| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 1250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4e-08, | |
| "logits/chosen": 1.175201177597046, | |
| "logits/rejected": 1.1360995769500732, | |
| "logps/chosen": -464.5503234863281, | |
| "logps/rejected": -320.97235107421875, | |
| "loss": 0.1112, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "logits/chosen": 1.2028732299804688, | |
| "logits/rejected": 1.0901875495910645, | |
| "logps/chosen": -496.0768127441406, | |
| "logps/rejected": -346.5080871582031, | |
| "loss": 0.1476, | |
| "rewards/accuracies": 0.3263888955116272, | |
| "rewards/chosen": -6.533772102557123e-05, | |
| "rewards/margins": -9.996786684496328e-05, | |
| "rewards/rejected": 3.4630156733328477e-05, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 8.000000000000001e-07, | |
| "logits/chosen": 1.1378858089447021, | |
| "logits/rejected": 1.0073139667510986, | |
| "logps/chosen": -502.5614318847656, | |
| "logps/rejected": -353.5082092285156, | |
| "loss": 0.1581, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": 2.500582195352763e-05, | |
| "rewards/margins": -1.3462633432936855e-05, | |
| "rewards/rejected": 3.8468460843432695e-05, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "logits/chosen": 1.1466845273971558, | |
| "logits/rejected": 1.0465342998504639, | |
| "logps/chosen": -480.00213623046875, | |
| "logps/rejected": -354.7355651855469, | |
| "loss": 0.1352, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": 0.00011489469034131616, | |
| "rewards/margins": 5.035057620261796e-05, | |
| "rewards/rejected": 6.454410322476178e-05, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "logits/chosen": 1.0807245969772339, | |
| "logits/rejected": 1.0223329067230225, | |
| "logps/chosen": -480.26605224609375, | |
| "logps/rejected": -352.97113037109375, | |
| "loss": 0.1494, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": 0.00020178986596874893, | |
| "rewards/margins": 5.12802398588974e-05, | |
| "rewards/rejected": 0.0001505096151959151, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "logits/chosen": 1.1475070714950562, | |
| "logits/rejected": 1.1084539890289307, | |
| "logps/chosen": -486.21875, | |
| "logps/rejected": -360.5682373046875, | |
| "loss": 0.1481, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": 0.0002630269154906273, | |
| "rewards/margins": 0.00012177646567579359, | |
| "rewards/rejected": 0.00014125046436674893, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "logits/chosen": 1.1609418392181396, | |
| "logits/rejected": 1.028495192527771, | |
| "logps/chosen": -490.59375, | |
| "logps/rejected": -344.62884521484375, | |
| "loss": 0.1493, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": 0.0005147407646290958, | |
| "rewards/margins": 0.0002236834552604705, | |
| "rewards/rejected": 0.0002910573093686253, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "logits/chosen": 1.0503188371658325, | |
| "logits/rejected": 1.0258784294128418, | |
| "logps/chosen": -476.06292724609375, | |
| "logps/rejected": -371.3968200683594, | |
| "loss": 0.151, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": 0.0010026374366134405, | |
| "rewards/margins": 0.00048288650577887893, | |
| "rewards/rejected": 0.0005197509890422225, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "logits/chosen": 1.1313526630401611, | |
| "logits/rejected": 1.0380247831344604, | |
| "logps/chosen": -487.41448974609375, | |
| "logps/rejected": -360.5604248046875, | |
| "loss": 0.1481, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.0016277392860502005, | |
| "rewards/margins": 0.0009324050624854863, | |
| "rewards/rejected": 0.0006953343981876969, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "logits/chosen": 1.1803278923034668, | |
| "logits/rejected": 1.005684494972229, | |
| "logps/chosen": -474.18878173828125, | |
| "logps/rejected": -357.3034973144531, | |
| "loss": 0.1409, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": 0.002065226435661316, | |
| "rewards/margins": 0.0008295776206068695, | |
| "rewards/rejected": 0.0012356489896774292, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.000000000000001e-06, | |
| "logits/chosen": 1.1424959897994995, | |
| "logits/rejected": 1.037285327911377, | |
| "logps/chosen": -485.64605712890625, | |
| "logps/rejected": -341.20599365234375, | |
| "loss": 0.1335, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.002735253656283021, | |
| "rewards/margins": 0.0017322547500953078, | |
| "rewards/rejected": 0.001002999022603035, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.4e-06, | |
| "logits/chosen": 1.0868542194366455, | |
| "logits/rejected": 1.029909372329712, | |
| "logps/chosen": -492.3543395996094, | |
| "logps/rejected": -320.5608215332031, | |
| "loss": 0.1337, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": 0.005243994295597076, | |
| "rewards/margins": 0.0035049724392592907, | |
| "rewards/rejected": 0.001739021041430533, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.800000000000001e-06, | |
| "logits/chosen": 1.1244226694107056, | |
| "logits/rejected": 1.0855529308319092, | |
| "logps/chosen": -462.7578125, | |
| "logps/rejected": -330.9637145996094, | |
| "loss": 0.1543, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": 0.006767258048057556, | |
| "rewards/margins": 0.004428135231137276, | |
| "rewards/rejected": 0.0023391232825815678, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.999756310023261e-06, | |
| "logits/chosen": 1.1845452785491943, | |
| "logits/rejected": 1.0370099544525146, | |
| "logps/chosen": -475.958251953125, | |
| "logps/rejected": -337.60552978515625, | |
| "loss": 0.1206, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.010690492577850819, | |
| "rewards/margins": 0.008029913529753685, | |
| "rewards/rejected": 0.0026605785824358463, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.997807075247147e-06, | |
| "logits/chosen": 1.1520451307296753, | |
| "logits/rejected": 1.0790607929229736, | |
| "logps/chosen": -459.16485595703125, | |
| "logps/rejected": -337.4107971191406, | |
| "loss": 0.1476, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": 0.011510485783219337, | |
| "rewards/margins": 0.008873937651515007, | |
| "rewards/rejected": 0.002636546967551112, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.993910125649561e-06, | |
| "logits/chosen": 1.1496937274932861, | |
| "logits/rejected": 1.0804468393325806, | |
| "logps/chosen": -469.76507568359375, | |
| "logps/rejected": -343.68109130859375, | |
| "loss": 0.1393, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": 0.01237241830676794, | |
| "rewards/margins": 0.01088970061391592, | |
| "rewards/rejected": 0.0014827173436060548, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.988068499954578e-06, | |
| "logits/chosen": 1.0382554531097412, | |
| "logits/rejected": 0.9245996475219727, | |
| "logps/chosen": -454.4856872558594, | |
| "logps/rejected": -329.7301330566406, | |
| "loss": 0.1229, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.011257833801209927, | |
| "rewards/margins": 0.010106958448886871, | |
| "rewards/rejected": 0.0011508769821375608, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.980286753286196e-06, | |
| "logits/chosen": 1.1908769607543945, | |
| "logits/rejected": 1.1219961643218994, | |
| "logps/chosen": -479.1846618652344, | |
| "logps/rejected": -360.87335205078125, | |
| "loss": 0.1445, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": 0.012364432215690613, | |
| "rewards/margins": 0.017936835065484047, | |
| "rewards/rejected": -0.0055724019184708595, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.970570953616383e-06, | |
| "logits/chosen": 1.0914846658706665, | |
| "logits/rejected": 1.023158311843872, | |
| "logps/chosen": -457.4100646972656, | |
| "logps/rejected": -361.01800537109375, | |
| "loss": 0.1289, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.009392337873578072, | |
| "rewards/margins": 0.020237213000655174, | |
| "rewards/rejected": -0.010844876058399677, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.958928677033465e-06, | |
| "logits/chosen": 1.1615819931030273, | |
| "logits/rejected": 1.0536638498306274, | |
| "logps/chosen": -483.87261962890625, | |
| "logps/rejected": -402.1483459472656, | |
| "loss": 0.1285, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": 0.005232605151832104, | |
| "rewards/margins": 0.028785843402147293, | |
| "rewards/rejected": -0.023553235456347466, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.9453690018345144e-06, | |
| "logits/chosen": 1.1142406463623047, | |
| "logits/rejected": 1.008840799331665, | |
| "logps/chosen": -462.1014709472656, | |
| "logps/rejected": -354.94744873046875, | |
| "loss": 0.132, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": 0.0019020054023712873, | |
| "rewards/margins": 0.02114967629313469, | |
| "rewards/rejected": -0.019247671589255333, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.9299025014463665e-06, | |
| "logits/chosen": 1.0533361434936523, | |
| "logits/rejected": 0.983782947063446, | |
| "logps/chosen": -465.71258544921875, | |
| "logps/rejected": -364.546875, | |
| "loss": 0.1351, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": 0.005812919698655605, | |
| "rewards/margins": 0.0318826325237751, | |
| "rewards/rejected": -0.02606971189379692, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.912541236180779e-06, | |
| "logits/chosen": 1.0279268026351929, | |
| "logits/rejected": 1.018845558166504, | |
| "logps/chosen": -454.9180603027344, | |
| "logps/rejected": -382.669677734375, | |
| "loss": 0.144, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": 0.007858935743570328, | |
| "rewards/margins": 0.029104083776474, | |
| "rewards/rejected": -0.02124514803290367, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.893298743830168e-06, | |
| "logits/chosen": 1.093308925628662, | |
| "logits/rejected": 0.9972928762435913, | |
| "logps/chosen": -486.96795654296875, | |
| "logps/rejected": -365.38653564453125, | |
| "loss": 0.141, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.0009560141479596496, | |
| "rewards/margins": 0.026852011680603027, | |
| "rewards/rejected": -0.025895997881889343, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.8721900291112415e-06, | |
| "logits/chosen": 1.1143492460250854, | |
| "logits/rejected": 1.011824369430542, | |
| "logps/chosen": -462.26202392578125, | |
| "logps/rejected": -372.9997863769531, | |
| "loss": 0.138, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": 0.010432013310492039, | |
| "rewards/margins": 0.028673967346549034, | |
| "rewards/rejected": -0.01824195310473442, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.849231551964771e-06, | |
| "logits/chosen": 1.0901610851287842, | |
| "logits/rejected": 0.9668253660202026, | |
| "logps/chosen": -450.9900817871094, | |
| "logps/rejected": -366.1516418457031, | |
| "loss": 0.121, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": 0.013042752631008625, | |
| "rewards/margins": 0.037192780524492264, | |
| "rewards/rejected": -0.024150025099515915, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.824441214720629e-06, | |
| "logits/chosen": 1.0770037174224854, | |
| "logits/rejected": 0.9468703269958496, | |
| "logps/chosen": -477.5690002441406, | |
| "logps/rejected": -331.53826904296875, | |
| "loss": 0.1342, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": 0.01135487761348486, | |
| "rewards/margins": 0.028281161561608315, | |
| "rewards/rejected": -0.01692628487944603, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.7978383481380865e-06, | |
| "logits/chosen": 1.0577478408813477, | |
| "logits/rejected": 0.9135805368423462, | |
| "logps/chosen": -453.3644104003906, | |
| "logps/rejected": -318.5481872558594, | |
| "loss": 0.1317, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.010126395151019096, | |
| "rewards/margins": 0.028996825218200684, | |
| "rewards/rejected": -0.018870430067181587, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.769443696332272e-06, | |
| "logits/chosen": 1.1663763523101807, | |
| "logits/rejected": 1.0382214784622192, | |
| "logps/chosen": -473.4112243652344, | |
| "logps/rejected": -363.4704284667969, | |
| "loss": 0.1224, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.01480911485850811, | |
| "rewards/margins": 0.03443712741136551, | |
| "rewards/rejected": -0.019628014415502548, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.7392794005985324e-06, | |
| "logits/chosen": 1.1509085893630981, | |
| "logits/rejected": 1.0418548583984375, | |
| "logps/chosen": -443.42169189453125, | |
| "logps/rejected": -352.11248779296875, | |
| "loss": 0.1454, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.01802745647728443, | |
| "rewards/margins": 0.037281010299921036, | |
| "rewards/rejected": -0.019253555685281754, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.707368982147318e-06, | |
| "logits/chosen": 1.066710114479065, | |
| "logits/rejected": 0.9809983372688293, | |
| "logps/chosen": -453.10662841796875, | |
| "logps/rejected": -366.565185546875, | |
| "loss": 0.1345, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": 0.012954890727996826, | |
| "rewards/margins": 0.02726684883236885, | |
| "rewards/rejected": -0.014311959967017174, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.673737323763048e-06, | |
| "logits/chosen": 1.031423568725586, | |
| "logits/rejected": 0.9241514205932617, | |
| "logps/chosen": -459.67901611328125, | |
| "logps/rejected": -367.7610778808594, | |
| "loss": 0.1287, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": 0.015265332534909248, | |
| "rewards/margins": 0.025473449379205704, | |
| "rewards/rejected": -0.01020811591297388, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.638410650401267e-06, | |
| "logits/chosen": 1.131606101989746, | |
| "logits/rejected": 1.0299171209335327, | |
| "logps/chosen": -478.57403564453125, | |
| "logps/rejected": -387.32318115234375, | |
| "loss": 0.1008, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": 0.026522139087319374, | |
| "rewards/margins": 0.04347400367259979, | |
| "rewards/rejected": -0.01695186272263527, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.601416508739211e-06, | |
| "logits/chosen": 1.1088566780090332, | |
| "logits/rejected": 0.9741806983947754, | |
| "logps/chosen": -460.74090576171875, | |
| "logps/rejected": -381.3876953125, | |
| "loss": 0.1276, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": 0.019920390099287033, | |
| "rewards/margins": 0.04846935719251633, | |
| "rewards/rejected": -0.028548967093229294, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.562783745695738e-06, | |
| "logits/chosen": 1.0776937007904053, | |
| "logits/rejected": 0.9450405836105347, | |
| "logps/chosen": -461.37115478515625, | |
| "logps/rejected": -381.6752624511719, | |
| "loss": 0.1173, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.01616843044757843, | |
| "rewards/margins": 0.04438379779458046, | |
| "rewards/rejected": -0.02821536734700203, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.522542485937369e-06, | |
| "logits/chosen": 0.9827763438224792, | |
| "logits/rejected": 0.854051947593689, | |
| "logps/chosen": -490.47021484375, | |
| "logps/rejected": -389.33197021484375, | |
| "loss": 0.1149, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": 0.018763616681098938, | |
| "rewards/margins": 0.06531712412834167, | |
| "rewards/rejected": -0.04655350744724274, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.4807241083879774e-06, | |
| "logits/chosen": 1.0331618785858154, | |
| "logits/rejected": 0.9459377527236938, | |
| "logps/chosen": -487.89007568359375, | |
| "logps/rejected": -382.6471862792969, | |
| "loss": 0.1528, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.005054169334471226, | |
| "rewards/margins": 0.03303053602576256, | |
| "rewards/rejected": -0.03808470442891121, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.437361221760449e-06, | |
| "logits/chosen": 0.9949855804443359, | |
| "logits/rejected": 0.8790602684020996, | |
| "logps/chosen": -473.7728576660156, | |
| "logps/rejected": -383.35467529296875, | |
| "loss": 0.1473, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.020323097705841064, | |
| "rewards/margins": 0.04043232277035713, | |
| "rewards/rejected": -0.020109226927161217, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.3924876391293915e-06, | |
| "logits/chosen": 1.079866647720337, | |
| "logits/rejected": 0.9681353569030762, | |
| "logps/chosen": -471.4029235839844, | |
| "logps/rejected": -399.40728759765625, | |
| "loss": 0.1079, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": 0.024924132972955704, | |
| "rewards/margins": 0.05287875980138779, | |
| "rewards/rejected": -0.02795463241636753, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.346138351564711e-06, | |
| "logits/chosen": 0.9959014654159546, | |
| "logits/rejected": 0.9022513628005981, | |
| "logps/chosen": -448.15350341796875, | |
| "logps/rejected": -344.8275146484375, | |
| "loss": 0.1143, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.024763161316514015, | |
| "rewards/margins": 0.03656725212931633, | |
| "rewards/rejected": -0.011804090812802315, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.2983495008466285e-06, | |
| "logits/chosen": 1.0768877267837524, | |
| "logits/rejected": 1.0071511268615723, | |
| "logps/chosen": -458.2196350097656, | |
| "logps/rejected": -370.3987731933594, | |
| "loss": 0.1363, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.023127641528844833, | |
| "rewards/margins": 0.03729063645005226, | |
| "rewards/rejected": -0.014162993058562279, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.249158351283414e-06, | |
| "logits/chosen": 1.0073800086975098, | |
| "logits/rejected": 0.93903648853302, | |
| "logps/chosen": -478.2962341308594, | |
| "logps/rejected": -367.376953125, | |
| "loss": 0.134, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.021456506103277206, | |
| "rewards/margins": 0.04471339285373688, | |
| "rewards/rejected": -0.02325688675045967, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.198603260653792e-06, | |
| "logits/chosen": 0.9648386240005493, | |
| "logits/rejected": 0.8528832197189331, | |
| "logps/chosen": -476.27001953125, | |
| "logps/rejected": -355.2597351074219, | |
| "loss": 0.1368, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": 0.018124157562851906, | |
| "rewards/margins": 0.03248673677444458, | |
| "rewards/rejected": -0.014362581074237823, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.146723650296701e-06, | |
| "logits/chosen": 1.0027981996536255, | |
| "logits/rejected": 0.8345006704330444, | |
| "logps/chosen": -468.65771484375, | |
| "logps/rejected": -377.0935363769531, | |
| "loss": 0.1226, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.01800473965704441, | |
| "rewards/margins": 0.04834091290831566, | |
| "rewards/rejected": -0.030336176976561546, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.093559974371725e-06, | |
| "logits/chosen": 0.9172792434692383, | |
| "logits/rejected": 0.8375406265258789, | |
| "logps/chosen": -466.71405029296875, | |
| "logps/rejected": -413.31903076171875, | |
| "loss": 0.1127, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": 0.005665500648319721, | |
| "rewards/margins": 0.05071156471967697, | |
| "rewards/rejected": -0.04504606872797012, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.039153688314146e-06, | |
| "logits/chosen": 0.9984884262084961, | |
| "logits/rejected": 0.8194684982299805, | |
| "logps/chosen": -474.7478942871094, | |
| "logps/rejected": -382.6129150390625, | |
| "loss": 0.1281, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.0018371727783232927, | |
| "rewards/margins": 0.054125331342220306, | |
| "rewards/rejected": -0.052288156002759933, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.983547216509254e-06, | |
| "logits/chosen": 0.9451436996459961, | |
| "logits/rejected": 0.8137146830558777, | |
| "logps/chosen": -462.51934814453125, | |
| "logps/rejected": -392.1611328125, | |
| "loss": 0.1388, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": 0.004337290767580271, | |
| "rewards/margins": 0.0444575771689415, | |
| "rewards/rejected": -0.04012028127908707, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.92678391921108e-06, | |
| "logits/chosen": 0.9080266952514648, | |
| "logits/rejected": 0.7958904504776001, | |
| "logps/chosen": -459.026123046875, | |
| "logps/rejected": -364.5742492675781, | |
| "loss": 0.133, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": 0.00981406681239605, | |
| "rewards/margins": 0.04570726677775383, | |
| "rewards/rejected": -0.03589319437742233, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.868908058731376e-06, | |
| "logits/chosen": 1.0409977436065674, | |
| "logits/rejected": 0.8651460409164429, | |
| "logps/chosen": -467.898681640625, | |
| "logps/rejected": -348.7823791503906, | |
| "loss": 0.1441, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": 0.005097855813801289, | |
| "rewards/margins": 0.03613067790865898, | |
| "rewards/rejected": -0.03103281930088997, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.8099647649251984e-06, | |
| "logits/chosen": 1.08036208152771, | |
| "logits/rejected": 1.0272912979125977, | |
| "logps/chosen": -471.68603515625, | |
| "logps/rejected": -395.9718322753906, | |
| "loss": 0.1202, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": 0.01659775711596012, | |
| "rewards/margins": 0.0396357998251915, | |
| "rewards/rejected": -0.023038046434521675, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "logits/chosen": 1.0317163467407227, | |
| "logits/rejected": 0.8968293070793152, | |
| "logps/chosen": -467.38671875, | |
| "logps/rejected": -378.4510192871094, | |
| "loss": 0.1347, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.018395045772194862, | |
| "rewards/margins": 0.029253458604216576, | |
| "rewards/rejected": -0.010858410969376564, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.689060522675689e-06, | |
| "logits/chosen": 0.9803446531295776, | |
| "logits/rejected": 0.8904365301132202, | |
| "logps/chosen": -434.4021911621094, | |
| "logps/rejected": -359.31134033203125, | |
| "loss": 0.125, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.019461948424577713, | |
| "rewards/margins": 0.04449521377682686, | |
| "rewards/rejected": -0.025033265352249146, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.627193851723577e-06, | |
| "logits/chosen": 0.9237390756607056, | |
| "logits/rejected": 0.758080780506134, | |
| "logps/chosen": -461.304443359375, | |
| "logps/rejected": -363.09490966796875, | |
| "loss": 0.1178, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.015167142264544964, | |
| "rewards/margins": 0.053153105080127716, | |
| "rewards/rejected": -0.03798595815896988, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.564448228912682e-06, | |
| "logits/chosen": 0.8934131860733032, | |
| "logits/rejected": 0.722560703754425, | |
| "logps/chosen": -473.1156311035156, | |
| "logps/rejected": -344.38458251953125, | |
| "loss": 0.1304, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": 0.011030582711100578, | |
| "rewards/margins": 0.026756322011351585, | |
| "rewards/rejected": -0.015725741162896156, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.5008725813922383e-06, | |
| "logits/chosen": 0.9554192423820496, | |
| "logits/rejected": 0.8335037231445312, | |
| "logps/chosen": -447.8621520996094, | |
| "logps/rejected": -343.98187255859375, | |
| "loss": 0.1268, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.019363736733794212, | |
| "rewards/margins": 0.048774898052215576, | |
| "rewards/rejected": -0.029411161318421364, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.436516483539781e-06, | |
| "logits/chosen": 0.9713459014892578, | |
| "logits/rejected": 0.851312518119812, | |
| "logps/chosen": -462.87078857421875, | |
| "logps/rejected": -350.909912109375, | |
| "loss": 0.1226, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.021459558978676796, | |
| "rewards/margins": 0.037843555212020874, | |
| "rewards/rejected": -0.016383999958634377, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.3714301183045382e-06, | |
| "logits/chosen": 0.9322063326835632, | |
| "logits/rejected": 0.8468168377876282, | |
| "logps/chosen": -451.48394775390625, | |
| "logps/rejected": -363.7984313964844, | |
| "loss": 0.1178, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": 0.025396600365638733, | |
| "rewards/margins": 0.04970560222864151, | |
| "rewards/rejected": -0.024309005588293076, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.3056642380762783e-06, | |
| "logits/chosen": 0.9000357389450073, | |
| "logits/rejected": 0.7963394522666931, | |
| "logps/chosen": -429.16839599609375, | |
| "logps/rejected": -349.9974670410156, | |
| "loss": 0.1462, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": 0.019953671842813492, | |
| "rewards/margins": 0.03375135362148285, | |
| "rewards/rejected": -0.013797680847346783, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.2392701251101172e-06, | |
| "logits/chosen": 0.8540957570075989, | |
| "logits/rejected": 0.801906406879425, | |
| "logps/chosen": -481.28302001953125, | |
| "logps/rejected": -398.15185546875, | |
| "loss": 0.1214, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": 0.023893360048532486, | |
| "rewards/margins": 0.04255540296435356, | |
| "rewards/rejected": -0.018662041053175926, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.1722995515381644e-06, | |
| "logits/chosen": 0.9093323945999146, | |
| "logits/rejected": 0.7911072373390198, | |
| "logps/chosen": -480.9325256347656, | |
| "logps/rejected": -374.56549072265625, | |
| "loss": 0.128, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.011839361861348152, | |
| "rewards/margins": 0.038532059639692307, | |
| "rewards/rejected": -0.026692699640989304, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.1048047389991693e-06, | |
| "logits/chosen": 0.9539586901664734, | |
| "logits/rejected": 0.832242488861084, | |
| "logps/chosen": -478.30950927734375, | |
| "logps/rejected": -388.95770263671875, | |
| "loss": 0.1141, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.013863136060535908, | |
| "rewards/margins": 0.05301787331700325, | |
| "rewards/rejected": -0.03915473818778992, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.0368383179176584e-06, | |
| "logits/chosen": 0.8396176099777222, | |
| "logits/rejected": 0.7142072319984436, | |
| "logps/chosen": -461.75445556640625, | |
| "logps/rejected": -371.7435607910156, | |
| "loss": 0.1417, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.02106183022260666, | |
| "rewards/margins": 0.050181370228528976, | |
| "rewards/rejected": -0.029119547456502914, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.9684532864643123e-06, | |
| "logits/chosen": 0.8672172427177429, | |
| "logits/rejected": 0.7649189233779907, | |
| "logps/chosen": -483.3880310058594, | |
| "logps/rejected": -401.4527587890625, | |
| "loss": 0.1144, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.0037638209760189056, | |
| "rewards/margins": 0.04919921234250069, | |
| "rewards/rejected": -0.04543538764119148, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.8997029692295875e-06, | |
| "logits/chosen": 0.8982945680618286, | |
| "logits/rejected": 0.7633360028266907, | |
| "logps/chosen": -453.81243896484375, | |
| "logps/rejected": -408.51947021484375, | |
| "loss": 0.1196, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.009529736824333668, | |
| "rewards/margins": 0.05482906103134155, | |
| "rewards/rejected": -0.04529931768774986, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.8306409756428067e-06, | |
| "logits/chosen": 0.9960956573486328, | |
| "logits/rejected": 0.7809866666793823, | |
| "logps/chosen": -486.93914794921875, | |
| "logps/rejected": -397.1849365234375, | |
| "loss": 0.0939, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": 0.016234764829277992, | |
| "rewards/margins": 0.06638605892658234, | |
| "rewards/rejected": -0.05015129595994949, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.761321158169134e-06, | |
| "logits/chosen": 0.9853866696357727, | |
| "logits/rejected": 0.7747653722763062, | |
| "logps/chosen": -485.00543212890625, | |
| "logps/rejected": -431.0743103027344, | |
| "loss": 0.117, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": 0.013629300519824028, | |
| "rewards/margins": 0.07910068333148956, | |
| "rewards/rejected": -0.06547138094902039, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.6917975703170466e-06, | |
| "logits/chosen": 0.9036257863044739, | |
| "logits/rejected": 0.7418056130409241, | |
| "logps/chosen": -483.9380798339844, | |
| "logps/rejected": -409.62030029296875, | |
| "loss": 0.1173, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.0023396513424813747, | |
| "rewards/margins": 0.05056001991033554, | |
| "rewards/rejected": -0.052899666130542755, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.6221244244890336e-06, | |
| "logits/chosen": 0.9365898370742798, | |
| "logits/rejected": 0.7452448606491089, | |
| "logps/chosen": -471.79486083984375, | |
| "logps/rejected": -403.4354248046875, | |
| "loss": 0.1282, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.00022033098503015935, | |
| "rewards/margins": 0.04656386375427246, | |
| "rewards/rejected": -0.04634353518486023, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5523560497083927e-06, | |
| "logits/chosen": 0.8479242324829102, | |
| "logits/rejected": 0.7285875082015991, | |
| "logps/chosen": -468.72674560546875, | |
| "logps/rejected": -373.24932861328125, | |
| "loss": 0.1469, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -0.005256607197225094, | |
| "rewards/margins": 0.03616983816027641, | |
| "rewards/rejected": -0.041426438838243484, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.482546849255096e-06, | |
| "logits/chosen": 0.865423321723938, | |
| "logits/rejected": 0.7504838705062866, | |
| "logps/chosen": -451.1360778808594, | |
| "logps/rejected": -367.5789489746094, | |
| "loss": 0.1223, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.014548267237842083, | |
| "rewards/margins": 0.04971490055322647, | |
| "rewards/rejected": -0.035166628658771515, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.4127512582437486e-06, | |
| "logits/chosen": 0.9130813479423523, | |
| "logits/rejected": 0.7521572709083557, | |
| "logps/chosen": -467.348388671875, | |
| "logps/rejected": -368.24505615234375, | |
| "loss": 0.1175, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.015271159820258617, | |
| "rewards/margins": 0.05196915939450264, | |
| "rewards/rejected": -0.036698006093502045, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.3430237011767166e-06, | |
| "logits/chosen": 0.8675438165664673, | |
| "logits/rejected": 0.7256118655204773, | |
| "logps/chosen": -460.74560546875, | |
| "logps/rejected": -374.3150329589844, | |
| "loss": 0.1129, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": 0.013036970980465412, | |
| "rewards/margins": 0.03715645894408226, | |
| "rewards/rejected": -0.024119490757584572, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.2734185495055503e-06, | |
| "logits/chosen": 0.935188889503479, | |
| "logits/rejected": 0.7854949831962585, | |
| "logps/chosen": -479.2552795410156, | |
| "logps/rejected": -388.42626953125, | |
| "loss": 0.1274, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": 0.014768016524612904, | |
| "rewards/margins": 0.04487290605902672, | |
| "rewards/rejected": -0.03010489046573639, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.2039900792337477e-06, | |
| "logits/chosen": 0.9271212816238403, | |
| "logits/rejected": 0.8139986991882324, | |
| "logps/chosen": -468.4580993652344, | |
| "logps/rejected": -374.9727783203125, | |
| "loss": 0.111, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.016965588554739952, | |
| "rewards/margins": 0.04319929704070091, | |
| "rewards/rejected": -0.02623371221125126, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.134792428593971e-06, | |
| "logits/chosen": 0.9355592727661133, | |
| "logits/rejected": 0.7876889705657959, | |
| "logps/chosen": -453.23345947265625, | |
| "logps/rejected": -374.57196044921875, | |
| "loss": 0.135, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": 0.007375324610620737, | |
| "rewards/margins": 0.034845925867557526, | |
| "rewards/rejected": -0.027470603585243225, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.0658795558326745e-06, | |
| "logits/chosen": 0.9382299184799194, | |
| "logits/rejected": 0.7689005732536316, | |
| "logps/chosen": -443.07867431640625, | |
| "logps/rejected": -385.6949157714844, | |
| "loss": 0.1109, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": 0.013460439629852772, | |
| "rewards/margins": 0.04786471277475357, | |
| "rewards/rejected": -0.034404270350933075, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.997305197135089e-06, | |
| "logits/chosen": 0.8568140864372253, | |
| "logits/rejected": 0.6882631182670593, | |
| "logps/chosen": -460.6326599121094, | |
| "logps/rejected": -352.6177673339844, | |
| "loss": 0.1218, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": 0.011636492796242237, | |
| "rewards/margins": 0.06006443500518799, | |
| "rewards/rejected": -0.048427946865558624, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.9291228247233607e-06, | |
| "logits/chosen": 0.8900226354598999, | |
| "logits/rejected": 0.7789937257766724, | |
| "logps/chosen": -469.8233337402344, | |
| "logps/rejected": -384.80474853515625, | |
| "loss": 0.1273, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": 0.00031064721406437457, | |
| "rewards/margins": 0.04238821938633919, | |
| "rewards/rejected": -0.042077574878931046, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.8613856051605242e-06, | |
| "logits/chosen": 0.9109789133071899, | |
| "logits/rejected": 0.7407966256141663, | |
| "logps/chosen": -455.23309326171875, | |
| "logps/rejected": -355.62457275390625, | |
| "loss": 0.1256, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": 0.007995473220944405, | |
| "rewards/margins": 0.03913170099258423, | |
| "rewards/rejected": -0.03113623335957527, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.7941463578928088e-06, | |
| "logits/chosen": 0.9689297676086426, | |
| "logits/rejected": 0.8149408102035522, | |
| "logps/chosen": -462.6448669433594, | |
| "logps/rejected": -381.04473876953125, | |
| "loss": 0.1324, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.009871283546090126, | |
| "rewards/margins": 0.04489940404891968, | |
| "rewards/rejected": -0.0350281223654747, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.7274575140626318e-06, | |
| "logits/chosen": 0.9453274011611938, | |
| "logits/rejected": 0.8604361414909363, | |
| "logps/chosen": -476.329833984375, | |
| "logps/rejected": -395.4358215332031, | |
| "loss": 0.1203, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.011838543228805065, | |
| "rewards/margins": 0.04980830103158951, | |
| "rewards/rejected": -0.03796975687146187, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.661371075624363e-06, | |
| "logits/chosen": 0.8852574229240417, | |
| "logits/rejected": 0.7384047508239746, | |
| "logps/chosen": -475.7542419433594, | |
| "logps/rejected": -397.722900390625, | |
| "loss": 0.1191, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": 0.0070769889280200005, | |
| "rewards/margins": 0.04614837095141411, | |
| "rewards/rejected": -0.03907138481736183, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.5959385747947697e-06, | |
| "logits/chosen": 0.8751562237739563, | |
| "logits/rejected": 0.7935196161270142, | |
| "logps/chosen": -469.99285888671875, | |
| "logps/rejected": -382.78521728515625, | |
| "loss": 0.1402, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.005080102477222681, | |
| "rewards/margins": 0.040591780096292496, | |
| "rewards/rejected": -0.035511672496795654, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.5312110338697427e-06, | |
| "logits/chosen": 0.8930953741073608, | |
| "logits/rejected": 0.7850891351699829, | |
| "logps/chosen": -459.15869140625, | |
| "logps/rejected": -398.9571838378906, | |
| "loss": 0.1006, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.012067661620676517, | |
| "rewards/margins": 0.05800219625234604, | |
| "rewards/rejected": -0.045934535562992096, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.467238925438646e-06, | |
| "logits/chosen": 0.8995454907417297, | |
| "logits/rejected": 0.8098067045211792, | |
| "logps/chosen": -474.0658264160156, | |
| "logps/rejected": -388.6312255859375, | |
| "loss": 0.124, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.018568776547908783, | |
| "rewards/margins": 0.056005608290433884, | |
| "rewards/rejected": -0.0374368354678154, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.4040721330273063e-06, | |
| "logits/chosen": 0.8805915117263794, | |
| "logits/rejected": 0.7850027680397034, | |
| "logps/chosen": -463.780029296875, | |
| "logps/rejected": -406.8705139160156, | |
| "loss": 0.1189, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.010667243972420692, | |
| "rewards/margins": 0.050372637808322906, | |
| "rewards/rejected": -0.039705388247966766, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.3417599122003464e-06, | |
| "logits/chosen": 0.8984284400939941, | |
| "logits/rejected": 0.7516420483589172, | |
| "logps/chosen": -491.55694580078125, | |
| "logps/rejected": -411.80413818359375, | |
| "loss": 0.1019, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.01107702311128378, | |
| "rewards/margins": 0.06236129254102707, | |
| "rewards/rejected": -0.051284272223711014, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.280350852153168e-06, | |
| "logits/chosen": 0.8879778981208801, | |
| "logits/rejected": 0.7307626008987427, | |
| "logps/chosen": -487.8158264160156, | |
| "logps/rejected": -379.3349609375, | |
| "loss": 0.1061, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": 0.012510605156421661, | |
| "rewards/margins": 0.04596661403775215, | |
| "rewards/rejected": -0.03345600515604019, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.2198928378235717e-06, | |
| "logits/chosen": 0.9071391224861145, | |
| "logits/rejected": 0.7923253774642944, | |
| "logps/chosen": -485.8133850097656, | |
| "logps/rejected": -390.81103515625, | |
| "loss": 0.1293, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": 0.007225564680993557, | |
| "rewards/margins": 0.057215720415115356, | |
| "rewards/rejected": -0.049990154802799225, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.160433012552508e-06, | |
| "logits/chosen": 0.7823120951652527, | |
| "logits/rejected": 0.7412742376327515, | |
| "logps/chosen": -452.2621154785156, | |
| "logps/rejected": -397.30126953125, | |
| "loss": 0.1244, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.0017619800055399537, | |
| "rewards/margins": 0.05425364896655083, | |
| "rewards/rejected": -0.05249166488647461, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.1020177413231334e-06, | |
| "logits/chosen": 0.8177992105484009, | |
| "logits/rejected": 0.724165141582489, | |
| "logps/chosen": -465.94244384765625, | |
| "logps/rejected": -404.88641357421875, | |
| "loss": 0.1332, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.007125381380319595, | |
| "rewards/margins": 0.05769032984972, | |
| "rewards/rejected": -0.05056494474411011, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.0446925746067768e-06, | |
| "logits/chosen": 0.9341510534286499, | |
| "logits/rejected": 0.7330499887466431, | |
| "logps/chosen": -458.97003173828125, | |
| "logps/rejected": -380.50836181640625, | |
| "loss": 0.121, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": 0.006901240907609463, | |
| "rewards/margins": 0.05187239125370979, | |
| "rewards/rejected": -0.044971149414777756, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.88502212844063e-07, | |
| "logits/chosen": 0.8761428594589233, | |
| "logits/rejected": 0.8241461515426636, | |
| "logps/chosen": -460.31109619140625, | |
| "logps/rejected": -379.6416015625, | |
| "loss": 0.105, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.009546898305416107, | |
| "rewards/margins": 0.046010397374629974, | |
| "rewards/rejected": -0.036463502794504166, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.334904715888496e-07, | |
| "logits/chosen": 0.9266616702079773, | |
| "logits/rejected": 0.7574371695518494, | |
| "logps/chosen": -491.9678649902344, | |
| "logps/rejected": -402.0565490722656, | |
| "loss": 0.1298, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": 0.0036029263865202665, | |
| "rewards/margins": 0.057801950722932816, | |
| "rewards/rejected": -0.05419902130961418, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.797002473421729e-07, | |
| "logits/chosen": 0.8647912740707397, | |
| "logits/rejected": 0.6777385473251343, | |
| "logps/chosen": -461.28460693359375, | |
| "logps/rejected": -378.638427734375, | |
| "loss": 0.1206, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": 0.017040668055415154, | |
| "rewards/margins": 0.04332052543759346, | |
| "rewards/rejected": -0.02627985179424286, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.271734841028553e-07, | |
| "logits/chosen": 0.9620984196662903, | |
| "logits/rejected": 0.7840194702148438, | |
| "logps/chosen": -478.987060546875, | |
| "logps/rejected": -392.83966064453125, | |
| "loss": 0.129, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.015017243102192879, | |
| "rewards/margins": 0.05055858939886093, | |
| "rewards/rejected": -0.0355413481593132, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.759511406608255e-07, | |
| "logits/chosen": 0.8557583689689636, | |
| "logits/rejected": 0.6884294748306274, | |
| "logps/chosen": -492.0303649902344, | |
| "logps/rejected": -396.324462890625, | |
| "loss": 0.1418, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": 0.004057017620652914, | |
| "rewards/margins": 0.02802330255508423, | |
| "rewards/rejected": -0.023966282606124878, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.260731586586983e-07, | |
| "logits/chosen": 0.9453495144844055, | |
| "logits/rejected": 0.8003050088882446, | |
| "logps/chosen": -483.55792236328125, | |
| "logps/rejected": -375.1344909667969, | |
| "loss": 0.1258, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.0010669174371287227, | |
| "rewards/margins": 0.046425946056842804, | |
| "rewards/rejected": -0.04535902291536331, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.775784314464717e-07, | |
| "logits/chosen": 0.8808594942092896, | |
| "logits/rejected": 0.7378624081611633, | |
| "logps/chosen": -463.3553771972656, | |
| "logps/rejected": -356.5357666015625, | |
| "loss": 0.1174, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": 0.007392757572233677, | |
| "rewards/margins": 0.05343680828809738, | |
| "rewards/rejected": -0.046044059097766876, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.305047737536707e-07, | |
| "logits/chosen": 0.905502200126648, | |
| "logits/rejected": 0.7184777855873108, | |
| "logps/chosen": -479.0455627441406, | |
| "logps/rejected": -375.6910095214844, | |
| "loss": 0.1193, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.006599182728677988, | |
| "rewards/margins": 0.0473155751824379, | |
| "rewards/rejected": -0.040716394782066345, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.848888922025553e-07, | |
| "logits/chosen": 0.9522923231124878, | |
| "logits/rejected": 0.8169105648994446, | |
| "logps/chosen": -464.9600524902344, | |
| "logps/rejected": -401.93902587890625, | |
| "loss": 0.1295, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": 0.014724774286150932, | |
| "rewards/margins": 0.05533941462635994, | |
| "rewards/rejected": -0.04061463475227356, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.407663566854008e-07, | |
| "logits/chosen": 0.9025578498840332, | |
| "logits/rejected": 0.775783360004425, | |
| "logps/chosen": -468.3810119628906, | |
| "logps/rejected": -405.38909912109375, | |
| "loss": 0.1254, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.008423840627074242, | |
| "rewards/margins": 0.041560642421245575, | |
| "rewards/rejected": -0.03313680365681648, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.981715726281666e-07, | |
| "logits/chosen": 0.881443977355957, | |
| "logits/rejected": 0.7977533340454102, | |
| "logps/chosen": -461.211181640625, | |
| "logps/rejected": -386.59503173828125, | |
| "loss": 0.1258, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.008033149875700474, | |
| "rewards/margins": 0.04333629831671715, | |
| "rewards/rejected": -0.03530315309762955, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.5713775416217884e-07, | |
| "logits/chosen": 0.9581049084663391, | |
| "logits/rejected": 0.8421887159347534, | |
| "logps/chosen": -473.08892822265625, | |
| "logps/rejected": -391.439208984375, | |
| "loss": 0.1228, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.005411247257143259, | |
| "rewards/margins": 0.045824769884347916, | |
| "rewards/rejected": -0.040413517504930496, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.1769689822475147e-07, | |
| "logits/chosen": 0.9711424112319946, | |
| "logits/rejected": 0.8476129770278931, | |
| "logps/chosen": -471.318359375, | |
| "logps/rejected": -386.77899169921875, | |
| "loss": 0.1284, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.004041860345751047, | |
| "rewards/margins": 0.04613048955798149, | |
| "rewards/rejected": -0.04208862781524658, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.798797596089351e-07, | |
| "logits/chosen": 0.9549845457077026, | |
| "logits/rejected": 0.7986112833023071, | |
| "logps/chosen": -479.4010314941406, | |
| "logps/rejected": -408.44085693359375, | |
| "loss": 0.13, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.021130617707967758, | |
| "rewards/margins": 0.054322708398103714, | |
| "rewards/rejected": -0.033192090690135956, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.4371582698185636e-07, | |
| "logits/chosen": 0.8994715809822083, | |
| "logits/rejected": 0.7577309608459473, | |
| "logps/chosen": -479.24200439453125, | |
| "logps/rejected": -382.0733947753906, | |
| "loss": 0.1226, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": 0.013279114849865437, | |
| "rewards/margins": 0.04814552143216133, | |
| "rewards/rejected": -0.03486640378832817, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 3.092332998903416e-07, | |
| "logits/chosen": 0.837253749370575, | |
| "logits/rejected": 0.7374977469444275, | |
| "logps/chosen": -454.10382080078125, | |
| "logps/rejected": -389.86248779296875, | |
| "loss": 0.1295, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": 0.0033629995305091143, | |
| "rewards/margins": 0.03582575172185898, | |
| "rewards/rejected": -0.03246275335550308, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.764590667717562e-07, | |
| "logits/chosen": 0.8877947926521301, | |
| "logits/rejected": 0.7853025794029236, | |
| "logps/chosen": -491.3389587402344, | |
| "logps/rejected": -418.23333740234375, | |
| "loss": 0.09, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": 0.0025968037080019712, | |
| "rewards/margins": 0.060978036373853683, | |
| "rewards/rejected": -0.058381229639053345, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.454186839872158e-07, | |
| "logits/chosen": 0.9278301000595093, | |
| "logits/rejected": 0.7884716391563416, | |
| "logps/chosen": -478.32537841796875, | |
| "logps/rejected": -417.85992431640625, | |
| "loss": 0.1139, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": 0.015951160341501236, | |
| "rewards/margins": 0.06339136511087418, | |
| "rewards/rejected": -0.04744020476937294, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.1613635589349756e-07, | |
| "logits/chosen": 0.9455703496932983, | |
| "logits/rejected": 0.7855480909347534, | |
| "logps/chosen": -479.3771057128906, | |
| "logps/rejected": -387.804931640625, | |
| "loss": 0.1409, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": 0.005107102915644646, | |
| "rewards/margins": 0.043499238789081573, | |
| "rewards/rejected": -0.03839213401079178, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.8863491596921745e-07, | |
| "logits/chosen": 0.9270991086959839, | |
| "logits/rejected": 0.8039152026176453, | |
| "logps/chosen": -476.14898681640625, | |
| "logps/rejected": -371.51458740234375, | |
| "loss": 0.1168, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.011659547686576843, | |
| "rewards/margins": 0.05175407603383064, | |
| "rewards/rejected": -0.0400945320725441, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.629358090099639e-07, | |
| "logits/chosen": 0.9182494878768921, | |
| "logits/rejected": 0.8193320035934448, | |
| "logps/chosen": -479.65570068359375, | |
| "logps/rejected": -407.64569091796875, | |
| "loss": 0.114, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": 0.01620074175298214, | |
| "rewards/margins": 0.046776242554187775, | |
| "rewards/rejected": -0.030575498938560486, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3905907440629752e-07, | |
| "logits/chosen": 0.8919647932052612, | |
| "logits/rejected": 0.8270837068557739, | |
| "logps/chosen": -475.1752014160156, | |
| "logps/rejected": -405.669921875, | |
| "loss": 0.1212, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": 0.00893275998532772, | |
| "rewards/margins": 0.04200071096420288, | |
| "rewards/rejected": -0.03306794911623001, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.1702333051763271e-07, | |
| "logits/chosen": 0.8223358392715454, | |
| "logits/rejected": 0.6920078992843628, | |
| "logps/chosen": -467.2176818847656, | |
| "logps/rejected": -372.9118347167969, | |
| "loss": 0.1354, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": 0.007803215179592371, | |
| "rewards/margins": 0.05598113685846329, | |
| "rewards/rejected": -0.04817792773246765, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 9.684576015420277e-08, | |
| "logits/chosen": 0.9437923431396484, | |
| "logits/rejected": 0.8367222547531128, | |
| "logps/chosen": -466.2740783691406, | |
| "logps/rejected": -400.8631286621094, | |
| "loss": 0.1299, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": 0.01643536426126957, | |
| "rewards/margins": 0.049150340259075165, | |
| "rewards/rejected": -0.03271497040987015, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.854209717842231e-08, | |
| "logits/chosen": 0.8424699902534485, | |
| "logits/rejected": 0.7011204361915588, | |
| "logps/chosen": -468.2311096191406, | |
| "logps/rejected": -381.13287353515625, | |
| "loss": 0.1157, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": 0.016398664563894272, | |
| "rewards/margins": 0.06015818193554878, | |
| "rewards/rejected": -0.04375952482223511, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 6.212661423609184e-08, | |
| "logits/chosen": 0.8363115191459656, | |
| "logits/rejected": 0.783504843711853, | |
| "logps/chosen": -489.8089294433594, | |
| "logps/rejected": -389.686767578125, | |
| "loss": 0.1083, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": 0.014369189739227295, | |
| "rewards/margins": 0.05540204793214798, | |
| "rewards/rejected": -0.041032858192920685, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.761211162702117e-08, | |
| "logits/chosen": 0.9592778086662292, | |
| "logits/rejected": 0.8283406496047974, | |
| "logps/chosen": -455.68017578125, | |
| "logps/rejected": -386.79205322265625, | |
| "loss": 0.1205, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": 0.014277097769081593, | |
| "rewards/margins": 0.05047346279025078, | |
| "rewards/rejected": -0.03619636222720146, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.5009907323737826e-08, | |
| "logits/chosen": 0.9002725481987, | |
| "logits/rejected": 0.7559747099876404, | |
| "logps/chosen": -469.6654357910156, | |
| "logps/rejected": -385.0262756347656, | |
| "loss": 0.1152, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": 0.006471751723438501, | |
| "rewards/margins": 0.04549337550997734, | |
| "rewards/rejected": -0.039021626114845276, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.4329828146074096e-08, | |
| "logits/chosen": 0.9445624351501465, | |
| "logits/rejected": 0.7747251391410828, | |
| "logps/chosen": -494.6180725097656, | |
| "logps/rejected": -381.2744445800781, | |
| "loss": 0.1219, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": 0.004516331944614649, | |
| "rewards/margins": 0.04709627851843834, | |
| "rewards/rejected": -0.04257994890213013, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.5580202098509078e-08, | |
| "logits/chosen": 0.9075485467910767, | |
| "logits/rejected": 0.791230320930481, | |
| "logps/chosen": -483.7403259277344, | |
| "logps/rejected": -374.10101318359375, | |
| "loss": 0.1168, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": 0.005151915363967419, | |
| "rewards/margins": 0.056254614144563675, | |
| "rewards/rejected": -0.05110269784927368, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.767851876239075e-09, | |
| "logits/chosen": 0.9745057225227356, | |
| "logits/rejected": 0.8449977040290833, | |
| "logps/chosen": -461.8253479003906, | |
| "logps/rejected": -399.06011962890625, | |
| "loss": 0.1325, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": 0.013572081923484802, | |
| "rewards/margins": 0.054333366453647614, | |
| "rewards/rejected": -0.04076128453016281, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.8980895450474455e-09, | |
| "logits/chosen": 0.8922117948532104, | |
| "logits/rejected": 0.6962934732437134, | |
| "logps/chosen": -454.9315490722656, | |
| "logps/rejected": -386.56591796875, | |
| "loss": 0.1104, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": 0.00724564865231514, | |
| "rewards/margins": 0.05632048100233078, | |
| "rewards/rejected": -0.04907483607530594, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.747123991141193e-10, | |
| "logits/chosen": 0.9504325985908508, | |
| "logits/rejected": 0.713564932346344, | |
| "logps/chosen": -479.7052307128906, | |
| "logps/rejected": -373.42852783203125, | |
| "loss": 0.1271, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": 0.010545340366661549, | |
| "rewards/margins": 0.05146101862192154, | |
| "rewards/rejected": -0.04091567546129227, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0, | |
| "logits/chosen": 0.8859880566596985, | |
| "logits/rejected": 0.726716160774231, | |
| "logps/chosen": -467.9039001464844, | |
| "logps/rejected": -371.0747375488281, | |
| "loss": 0.1236, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": 0.0073653412982821465, | |
| "rewards/margins": 0.05689455196261406, | |
| "rewards/rejected": -0.04952920600771904, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 1250, | |
| "total_flos": 0.0, | |
| "train_loss": 0.12703038393855096, | |
| "train_runtime": 9089.2921, | |
| "train_samples_per_second": 2.2, | |
| "train_steps_per_second": 0.138 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |