | |
| "grad_norm": 0.018059343099594116, | |
| "learning_rate": 3.4962713026158697e-07, | |
| "logits/chosen": -0.20840252935886383, | |
| "logits/rejected": -0.28752243518829346, | |
| "logps/chosen": -1003.153564453125, | |
| "logps/rejected": -1656.9368896484375, | |
| "loss": 0.0305, | |
| "rewards/accuracies": 0.956250011920929, | |
| "rewards/chosen": 9.164114952087402, | |
| "rewards/margins": 43.66472625732422, | |
| "rewards/rejected": -34.5006103515625, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.478405315614618, | |
| "grad_norm": 0.06759638339281082, | |
| "learning_rate": 3.2053896575809426e-07, | |
| "logits/chosen": -0.21498829126358032, | |
| "logits/rejected": -0.2865886390209198, | |
| "logps/chosen": -997.242919921875, | |
| "logps/rejected": -1629.5584716796875, | |
| "loss": 0.0336, | |
| "rewards/accuracies": 0.9515625238418579, | |
| "rewards/chosen": 9.193347930908203, | |
| "rewards/margins": 42.87928771972656, | |
| "rewards/rejected": -33.68593978881836, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.544850498338871, | |
| "grad_norm": 0.02950510010123253, | |
| "learning_rate": 2.9263101785268253e-07, | |
| "logits/chosen": -0.20270058512687683, | |
| "logits/rejected": -0.2855191230773926, | |
| "logps/chosen": -989.1177368164062, | |
| "logps/rejected": -1705.371337890625, | |
| "loss": 0.0317, | |
| "rewards/accuracies": 0.9546875357627869, | |
| "rewards/chosen": 9.245896339416504, | |
| "rewards/margins": 46.25841522216797, | |
| "rewards/rejected": -37.01251983642578, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.611295681063122, | |
| "grad_norm": 0.023194456472992897, | |
| "learning_rate": 2.6591839919146963e-07, | |
| "logits/chosen": -0.22827497124671936, | |
| "logits/rejected": -0.29209384322166443, | |
| "logps/chosen": -974.6512451171875, | |
| "logps/rejected": -1619.2587890625, | |
| "loss": 0.0337, | |
| "rewards/accuracies": 0.9515624642372131, | |
| "rewards/chosen": 9.299782752990723, | |
| "rewards/margins": 42.95219039916992, | |
| "rewards/rejected": -33.65241241455078, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.677740863787376, | |
| "grad_norm": 0.052743032574653625, | |
| "learning_rate": 2.404155751286988e-07, | |
| "logits/chosen": -0.1986398845911026, | |
| "logits/rejected": -0.28585973381996155, | |
| "logps/chosen": -958.5393676757812, | |
| "logps/rejected": -1646.5435791015625, | |
| "loss": 0.0217, | |
| "rewards/accuracies": 0.9687501192092896, | |
| "rewards/chosen": 9.183121681213379, | |
| "rewards/margins": 43.7320671081543, | |
| "rewards/rejected": -34.5489387512207, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 8.744186046511627, | |
| "grad_norm": 0.0289428923279047, | |
| "learning_rate": 2.1613635589349756e-07, | |
| "logits/chosen": -0.21537990868091583, | |
| "logits/rejected": -0.28249984979629517, | |
| "logps/chosen": -1000.1814575195312, | |
| "logps/rejected": -1656.298095703125, | |
| "loss": 0.0255, | |
| "rewards/accuracies": 0.964062511920929, | |
| "rewards/chosen": 9.053632736206055, | |
| "rewards/margins": 44.6961555480957, | |
| "rewards/rejected": -35.64251708984375, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 8.81063122923588, | |
| "grad_norm": 0.03176463395357132, | |
| "learning_rate": 1.9309388911139427e-07, | |
| "logits/chosen": -0.220908984541893, | |
| "logits/rejected": -0.29246407747268677, | |
| "logps/chosen": -953.7644653320312, | |
| "logps/rejected": -1591.0057373046875, | |
| "loss": 0.038, | |
| "rewards/accuracies": 0.9453125596046448, | |
| "rewards/chosen": 9.227982521057129, | |
| "rewards/margins": 43.120140075683594, | |
| "rewards/rejected": -33.89215850830078, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 8.877076411960132, | |
| "grad_norm": 0.0467684343457222, | |
| "learning_rate": 1.713006526846439e-07, | |
| "logits/chosen": -0.20715700089931488, | |
| "logits/rejected": -0.279428094625473, | |
| "logps/chosen": -988.7305297851562, | |
| "logps/rejected": -1678.761474609375, | |
| "loss": 0.0197, | |
| "rewards/accuracies": 0.9718750715255737, | |
| "rewards/chosen": 9.498176574707031, | |
| "rewards/margins": 45.80161666870117, | |
| "rewards/rejected": -36.30343246459961, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 8.943521594684386, | |
| "grad_norm": 0.012005633674561977, | |
| "learning_rate": 1.507684480352292e-07, | |
| "logits/chosen": -0.2161640077829361, | |
| "logits/rejected": -0.2925984859466553, | |
| "logps/chosen": -945.5839233398438, | |
| "logps/rejected": -1591.4210205078125, | |
| "loss": 0.037, | |
| "rewards/accuracies": 0.948437511920929, | |
| "rewards/chosen": 8.958191871643066, | |
| "rewards/margins": 41.801204681396484, | |
| "rewards/rejected": -32.843013763427734, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.006644518272426, | |
| "grad_norm": 0.016238396987318993, | |
| "learning_rate": 1.31508393714177e-07, | |
| "logits/chosen": -0.22952061891555786, | |
| "logits/rejected": -0.2851778268814087, | |
| "logps/chosen": -987.5601806640625, | |
| "logps/rejected": -1585.639892578125, | |
| "loss": 0.0294, | |
| "rewards/accuracies": 0.955592155456543, | |
| "rewards/chosen": 9.82602310180664, | |
| "rewards/margins": 42.44609069824219, | |
| "rewards/rejected": -32.62006378173828, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.073089700996677, | |
| "grad_norm": 0.022422021254897118, | |
| "learning_rate": 1.1353091938067024e-07, | |
| "logits/chosen": -0.22749130427837372, | |
| "logits/rejected": -0.29552769660949707, | |
| "logps/chosen": -969.6675415039062, | |
| "logps/rejected": -1588.7435302734375, | |
| "loss": 0.0402, | |
| "rewards/accuracies": 0.942187488079071, | |
| "rewards/chosen": 9.359356880187988, | |
| "rewards/margins": 42.008140563964844, | |
| "rewards/rejected": -32.64878845214844, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.13953488372093, | |
| "grad_norm": 0.005254440009593964, | |
| "learning_rate": 9.684576015420277e-08, | |
| "logits/chosen": -0.2245253324508667, | |
| "logits/rejected": -0.2839411795139313, | |
| "logps/chosen": -979.505615234375, | |
| "logps/rejected": -1624.649169921875, | |
| "loss": 0.0272, | |
| "rewards/accuracies": 0.9609375596046448, | |
| "rewards/chosen": 9.78758716583252, | |
| "rewards/margins": 43.68372344970703, | |
| "rewards/rejected": -33.89613723754883, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.205980066445182, | |
| "grad_norm": 0.02063533291220665, | |
| "learning_rate": 8.146195134284052e-08, | |
| "logits/chosen": -0.21092993021011353, | |
| "logits/rejected": -0.28807392716407776, | |
| "logps/chosen": -975.6532592773438, | |
| "logps/rejected": -1666.35107421875, | |
| "loss": 0.0261, | |
| "rewards/accuracies": 0.9625000953674316, | |
| "rewards/chosen": 9.447478294372559, | |
| "rewards/margins": 43.728919982910156, | |
| "rewards/rejected": -34.28144073486328, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.272425249169435, | |
| "grad_norm": 0.09783101826906204, | |
| "learning_rate": 6.738782355044048e-08, | |
| "logits/chosen": -0.2123195081949234, | |
| "logits/rejected": -0.2877689301967621, | |
| "logps/chosen": -948.8867797851562, | |
| "logps/rejected": -1596.3294677734375, | |
| "loss": 0.0316, | |
| "rewards/accuracies": 0.9546874761581421, | |
| "rewards/chosen": 9.14689826965332, | |
| "rewards/margins": 42.3780517578125, | |
| "rewards/rejected": -33.23115158081055, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.338870431893687, | |
| "grad_norm": 0.02494479902088642, | |
| "learning_rate": 5.463099816548578e-08, | |
| "logits/chosen": -0.21241770684719086, | |
| "logits/rejected": -0.2796427309513092, | |
| "logps/chosen": -1045.521484375, | |
| "logps/rejected": -1674.2296142578125, | |
| "loss": 0.0186, | |
| "rewards/accuracies": 0.973437488079071, | |
| "rewards/chosen": 9.795973777770996, | |
| "rewards/margins": 45.49605941772461, | |
| "rewards/rejected": -35.70008850097656, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.40531561461794, | |
| "grad_norm": 0.004993575159460306, | |
| "learning_rate": 4.319838323396691e-08, | |
| "logits/chosen": -0.21156074106693268, | |
| "logits/rejected": -0.2889597713947296, | |
| "logps/chosen": -934.5477294921875, | |
| "logps/rejected": -1561.5242919921875, | |
| "loss": 0.0295, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 8.935663223266602, | |
| "rewards/margins": 42.12391662597656, | |
| "rewards/rejected": -33.18825149536133, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.471760797342192, | |
| "grad_norm": 0.003986136522144079, | |
| "learning_rate": 3.309616971855195e-08, | |
| "logits/chosen": -0.22422273457050323, | |
| "logits/rejected": -0.2854507267475128, | |
| "logps/chosen": -1002.1636352539062, | |
| "logps/rejected": -1622.54150390625, | |
| "loss": 0.0293, | |
| "rewards/accuracies": 0.9593750238418579, | |
| "rewards/chosen": 9.735203742980957, | |
| "rewards/margins": 43.484405517578125, | |
| "rewards/rejected": -33.749202728271484, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.538205980066445, | |
| "grad_norm": 0.006039198487997055, | |
| "learning_rate": 2.4329828146074096e-08, | |
| "logits/chosen": -0.21182821691036224, | |
| "logits/rejected": -0.2917216420173645, | |
| "logps/chosen": -967.4940795898438, | |
| "logps/rejected": -1642.61474609375, | |
| "loss": 0.0349, | |
| "rewards/accuracies": 0.949999988079071, | |
| "rewards/chosen": 9.229864120483398, | |
| "rewards/margins": 43.08539962768555, | |
| "rewards/rejected": -33.85553741455078, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 9.604651162790697, | |
| "grad_norm": 0.02733450196683407, | |
| "learning_rate": 1.6904105645142443e-08, | |
| "logits/chosen": -0.2116456925868988, | |
| "logits/rejected": -0.28832632303237915, | |
| "logps/chosen": -948.6715087890625, | |
| "logps/rejected": -1624.857666015625, | |
| "loss": 0.0298, | |
| "rewards/accuracies": 0.957812488079071, | |
| "rewards/chosen": 9.185124397277832, | |
| "rewards/margins": 43.26292419433594, | |
| "rewards/rejected": -34.077796936035156, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.67109634551495, | |
| "grad_norm": 0.01584717258810997, | |
| "learning_rate": 1.0823023375489128e-08, | |
| "logits/chosen": -0.20935006439685822, | |
| "logits/rejected": -0.2797791361808777, | |
| "logps/chosen": -1003.3470458984375, | |
| "logps/rejected": -1671.807373046875, | |
| "loss": 0.0207, | |
| "rewards/accuracies": 0.9703124761581421, | |
| "rewards/chosen": 9.804126739501953, | |
| "rewards/margins": 44.94650650024414, | |
| "rewards/rejected": -35.14237976074219, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 9.737541528239202, | |
| "grad_norm": 0.06928031146526337, | |
| "learning_rate": 6.089874350439507e-09, | |
| "logits/chosen": -0.20858515799045563, | |
| "logits/rejected": -0.2875511348247528, | |
| "logps/chosen": -989.045654296875, | |
| "logps/rejected": -1670.74755859375, | |
| "loss": 0.0338, | |
| "rewards/accuracies": 0.9515625238418579, | |
| "rewards/chosen": 8.989653587341309, | |
| "rewards/margins": 45.47593307495117, | |
| "rewards/rejected": -36.48627853393555, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 9.803986710963455, | |
| "grad_norm": 0.0877932533621788, | |
| "learning_rate": 2.7072216536885855e-09, | |
| "logits/chosen": -0.21547669172286987, | |
| "logits/rejected": -0.2913072109222412, | |
| "logps/chosen": -976.333740234375, | |
| "logps/rejected": -1599.1285400390625, | |
| "loss": 0.0359, | |
| "rewards/accuracies": 0.9484375715255737, | |
| "rewards/chosen": 9.221257209777832, | |
| "rewards/margins": 42.044002532958984, | |
| "rewards/rejected": -32.82274627685547, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 9.870431893687707, | |
| "grad_norm": 0.000816255051176995, | |
| "learning_rate": 6.768970513457151e-10, | |
| "logits/chosen": -0.2026660144329071, | |
| "logits/rejected": -0.2809537649154663, | |
| "logps/chosen": -1004.3927001953125, | |
| "logps/rejected": -1700.61376953125, | |
| "loss": 0.0196, | |
| "rewards/accuracies": 0.971875011920929, | |
| "rewards/chosen": 9.102788925170898, | |
| "rewards/margins": 45.093666076660156, | |
| "rewards/rejected": -35.99087905883789, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 9.93687707641196, | |
| "grad_norm": 0.11716156452894211, | |
| "learning_rate": 0.0, | |
| "logits/chosen": -0.21064449846744537, | |
| "logits/rejected": -0.28609463572502136, | |
| "logps/chosen": -1000.4492797851562, | |
| "logps/rejected": -1652.978515625, | |
| "loss": 0.0272, | |
| "rewards/accuracies": 0.9609375, | |
| "rewards/chosen": 9.290279388427734, | |
| "rewards/margins": 44.490787506103516, | |
| "rewards/rejected": -35.20050811767578, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.93687707641196, | |
| "step": 1500, | |
| "total_flos": 2.9219422466201354e+19, | |
| "train_loss": 0.07426366006334623, | |
| "train_runtime": 24708.7747, | |
| "train_samples_per_second": 3.897, | |
| "train_steps_per_second": 0.061 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.9219422466201354e+19, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |