| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9973828840617638, | |
| "eval_steps": 10000, | |
| "global_step": 954, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.0416666666666667e-07, | |
| "logits/chosen": 0.17704486846923828, | |
| "logits/rejected": 0.25409135222435, | |
| "logps/chosen": -354.4068603515625, | |
| "logps/rejected": -305.2366638183594, | |
| "loss": 0.1821, | |
| "rewards/accuracies": 0.4312500059604645, | |
| "rewards/chosen": -6.60312725813128e-05, | |
| "rewards/margins": 0.00012125837383791804, | |
| "rewards/rejected": -0.00018728969735093415, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "logits/chosen": 0.07091161608695984, | |
| "logits/rejected": 0.1985362321138382, | |
| "logps/chosen": -316.65069580078125, | |
| "logps/rejected": -276.1200866699219, | |
| "loss": 0.182, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": 0.0008458361262455583, | |
| "rewards/margins": 0.0016920112539082766, | |
| "rewards/rejected": -0.0008461751276627183, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "logits/chosen": 0.17787829041481018, | |
| "logits/rejected": 0.2488478720188141, | |
| "logps/chosen": -294.9706115722656, | |
| "logps/rejected": -298.59521484375, | |
| "loss": 0.1822, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -3.700423985719681e-05, | |
| "rewards/margins": 0.0029355171136558056, | |
| "rewards/rejected": -0.0029725211206823587, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "logits/chosen": 0.09609868377447128, | |
| "logits/rejected": 0.21795693039894104, | |
| "logps/chosen": -347.44097900390625, | |
| "logps/rejected": -320.9972839355469, | |
| "loss": 0.1877, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": 0.0013125470140948892, | |
| "rewards/margins": 0.00661453977227211, | |
| "rewards/rejected": -0.005301993805915117, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 5.208333333333334e-07, | |
| "logits/chosen": 0.1497882902622223, | |
| "logits/rejected": 0.240590900182724, | |
| "logps/chosen": -311.1229553222656, | |
| "logps/rejected": -286.51702880859375, | |
| "loss": 0.1814, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": -0.005703258328139782, | |
| "rewards/margins": 0.022644545882940292, | |
| "rewards/rejected": -0.02834780514240265, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 6.249999999999999e-07, | |
| "logits/chosen": 0.13869214057922363, | |
| "logits/rejected": 0.28307411074638367, | |
| "logps/chosen": -295.9754638671875, | |
| "logps/rejected": -281.43798828125, | |
| "loss": 0.1766, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -0.03096725046634674, | |
| "rewards/margins": 0.028959080576896667, | |
| "rewards/rejected": -0.059926338493824005, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 7.291666666666666e-07, | |
| "logits/chosen": 0.18460798263549805, | |
| "logits/rejected": 0.2718513607978821, | |
| "logps/chosen": -335.46148681640625, | |
| "logps/rejected": -330.33404541015625, | |
| "loss": 0.174, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.057377688586711884, | |
| "rewards/margins": 0.05648452043533325, | |
| "rewards/rejected": -0.11386220157146454, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.333333333333333e-07, | |
| "logits/chosen": 0.29816848039627075, | |
| "logits/rejected": 0.4011983871459961, | |
| "logps/chosen": -330.4580383300781, | |
| "logps/rejected": -311.96490478515625, | |
| "loss": 0.159, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -0.11794394254684448, | |
| "rewards/margins": 0.13102997839450836, | |
| "rewards/rejected": -0.24897389113903046, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.374999999999999e-07, | |
| "logits/chosen": 0.2283201515674591, | |
| "logits/rejected": 0.37335914373397827, | |
| "logps/chosen": -358.6737365722656, | |
| "logps/rejected": -304.0804138183594, | |
| "loss": 0.1421, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -0.21732211112976074, | |
| "rewards/margins": 0.15273679792881012, | |
| "rewards/rejected": -0.37005892395973206, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.999463737538052e-07, | |
| "logits/chosen": 0.2938156723976135, | |
| "logits/rejected": 0.46553492546081543, | |
| "logps/chosen": -361.78338623046875, | |
| "logps/rejected": -343.25750732421875, | |
| "loss": 0.1217, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": -0.27221935987472534, | |
| "rewards/margins": 0.23653486371040344, | |
| "rewards/rejected": -0.5087541937828064, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.993432105822034e-07, | |
| "logits/chosen": 0.31155580282211304, | |
| "logits/rejected": 0.3508353531360626, | |
| "logps/chosen": -353.184814453125, | |
| "logps/rejected": -366.32720947265625, | |
| "loss": 0.106, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -0.40565404295921326, | |
| "rewards/margins": 0.2631165683269501, | |
| "rewards/rejected": -0.6687706708908081, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.980706626858607e-07, | |
| "logits/chosen": 0.26659709215164185, | |
| "logits/rejected": 0.3288796842098236, | |
| "logps/chosen": -374.50274658203125, | |
| "logps/rejected": -403.8424377441406, | |
| "loss": 0.0951, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.5394914150238037, | |
| "rewards/margins": 0.28696924448013306, | |
| "rewards/rejected": -0.8264607191085815, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.961304359538434e-07, | |
| "logits/chosen": 0.1616436094045639, | |
| "logits/rejected": 0.2970871031284332, | |
| "logps/chosen": -396.555419921875, | |
| "logps/rejected": -362.3848876953125, | |
| "loss": 0.0934, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.5805934071540833, | |
| "rewards/margins": 0.19475166499614716, | |
| "rewards/rejected": -0.775344967842102, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.935251313189563e-07, | |
| "logits/chosen": 0.1485656201839447, | |
| "logits/rejected": 0.2714545428752899, | |
| "logps/chosen": -384.0659484863281, | |
| "logps/rejected": -346.6048278808594, | |
| "loss": 0.0933, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.5202253460884094, | |
| "rewards/margins": 0.24675369262695312, | |
| "rewards/rejected": -0.766978919506073, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.902582412711118e-07, | |
| "logits/chosen": 0.12988325953483582, | |
| "logits/rejected": 0.1523539423942566, | |
| "logps/chosen": -379.16839599609375, | |
| "logps/rejected": -395.9466552734375, | |
| "loss": 0.1019, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.4386775493621826, | |
| "rewards/margins": 0.37129276990890503, | |
| "rewards/rejected": -0.8099702596664429, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.86334145175542e-07, | |
| "logits/chosen": 0.06655962765216827, | |
| "logits/rejected": 0.09024105966091156, | |
| "logps/chosen": -341.7105407714844, | |
| "logps/rejected": -360.19805908203125, | |
| "loss": 0.0937, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.3964901566505432, | |
| "rewards/margins": 0.3985019028186798, | |
| "rewards/rejected": -0.7949920892715454, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.817581034021272e-07, | |
| "logits/chosen": 0.16973164677619934, | |
| "logits/rejected": 0.21836213767528534, | |
| "logps/chosen": -398.22369384765625, | |
| "logps/rejected": -417.8206481933594, | |
| "loss": 0.081, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.6185532808303833, | |
| "rewards/margins": 0.4811604917049408, | |
| "rewards/rejected": -1.0997138023376465, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.765362502737097e-07, | |
| "logits/chosen": 0.09212584793567657, | |
| "logits/rejected": 0.23974208533763885, | |
| "logps/chosen": -388.64910888671875, | |
| "logps/rejected": -411.5782775878906, | |
| "loss": 0.0713, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -0.6261709928512573, | |
| "rewards/margins": 0.4908596873283386, | |
| "rewards/rejected": -1.1170307397842407, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.706755858428485e-07, | |
| "logits/chosen": 0.1811675727367401, | |
| "logits/rejected": 0.27236208319664, | |
| "logps/chosen": -419.11376953125, | |
| "logps/rejected": -437.33843994140625, | |
| "loss": 0.0681, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.8445426225662231, | |
| "rewards/margins": 0.4015916883945465, | |
| "rewards/rejected": -1.2461342811584473, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 9.641839665080363e-07, | |
| "logits/chosen": 0.14256766438484192, | |
| "logits/rejected": 0.2711044251918793, | |
| "logps/chosen": -414.55975341796875, | |
| "logps/rejected": -416.9037170410156, | |
| "loss": 0.0675, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.7406997680664062, | |
| "rewards/margins": 0.48706990480422974, | |
| "rewards/rejected": -1.2277696132659912, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.570700944819582e-07, | |
| "logits/chosen": 0.23208096623420715, | |
| "logits/rejected": 0.35697174072265625, | |
| "logps/chosen": -382.19970703125, | |
| "logps/rejected": -386.50701904296875, | |
| "loss": 0.0708, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.6804240942001343, | |
| "rewards/margins": 0.48590850830078125, | |
| "rewards/rejected": -1.166332721710205, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 9.493435061259129e-07, | |
| "logits/chosen": 0.13639363646507263, | |
| "logits/rejected": 0.23731064796447754, | |
| "logps/chosen": -382.42022705078125, | |
| "logps/rejected": -369.6554870605469, | |
| "loss": 0.0763, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.6574115753173828, | |
| "rewards/margins": 0.40243881940841675, | |
| "rewards/rejected": -1.0598504543304443, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.4101455916603e-07, | |
| "logits/chosen": 0.1799091249704361, | |
| "logits/rejected": 0.2304597645998001, | |
| "logps/chosen": -416.672607421875, | |
| "logps/rejected": -420.39862060546875, | |
| "loss": 0.0668, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -0.9061130285263062, | |
| "rewards/margins": 0.46666598320007324, | |
| "rewards/rejected": -1.3727790117263794, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.320944188084241e-07, | |
| "logits/chosen": 0.08318189531564713, | |
| "logits/rejected": 0.13486048579216003, | |
| "logps/chosen": -408.77545166015625, | |
| "logps/rejected": -427.9566345214844, | |
| "loss": 0.0639, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -0.968237042427063, | |
| "rewards/margins": 0.2922549843788147, | |
| "rewards/rejected": -1.260491967201233, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.225950427718974e-07, | |
| "logits/chosen": 0.051157813519239426, | |
| "logits/rejected": 0.1319509893655777, | |
| "logps/chosen": -385.2474670410156, | |
| "logps/rejected": -402.11126708984375, | |
| "loss": 0.0631, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": -0.7319932579994202, | |
| "rewards/margins": 0.468679815530777, | |
| "rewards/rejected": -1.2006731033325195, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.125291652582547e-07, | |
| "logits/chosen": 0.013853952288627625, | |
| "logits/rejected": 0.10071275383234024, | |
| "logps/chosen": -445.53607177734375, | |
| "logps/rejected": -434.2711486816406, | |
| "loss": 0.0641, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.9089228510856628, | |
| "rewards/margins": 0.4331666827201843, | |
| "rewards/rejected": -1.3420894145965576, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 9.019102798817195e-07, | |
| "logits/chosen": 0.1297096163034439, | |
| "logits/rejected": 0.1613592505455017, | |
| "logps/chosen": -403.47393798828125, | |
| "logps/rejected": -446.1951599121094, | |
| "loss": 0.0685, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": -0.7434005737304688, | |
| "rewards/margins": 0.6140644550323486, | |
| "rewards/rejected": -1.357465147972107, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.90752621580335e-07, | |
| "logits/chosen": 0.16231071949005127, | |
| "logits/rejected": 0.1873283088207245, | |
| "logps/chosen": -362.4006652832031, | |
| "logps/rejected": -398.279296875, | |
| "loss": 0.0751, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -0.6591774225234985, | |
| "rewards/margins": 0.41294485330581665, | |
| "rewards/rejected": -1.07212233543396, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 8.79071147533597e-07, | |
| "logits/chosen": 0.14204099774360657, | |
| "logits/rejected": 0.20997166633605957, | |
| "logps/chosen": -424.5856018066406, | |
| "logps/rejected": -456.9698181152344, | |
| "loss": 0.0642, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.7815448045730591, | |
| "rewards/margins": 0.5602203011512756, | |
| "rewards/rejected": -1.34176504611969, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 8.668815171119019e-07, | |
| "logits/chosen": 0.2026984989643097, | |
| "logits/rejected": 0.23374077677726746, | |
| "logps/chosen": -380.8060607910156, | |
| "logps/rejected": -468.7802734375, | |
| "loss": 0.0554, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.8905105590820312, | |
| "rewards/margins": 0.5638677477836609, | |
| "rewards/rejected": -1.454378366470337, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 8.54200070884685e-07, | |
| "logits/chosen": 0.23336808383464813, | |
| "logits/rejected": 0.25176650285720825, | |
| "logps/chosen": -385.24676513671875, | |
| "logps/rejected": -462.87322998046875, | |
| "loss": 0.0565, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -0.8951492309570312, | |
| "rewards/margins": 0.6165014505386353, | |
| "rewards/rejected": -1.5116506814956665, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 8.410438087153911e-07, | |
| "logits/chosen": 0.22913236916065216, | |
| "logits/rejected": 0.3360585570335388, | |
| "logps/chosen": -383.767578125, | |
| "logps/rejected": -424.25067138671875, | |
| "loss": 0.0641, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.6813658475875854, | |
| "rewards/margins": 0.6591276526451111, | |
| "rewards/rejected": -1.3404934406280518, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 8.274303669726426e-07, | |
| "logits/chosen": 0.22990348935127258, | |
| "logits/rejected": 0.3006184697151184, | |
| "logps/chosen": -366.43499755859375, | |
| "logps/rejected": -444.06536865234375, | |
| "loss": 0.0636, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.6766657829284668, | |
| "rewards/margins": 0.6564770936965942, | |
| "rewards/rejected": -1.333142876625061, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 8.133779948881513e-07, | |
| "logits/chosen": 0.22257550060749054, | |
| "logits/rejected": 0.3241097033023834, | |
| "logps/chosen": -360.141845703125, | |
| "logps/rejected": -405.85711669921875, | |
| "loss": 0.0662, | |
| "rewards/accuracies": 0.78125, | |
| "rewards/chosen": -0.7344536781311035, | |
| "rewards/margins": 0.7157880067825317, | |
| "rewards/rejected": -1.4502416849136353, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 7.989055300930704e-07, | |
| "logits/chosen": 0.1499968320131302, | |
| "logits/rejected": 0.15372925996780396, | |
| "logps/chosen": -388.67559814453125, | |
| "logps/rejected": -462.0445251464844, | |
| "loss": 0.0644, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.8717344403266907, | |
| "rewards/margins": 0.6429644227027893, | |
| "rewards/rejected": -1.51469886302948, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.840323733655778e-07, | |
| "logits/chosen": 0.08885981142520905, | |
| "logits/rejected": 0.19541098177433014, | |
| "logps/chosen": -407.87286376953125, | |
| "logps/rejected": -420.4515686035156, | |
| "loss": 0.0583, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -0.797155499458313, | |
| "rewards/margins": 0.5855330228805542, | |
| "rewards/rejected": -1.3826884031295776, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.687784626235447e-07, | |
| "logits/chosen": 0.05912008136510849, | |
| "logits/rejected": 0.17702099680900574, | |
| "logps/chosen": -428.82354736328125, | |
| "logps/rejected": -466.0895080566406, | |
| "loss": 0.0599, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": -0.803920567035675, | |
| "rewards/margins": 0.7507921457290649, | |
| "rewards/rejected": -1.5547125339508057, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.531642461971514e-07, | |
| "logits/chosen": 0.11388075351715088, | |
| "logits/rejected": 0.1931450068950653, | |
| "logps/chosen": -388.9282531738281, | |
| "logps/rejected": -427.1614685058594, | |
| "loss": 0.0578, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": -0.9585503339767456, | |
| "rewards/margins": 0.5912213325500488, | |
| "rewards/rejected": -1.5497716665267944, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 7.372106554172801e-07, | |
| "logits/chosen": -0.049389470368623734, | |
| "logits/rejected": 0.10218650102615356, | |
| "logps/chosen": -443.7737731933594, | |
| "logps/rejected": -484.5735778808594, | |
| "loss": 0.0446, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -1.0208237171173096, | |
| "rewards/margins": 0.8150562047958374, | |
| "rewards/rejected": -1.835879921913147, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 7.209390765564318e-07, | |
| "logits/chosen": 0.07526848465204239, | |
| "logits/rejected": 0.1457681804895401, | |
| "logps/chosen": -430.77130126953125, | |
| "logps/rejected": -478.53118896484375, | |
| "loss": 0.0488, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -1.137662410736084, | |
| "rewards/margins": 0.6997725963592529, | |
| "rewards/rejected": -1.837435007095337, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.043713221597773e-07, | |
| "logits/chosen": -0.014962440356612206, | |
| "logits/rejected": 0.049673158675432205, | |
| "logps/chosen": -394.35980224609375, | |
| "logps/rejected": -455.79168701171875, | |
| "loss": 0.0469, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.0516221523284912, | |
| "rewards/margins": 0.6002627015113831, | |
| "rewards/rejected": -1.65188467502594, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.875296018047809e-07, | |
| "logits/chosen": 0.1113734096288681, | |
| "logits/rejected": 0.17297616600990295, | |
| "logps/chosen": -371.1769104003906, | |
| "logps/rejected": -433.82763671875, | |
| "loss": 0.057, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -0.7784308791160583, | |
| "rewards/margins": 0.7032991647720337, | |
| "rewards/rejected": -1.4817302227020264, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 6.704364923285857e-07, | |
| "logits/chosen": 0.08021976053714752, | |
| "logits/rejected": 0.09611347317695618, | |
| "logps/chosen": -433.26898193359375, | |
| "logps/rejected": -482.2544860839844, | |
| "loss": 0.0623, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -0.9919212460517883, | |
| "rewards/margins": 0.5928072333335876, | |
| "rewards/rejected": -1.584728479385376, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 6.531149075630796e-07, | |
| "logits/chosen": 0.06492827087640762, | |
| "logits/rejected": 0.09372309595346451, | |
| "logps/chosen": -369.0657958984375, | |
| "logps/rejected": -427.1637268066406, | |
| "loss": 0.0602, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": -0.8450859785079956, | |
| "rewards/margins": 0.6487796902656555, | |
| "rewards/rejected": -1.4938656091690063, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 6.355880676182085e-07, | |
| "logits/chosen": 0.015085640363395214, | |
| "logits/rejected": 0.1697283238172531, | |
| "logps/chosen": -454.42071533203125, | |
| "logps/rejected": -461.6656799316406, | |
| "loss": 0.0537, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -1.0446925163269043, | |
| "rewards/margins": 0.7324589490890503, | |
| "rewards/rejected": -1.7771514654159546, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 6.178794677547137e-07, | |
| "logits/chosen": 0.052903078496456146, | |
| "logits/rejected": 0.21909013390541077, | |
| "logps/chosen": -389.771728515625, | |
| "logps/rejected": -432.63311767578125, | |
| "loss": 0.0475, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -0.918341338634491, | |
| "rewards/margins": 0.7504295706748962, | |
| "rewards/rejected": -1.6687707901000977, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 6.000128468880222e-07, | |
| "logits/chosen": 0.0020152360666543245, | |
| "logits/rejected": 0.10528425872325897, | |
| "logps/chosen": -439.73016357421875, | |
| "logps/rejected": -486.3055114746094, | |
| "loss": 0.0531, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -1.0058103799819946, | |
| "rewards/margins": 0.8824182748794556, | |
| "rewards/rejected": -1.8882286548614502, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5.820121557655108e-07, | |
| "logits/chosen": 0.03267590329051018, | |
| "logits/rejected": 0.10403893887996674, | |
| "logps/chosen": -426.3312072753906, | |
| "logps/rejected": -521.575439453125, | |
| "loss": 0.0497, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -0.897496223449707, | |
| "rewards/margins": 1.0473217964172363, | |
| "rewards/rejected": -1.9448179006576538, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.639015248598023e-07, | |
| "logits/chosen": -0.05066138505935669, | |
| "logits/rejected": 0.0016520231729373336, | |
| "logps/chosen": -459.2066955566406, | |
| "logps/rejected": -572.3805541992188, | |
| "loss": 0.0254, | |
| "rewards/accuracies": 0.768750011920929, | |
| "rewards/chosen": -1.404326319694519, | |
| "rewards/margins": 1.2682745456695557, | |
| "rewards/rejected": -2.6726012229919434, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 5.457052320211339e-07, | |
| "logits/chosen": 0.10663177818059921, | |
| "logits/rejected": 0.143524631857872, | |
| "logps/chosen": -454.5547790527344, | |
| "logps/rejected": -574.3235473632812, | |
| "loss": 0.0198, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -1.592284083366394, | |
| "rewards/margins": 1.2184875011444092, | |
| "rewards/rejected": -2.8107717037200928, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.274476699321637e-07, | |
| "logits/chosen": -0.019788045436143875, | |
| "logits/rejected": 0.12656378746032715, | |
| "logps/chosen": -488.24627685546875, | |
| "logps/rejected": -596.00537109375, | |
| "loss": 0.015, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.8213142156600952, | |
| "rewards/margins": 1.3653538227081299, | |
| "rewards/rejected": -3.1866683959960938, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.091533134088387e-07, | |
| "logits/chosen": -0.0814504474401474, | |
| "logits/rejected": 0.05524957925081253, | |
| "logps/chosen": -552.7730712890625, | |
| "logps/rejected": -634.5548095703125, | |
| "loss": 0.0147, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -2.0995850563049316, | |
| "rewards/margins": 1.1655638217926025, | |
| "rewards/rejected": -3.2651493549346924, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.908466865911614e-07, | |
| "logits/chosen": 0.03363295644521713, | |
| "logits/rejected": 0.043015364557504654, | |
| "logps/chosen": -468.89593505859375, | |
| "logps/rejected": -560.2864990234375, | |
| "loss": 0.0174, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -1.5512639284133911, | |
| "rewards/margins": 1.2513355016708374, | |
| "rewards/rejected": -2.8025994300842285, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 4.7255233006783624e-07, | |
| "logits/chosen": -0.03754299506545067, | |
| "logits/rejected": 0.08725563436746597, | |
| "logps/chosen": -456.68243408203125, | |
| "logps/rejected": -549.9105224609375, | |
| "loss": 0.0178, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -1.657478928565979, | |
| "rewards/margins": 1.0530353784561157, | |
| "rewards/rejected": -2.7105140686035156, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.5429476797886617e-07, | |
| "logits/chosen": 0.0340617299079895, | |
| "logits/rejected": 0.1264275759458542, | |
| "logps/chosen": -469.5687561035156, | |
| "logps/rejected": -592.4705810546875, | |
| "loss": 0.0185, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.408406138420105, | |
| "rewards/margins": 1.4667712450027466, | |
| "rewards/rejected": -2.8751769065856934, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.3609847514019763e-07, | |
| "logits/chosen": 0.0167356226593256, | |
| "logits/rejected": 0.032135289162397385, | |
| "logps/chosen": -480.41278076171875, | |
| "logps/rejected": -577.2174072265625, | |
| "loss": 0.0165, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -1.5578255653381348, | |
| "rewards/margins": 1.0947318077087402, | |
| "rewards/rejected": -2.652557611465454, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.179878442344892e-07, | |
| "logits/chosen": 0.10041844844818115, | |
| "logits/rejected": 0.16732005774974823, | |
| "logps/chosen": -453.9161071777344, | |
| "logps/rejected": -615.6796875, | |
| "loss": 0.0153, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -1.7070415019989014, | |
| "rewards/margins": 1.4755295515060425, | |
| "rewards/rejected": -3.1825711727142334, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.9998715311197783e-07, | |
| "logits/chosen": 0.1310591995716095, | |
| "logits/rejected": 0.20585906505584717, | |
| "logps/chosen": -493.8118591308594, | |
| "logps/rejected": -631.4963989257812, | |
| "loss": 0.015, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -1.7850983142852783, | |
| "rewards/margins": 1.443263292312622, | |
| "rewards/rejected": -3.228361129760742, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.821205322452863e-07, | |
| "logits/chosen": 0.22954685986042023, | |
| "logits/rejected": 0.2483092099428177, | |
| "logps/chosen": -473.4378967285156, | |
| "logps/rejected": -605.134033203125, | |
| "loss": 0.0149, | |
| "rewards/accuracies": 0.7562500238418579, | |
| "rewards/chosen": -1.700280785560608, | |
| "rewards/margins": 1.460669755935669, | |
| "rewards/rejected": -3.1609506607055664, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.6441193238179146e-07, | |
| "logits/chosen": 0.13607949018478394, | |
| "logits/rejected": 0.1680508852005005, | |
| "logps/chosen": -451.55340576171875, | |
| "logps/rejected": -627.7686157226562, | |
| "loss": 0.0147, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -1.6148862838745117, | |
| "rewards/margins": 1.678989052772522, | |
| "rewards/rejected": -3.2938759326934814, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.4688509243692034e-07, | |
| "logits/chosen": 0.04345204681158066, | |
| "logits/rejected": 0.13040025532245636, | |
| "logps/chosen": -461.54095458984375, | |
| "logps/rejected": -684.9581909179688, | |
| "loss": 0.0153, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": -1.6480602025985718, | |
| "rewards/margins": 1.6946277618408203, | |
| "rewards/rejected": -3.3426880836486816, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.295635076714144e-07, | |
| "logits/chosen": 0.18233785033226013, | |
| "logits/rejected": 0.19972297549247742, | |
| "logps/chosen": -408.9209899902344, | |
| "logps/rejected": -547.9658813476562, | |
| "loss": 0.0143, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -1.6356074810028076, | |
| "rewards/margins": 1.3703811168670654, | |
| "rewards/rejected": -3.005988597869873, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.12470398195219e-07, | |
| "logits/chosen": 0.15017299354076385, | |
| "logits/rejected": 0.07167269289493561, | |
| "logps/chosen": -474.58172607421875, | |
| "logps/rejected": -649.4796142578125, | |
| "loss": 0.0129, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -1.6831333637237549, | |
| "rewards/margins": 1.4837870597839355, | |
| "rewards/rejected": -3.1669201850891113, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.956286778402226e-07, | |
| "logits/chosen": 0.03866753727197647, | |
| "logits/rejected": 0.20129835605621338, | |
| "logps/chosen": -546.3468017578125, | |
| "logps/rejected": -608.462646484375, | |
| "loss": 0.0126, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -1.7091865539550781, | |
| "rewards/margins": 1.3178246021270752, | |
| "rewards/rejected": -3.0270111560821533, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.7906092344356826e-07, | |
| "logits/chosen": 0.2127591073513031, | |
| "logits/rejected": 0.24179625511169434, | |
| "logps/chosen": -462.47412109375, | |
| "logps/rejected": -581.084228515625, | |
| "loss": 0.014, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": -1.751960039138794, | |
| "rewards/margins": 1.4448457956314087, | |
| "rewards/rejected": -3.196805953979492, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.6278934458271996e-07, | |
| "logits/chosen": 0.09269841015338898, | |
| "logits/rejected": 0.2964209318161011, | |
| "logps/chosen": -479.434326171875, | |
| "logps/rejected": -605.9524536132812, | |
| "loss": 0.0123, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -1.8022867441177368, | |
| "rewards/margins": 1.3753817081451416, | |
| "rewards/rejected": -3.177668333053589, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.468357538028487e-07, | |
| "logits/chosen": 0.16141146421432495, | |
| "logits/rejected": 0.18542757630348206, | |
| "logps/chosen": -487.90277099609375, | |
| "logps/rejected": -652.5034790039062, | |
| "loss": 0.0107, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -1.9332258701324463, | |
| "rewards/margins": 1.736053705215454, | |
| "rewards/rejected": -3.6692795753479004, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2.312215373764551e-07, | |
| "logits/chosen": 0.07799498736858368, | |
| "logits/rejected": 0.17718131840229034, | |
| "logps/chosen": -603.2567138671875, | |
| "logps/rejected": -699.2156372070312, | |
| "loss": 0.0101, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -2.1482930183410645, | |
| "rewards/margins": 1.3787685632705688, | |
| "rewards/rejected": -3.5270614624023438, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 2.1596762663442213e-07, | |
| "logits/chosen": 0.2014874666929245, | |
| "logits/rejected": 0.3246391713619232, | |
| "logps/chosen": -489.08349609375, | |
| "logps/rejected": -607.5847778320312, | |
| "loss": 0.0096, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -2.083740711212158, | |
| "rewards/margins": 1.446257472038269, | |
| "rewards/rejected": -3.5299980640411377, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 2.0109446990692963e-07, | |
| "logits/chosen": 0.09734896570444107, | |
| "logits/rejected": 0.16283641755580902, | |
| "logps/chosen": -540.1688232421875, | |
| "logps/rejected": -701.462890625, | |
| "loss": 0.0094, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -2.07643985748291, | |
| "rewards/margins": 1.7090556621551514, | |
| "rewards/rejected": -3.7854957580566406, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.8662200511184872e-07, | |
| "logits/chosen": 0.07912759482860565, | |
| "logits/rejected": 0.19963078200817108, | |
| "logps/chosen": -491.30426025390625, | |
| "logps/rejected": -630.0563354492188, | |
| "loss": 0.0099, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.9977525472640991, | |
| "rewards/margins": 1.5802443027496338, | |
| "rewards/rejected": -3.5779967308044434, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.725696330273575e-07, | |
| "logits/chosen": 0.14783975481987, | |
| "logits/rejected": 0.27563345432281494, | |
| "logps/chosen": -530.8796997070312, | |
| "logps/rejected": -640.3440551757812, | |
| "loss": 0.0107, | |
| "rewards/accuracies": 0.7562500238418579, | |
| "rewards/chosen": -2.036653518676758, | |
| "rewards/margins": 1.323557734489441, | |
| "rewards/rejected": -3.3602116107940674, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.589561912846089e-07, | |
| "logits/chosen": 0.16717246174812317, | |
| "logits/rejected": 0.2920343279838562, | |
| "logps/chosen": -499.3802795410156, | |
| "logps/rejected": -612.64892578125, | |
| "loss": 0.012, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -2.0618550777435303, | |
| "rewards/margins": 1.435462236404419, | |
| "rewards/rejected": -3.4973175525665283, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.4579992911531496e-07, | |
| "logits/chosen": 0.1249130517244339, | |
| "logits/rejected": 0.23616066575050354, | |
| "logps/chosen": -575.0750732421875, | |
| "logps/rejected": -649.9669189453125, | |
| "loss": 0.0106, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -2.2815146446228027, | |
| "rewards/margins": 1.226216197013855, | |
| "rewards/rejected": -3.5077309608459473, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.3311848288809813e-07, | |
| "logits/chosen": 0.21837782859802246, | |
| "logits/rejected": 0.31546956300735474, | |
| "logps/chosen": -510.7059020996094, | |
| "logps/rejected": -609.2933959960938, | |
| "loss": 0.0119, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -1.790372610092163, | |
| "rewards/margins": 1.2426694631576538, | |
| "rewards/rejected": -3.0330421924591064, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.209288524664029e-07, | |
| "logits/chosen": 0.14562873542308807, | |
| "logits/rejected": 0.3084864318370819, | |
| "logps/chosen": -622.6912841796875, | |
| "logps/rejected": -749.8731689453125, | |
| "loss": 0.0131, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -2.2252538204193115, | |
| "rewards/margins": 1.5818650722503662, | |
| "rewards/rejected": -3.8071188926696777, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 1.0924737841966497e-07, | |
| "logits/chosen": 0.1799144446849823, | |
| "logits/rejected": 0.354133278131485, | |
| "logps/chosen": -585.0472412109375, | |
| "logps/rejected": -712.3133544921875, | |
| "loss": 0.0107, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -2.1570990085601807, | |
| "rewards/margins": 1.6586040258407593, | |
| "rewards/rejected": -3.8157036304473877, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.808972011828054e-08, | |
| "logits/chosen": 0.20896565914154053, | |
| "logits/rejected": 0.1832619458436966, | |
| "logps/chosen": -474.9366149902344, | |
| "logps/rejected": -665.3892822265625, | |
| "loss": 0.0099, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.9308887720108032, | |
| "rewards/margins": 1.5281493663787842, | |
| "rewards/rejected": -3.459038257598877, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 8.747083474174527e-08, | |
| "logits/chosen": 0.25221484899520874, | |
| "logits/rejected": 0.3025228679180145, | |
| "logps/chosen": -486.76678466796875, | |
| "logps/rejected": -610.9810791015625, | |
| "loss": 0.01, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -1.9139289855957031, | |
| "rewards/margins": 1.4173685312271118, | |
| "rewards/rejected": -3.3312973976135254, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 7.740495722810269e-08, | |
| "logits/chosen": 0.12703558802604675, | |
| "logits/rejected": 0.25433093309402466, | |
| "logps/chosen": -528.8013916015625, | |
| "logps/rejected": -645.4374389648438, | |
| "loss": 0.01, | |
| "rewards/accuracies": 0.768750011920929, | |
| "rewards/chosen": -2.129984140396118, | |
| "rewards/margins": 1.322923183441162, | |
| "rewards/rejected": -3.452907085418701, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 6.790558119157597e-08, | |
| "logits/chosen": 0.1941952407360077, | |
| "logits/rejected": 0.36538344621658325, | |
| "logps/chosen": -536.0458374023438, | |
| "logps/rejected": -630.6697387695312, | |
| "loss": 0.0111, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -1.9618316888809204, | |
| "rewards/margins": 1.3840124607086182, | |
| "rewards/rejected": -3.34584379196167, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 5.898544083397e-08, | |
| "logits/chosen": 0.1936766654253006, | |
| "logits/rejected": 0.22626741230487823, | |
| "logps/chosen": -482.18902587890625, | |
| "logps/rejected": -640.9258422851562, | |
| "loss": 0.0113, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.822951078414917, | |
| "rewards/margins": 1.679091215133667, | |
| "rewards/rejected": -3.502042055130005, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 5.065649387408705e-08, | |
| "logits/chosen": 0.16037659347057343, | |
| "logits/rejected": 0.23867423832416534, | |
| "logps/chosen": -536.796630859375, | |
| "logps/rejected": -645.6795654296875, | |
| "loss": 0.0119, | |
| "rewards/accuracies": 0.7562500238418579, | |
| "rewards/chosen": -2.09273624420166, | |
| "rewards/margins": 1.3475998640060425, | |
| "rewards/rejected": -3.440336227416992, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 4.292990551804171e-08, | |
| "logits/chosen": 0.11955185234546661, | |
| "logits/rejected": 0.2987907826900482, | |
| "logps/chosen": -521.8675537109375, | |
| "logps/rejected": -622.3560791015625, | |
| "loss": 0.0115, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.9727070331573486, | |
| "rewards/margins": 1.207002878189087, | |
| "rewards/rejected": -3.1797099113464355, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.581603349196371e-08, | |
| "logits/chosen": 0.12183141708374023, | |
| "logits/rejected": 0.24950018525123596, | |
| "logps/chosen": -529.2427978515625, | |
| "logps/rejected": -662.9299926757812, | |
| "loss": 0.0112, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/chosen": -2.024509906768799, | |
| "rewards/margins": 1.5907318592071533, | |
| "rewards/rejected": -3.615241289138794, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.9324414157151367e-08, | |
| "logits/chosen": 0.11247365176677704, | |
| "logits/rejected": 0.28803473711013794, | |
| "logps/chosen": -538.6015625, | |
| "logps/rejected": -616.6097412109375, | |
| "loss": 0.0105, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": -2.005286693572998, | |
| "rewards/margins": 1.320533037185669, | |
| "rewards/rejected": -3.325819492340088, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 2.3463749726290284e-08, | |
| "logits/chosen": 0.09726160764694214, | |
| "logits/rejected": 0.3085189759731293, | |
| "logps/chosen": -527.7420043945312, | |
| "logps/rejected": -666.7064208984375, | |
| "loss": 0.0114, | |
| "rewards/accuracies": 0.78125, | |
| "rewards/chosen": -1.963595986366272, | |
| "rewards/margins": 1.6061077117919922, | |
| "rewards/rejected": -3.5697035789489746, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.824189659787284e-08, | |
| "logits/chosen": 0.19652321934700012, | |
| "logits/rejected": 0.2885872423648834, | |
| "logps/chosen": -515.560546875, | |
| "logps/rejected": -641.10791015625, | |
| "loss": 0.0111, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -1.9605176448822021, | |
| "rewards/margins": 1.3721264600753784, | |
| "rewards/rejected": -3.33264422416687, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.3665854824458035e-08, | |
| "logits/chosen": 0.16733339428901672, | |
| "logits/rejected": 0.3634529113769531, | |
| "logps/chosen": -542.18505859375, | |
| "logps/rejected": -629.7310791015625, | |
| "loss": 0.0115, | |
| "rewards/accuracies": 0.768750011920929, | |
| "rewards/chosen": -2.0391831398010254, | |
| "rewards/margins": 1.1835925579071045, | |
| "rewards/rejected": -3.2227752208709717, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 9.741758728888217e-09, | |
| "logits/chosen": 0.08950433880090714, | |
| "logits/rejected": 0.2665843069553375, | |
| "logps/chosen": -533.1641845703125, | |
| "logps/rejected": -621.0523681640625, | |
| "loss": 0.0113, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -1.9605424404144287, | |
| "rewards/margins": 1.1125773191452026, | |
| "rewards/rejected": -3.073119640350342, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 6.474868681043577e-09, | |
| "logits/chosen": 0.13345034420490265, | |
| "logits/rejected": 0.2458508014678955, | |
| "logps/chosen": -523.0572509765625, | |
| "logps/rejected": -666.5548706054688, | |
| "loss": 0.0107, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -2.094968557357788, | |
| "rewards/margins": 1.4136923551559448, | |
| "rewards/rejected": -3.5086607933044434, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 3.869564046156459e-09, | |
| "logits/chosen": 0.17636564373970032, | |
| "logits/rejected": 0.24904970824718475, | |
| "logps/chosen": -521.7586669921875, | |
| "logps/rejected": -661.547119140625, | |
| "loss": 0.0115, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -2.0953400135040283, | |
| "rewards/margins": 1.3953152894973755, | |
| "rewards/rejected": -3.4906551837921143, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.929337314139412e-09, | |
| "logits/chosen": 0.1708141714334488, | |
| "logits/rejected": 0.2874212861061096, | |
| "logps/chosen": -481.3929138183594, | |
| "logps/rejected": -591.492431640625, | |
| "loss": 0.0107, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -1.8482071161270142, | |
| "rewards/margins": 1.3176212310791016, | |
| "rewards/rejected": -3.165828227996826, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.567894177967325e-10, | |
| "logits/chosen": 0.1810809224843979, | |
| "logits/rejected": 0.3499010503292084, | |
| "logps/chosen": -509.21966552734375, | |
| "logps/rejected": -619.0591430664062, | |
| "loss": 0.0119, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -1.7878868579864502, | |
| "rewards/margins": 1.3797376155853271, | |
| "rewards/rejected": -3.1676242351531982, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 5.3626246194704575e-11, | |
| "logits/chosen": 0.12432925403118134, | |
| "logits/rejected": 0.1847553700208664, | |
| "logps/chosen": -471.4737854003906, | |
| "logps/rejected": -620.7115478515625, | |
| "loss": 0.0121, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.8229620456695557, | |
| "rewards/margins": 1.5415856838226318, | |
| "rewards/rejected": -3.3645477294921875, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 954, | |
| "total_flos": 0.0, | |
| "train_loss": 0.050850671487596796, | |
| "train_runtime": 12712.7589, | |
| "train_samples_per_second": 9.618, | |
| "train_steps_per_second": 0.075 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 954, | |
| "num_train_epochs": 2, | |
| "save_steps": 10000, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |