| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9994666666666666, | |
| "eval_steps": 500, | |
| "global_step": 937, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 5.319148936170213e-08, | |
| "logits/chosen": 0.4053989052772522, | |
| "logits/rejected": 0.1312936246395111, | |
| "logps/chosen": -434.00537109375, | |
| "logps/rejected": -516.5983276367188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5.319148936170213e-07, | |
| "logits/chosen": 0.15122921764850616, | |
| "logits/rejected": 0.29006803035736084, | |
| "logps/chosen": -365.09381103515625, | |
| "logps/rejected": -352.4356994628906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.0638297872340427e-06, | |
| "logits/chosen": 0.11831488460302353, | |
| "logits/rejected": 0.20268961787223816, | |
| "logps/chosen": -339.44415283203125, | |
| "logps/rejected": -347.7588806152344, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.595744680851064e-06, | |
| "logits/chosen": 0.25372257828712463, | |
| "logits/rejected": 0.2503891885280609, | |
| "logps/chosen": -382.58709716796875, | |
| "logps/rejected": -363.3986511230469, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.1276595744680853e-06, | |
| "logits/chosen": 0.15515519678592682, | |
| "logits/rejected": 0.2712388336658478, | |
| "logps/chosen": -401.01165771484375, | |
| "logps/rejected": -385.3412170410156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.6595744680851065e-06, | |
| "logits/chosen": 0.23549167811870575, | |
| "logits/rejected": 0.27643483877182007, | |
| "logps/chosen": -440.90948486328125, | |
| "logps/rejected": -431.11541748046875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.191489361702128e-06, | |
| "logits/chosen": 0.17448191344738007, | |
| "logits/rejected": 0.3222740590572357, | |
| "logps/chosen": -408.9176940917969, | |
| "logps/rejected": -412.7042541503906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.723404255319149e-06, | |
| "logits/chosen": 0.131931334733963, | |
| "logits/rejected": 0.2314920872449875, | |
| "logps/chosen": -354.6463623046875, | |
| "logps/rejected": -355.9555969238281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.255319148936171e-06, | |
| "logits/chosen": 0.15874435007572174, | |
| "logits/rejected": 0.26632028818130493, | |
| "logps/chosen": -392.85498046875, | |
| "logps/rejected": -379.8575744628906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.787234042553192e-06, | |
| "logits/chosen": 0.26180005073547363, | |
| "logits/rejected": 0.19905249774456024, | |
| "logps/chosen": -392.9334716796875, | |
| "logps/rejected": -414.093017578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.999375059004058e-06, | |
| "logits/chosen": 0.1863372027873993, | |
| "logits/rejected": 0.19517865777015686, | |
| "logps/chosen": -417.194580078125, | |
| "logps/rejected": -409.6343688964844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.9955571065548795e-06, | |
| "logits/chosen": 0.2684422731399536, | |
| "logits/rejected": 0.19285789132118225, | |
| "logps/chosen": -419.1029357910156, | |
| "logps/rejected": -400.07568359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.9882736864879e-06, | |
| "logits/chosen": 0.1304994523525238, | |
| "logits/rejected": 0.3027098774909973, | |
| "logps/chosen": -414.75128173828125, | |
| "logps/rejected": -442.08343505859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.977534912960124e-06, | |
| "logits/chosen": 0.21744604408740997, | |
| "logits/rejected": 0.3507159650325775, | |
| "logps/chosen": -412.56658935546875, | |
| "logps/rejected": -398.8998107910156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.963355698422092e-06, | |
| "logits/chosen": 0.2333650141954422, | |
| "logits/rejected": 0.2951519191265106, | |
| "logps/chosen": -384.1523132324219, | |
| "logps/rejected": -364.6881408691406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.945755732909625e-06, | |
| "logits/chosen": 0.12179046869277954, | |
| "logits/rejected": 0.17499811947345734, | |
| "logps/chosen": -382.7596740722656, | |
| "logps/rejected": -385.5395812988281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.924759456701167e-06, | |
| "logits/chosen": 0.2078637331724167, | |
| "logits/rejected": 0.29525110125541687, | |
| "logps/chosen": -423.19097900390625, | |
| "logps/rejected": -411.33038330078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.900396026378671e-06, | |
| "logits/chosen": 0.17098727822303772, | |
| "logits/rejected": 0.3464776873588562, | |
| "logps/chosen": -440.69927978515625, | |
| "logps/rejected": -413.71343994140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.872699274339169e-06, | |
| "logits/chosen": 0.2748049795627594, | |
| "logits/rejected": 0.28834646940231323, | |
| "logps/chosen": -381.0144958496094, | |
| "logps/rejected": -380.5312805175781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.8417076618132434e-06, | |
| "logits/chosen": 0.21008631587028503, | |
| "logits/rejected": 0.2946034371852875, | |
| "logps/chosen": -484.1692810058594, | |
| "logps/rejected": -452.757080078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.807464225455655e-06, | |
| "logits/chosen": 0.14078669250011444, | |
| "logits/rejected": 0.28049755096435547, | |
| "logps/chosen": -404.1522216796875, | |
| "logps/rejected": -421.33856201171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.770016517582283e-06, | |
| "logits/chosen": 0.2193254977464676, | |
| "logits/rejected": 0.22546172142028809, | |
| "logps/chosen": -401.60546875, | |
| "logps/rejected": -405.28582763671875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.7294165401363616e-06, | |
| "logits/chosen": 0.22740697860717773, | |
| "logits/rejected": 0.2075118124485016, | |
| "logps/chosen": -401.9066467285156, | |
| "logps/rejected": -396.5219421386719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.68572067247573e-06, | |
| "logits/chosen": 0.2253454476594925, | |
| "logits/rejected": 0.24970141053199768, | |
| "logps/chosen": -399.9390563964844, | |
| "logps/rejected": -400.8056335449219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.638989593081364e-06, | |
| "logits/chosen": 0.09361274540424347, | |
| "logits/rejected": 0.2551277279853821, | |
| "logps/chosen": -377.5755310058594, | |
| "logps/rejected": -380.84381103515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.5892881952959015e-06, | |
| "logits/chosen": 0.21397773921489716, | |
| "logits/rejected": 0.27325302362442017, | |
| "logps/chosen": -414.1591796875, | |
| "logps/rejected": -400.00323486328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.536685497209182e-06, | |
| "logits/chosen": 0.20620456337928772, | |
| "logits/rejected": 0.25625258684158325, | |
| "logps/chosen": -431.2455139160156, | |
| "logps/rejected": -444.0306701660156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.481254545815943e-06, | |
| "logits/chosen": 0.17776310443878174, | |
| "logits/rejected": 0.22469878196716309, | |
| "logps/chosen": -433.8826599121094, | |
| "logps/rejected": -453.208740234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.42307231557875e-06, | |
| "logits/chosen": 0.16825468838214874, | |
| "logits/rejected": 0.25171762704849243, | |
| "logps/chosen": -408.1540832519531, | |
| "logps/rejected": -394.03704833984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.3622196015370305e-06, | |
| "logits/chosen": 0.17698076367378235, | |
| "logits/rejected": 0.2516983151435852, | |
| "logps/chosen": -402.94781494140625, | |
| "logps/rejected": -432.970947265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.298780907110648e-06, | |
| "logits/chosen": 0.2270512878894806, | |
| "logits/rejected": 0.2209252119064331, | |
| "logps/chosen": -423.2540588378906, | |
| "logps/rejected": -388.04620361328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.23284432675381e-06, | |
| "logits/chosen": 0.1234111562371254, | |
| "logits/rejected": 0.27970707416534424, | |
| "logps/chosen": -379.5980529785156, | |
| "logps/rejected": -379.2107849121094, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.164501423622277e-06, | |
| "logits/chosen": 0.20214669406414032, | |
| "logits/rejected": 0.2584760785102844, | |
| "logps/chosen": -382.29937744140625, | |
| "logps/rejected": -370.25616455078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.0938471024237355e-06, | |
| "logits/chosen": 0.24651730060577393, | |
| "logits/rejected": 0.24896959960460663, | |
| "logps/chosen": -437.11859130859375, | |
| "logps/rejected": -405.49566650390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.020979477627907e-06, | |
| "logits/chosen": 0.28022533655166626, | |
| "logits/rejected": 0.31104576587677, | |
| "logps/chosen": -419.55438232421875, | |
| "logps/rejected": -423.57391357421875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.9459997372194105e-06, | |
| "logits/chosen": 0.18333196640014648, | |
| "logits/rejected": 0.3204515278339386, | |
| "logps/chosen": -437.1458435058594, | |
| "logps/rejected": -431.79461669921875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.869012002182573e-06, | |
| "logits/chosen": 0.11528631299734116, | |
| "logits/rejected": 0.2239673137664795, | |
| "logps/chosen": -450.3658142089844, | |
| "logps/rejected": -414.203369140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.7901231819133104e-06, | |
| "logits/chosen": 0.24927763640880585, | |
| "logits/rejected": 0.26546379923820496, | |
| "logps/chosen": -399.69439697265625, | |
| "logps/rejected": -403.4917907714844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.709442825758875e-06, | |
| "logits/chosen": 0.11166371405124664, | |
| "logits/rejected": 0.2701989710330963, | |
| "logps/chosen": -404.04449462890625, | |
| "logps/rejected": -379.7621765136719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.6270829708916113e-06, | |
| "logits/chosen": 0.17965979874134064, | |
| "logits/rejected": 0.2478092461824417, | |
| "logps/chosen": -421.53173828125, | |
| "logps/rejected": -430.0699768066406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.543157986727991e-06, | |
| "logits/chosen": 0.2598588168621063, | |
| "logits/rejected": 0.292039692401886, | |
| "logps/chosen": -423.152099609375, | |
| "logps/rejected": -419.819091796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.4577844161089614e-06, | |
| "logits/chosen": 0.2145676165819168, | |
| "logits/rejected": 0.2213706225156784, | |
| "logps/chosen": -421.8087463378906, | |
| "logps/rejected": -441.27716064453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.3710808134621577e-06, | |
| "logits/chosen": 0.23070721328258514, | |
| "logits/rejected": 0.2725136876106262, | |
| "logps/chosen": -446.6600036621094, | |
| "logps/rejected": -443.410400390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.2831675801707126e-06, | |
| "logits/chosen": 0.22015449404716492, | |
| "logits/rejected": 0.22397899627685547, | |
| "logps/chosen": -367.9841613769531, | |
| "logps/rejected": -366.0374450683594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.194166797377289e-06, | |
| "logits/chosen": 0.22503578662872314, | |
| "logits/rejected": 0.2497364729642868, | |
| "logps/chosen": -451.5084533691406, | |
| "logps/rejected": -441.224365234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.104202056455501e-06, | |
| "logits/chosen": 0.21670684218406677, | |
| "logits/rejected": 0.2709944248199463, | |
| "logps/chosen": -425.6534729003906, | |
| "logps/rejected": -408.99920654296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.013398287384144e-06, | |
| "logits/chosen": 0.2238447666168213, | |
| "logits/rejected": 0.28349417448043823, | |
| "logps/chosen": -438.446533203125, | |
| "logps/rejected": -419.7816467285156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.9218815852625717e-06, | |
| "logits/chosen": 0.21700212359428406, | |
| "logits/rejected": 0.250012069940567, | |
| "logps/chosen": -414.27703857421875, | |
| "logps/rejected": -413.127197265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.829779035208113e-06, | |
| "logits/chosen": 0.18912899494171143, | |
| "logits/rejected": 0.30137619376182556, | |
| "logps/chosen": -388.0596923828125, | |
| "logps/rejected": -392.78375244140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.737218535878705e-06, | |
| "logits/chosen": 0.257573664188385, | |
| "logits/rejected": 0.27977752685546875, | |
| "logps/chosen": -377.6993713378906, | |
| "logps/rejected": -384.7877197265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.64432862186579e-06, | |
| "logits/chosen": 0.18493638932704926, | |
| "logits/rejected": 0.15904666483402252, | |
| "logps/chosen": -369.86480712890625, | |
| "logps/rejected": -374.7712097167969, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.551238285204126e-06, | |
| "logits/chosen": 0.2099113166332245, | |
| "logits/rejected": 0.26101061701774597, | |
| "logps/chosen": -445.0758361816406, | |
| "logps/rejected": -431.1841735839844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.4580767962463688e-06, | |
| "logits/chosen": 0.15272489190101624, | |
| "logits/rejected": 0.2410932034254074, | |
| "logps/chosen": -411.3912658691406, | |
| "logps/rejected": -378.7525939941406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.3649735241511546e-06, | |
| "logits/chosen": 0.2886171042919159, | |
| "logits/rejected": 0.24891161918640137, | |
| "logps/chosen": -412.112548828125, | |
| "logps/rejected": -403.3191223144531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.2720577572339914e-06, | |
| "logits/chosen": 0.14852824807167053, | |
| "logits/rejected": 0.23693397641181946, | |
| "logps/chosen": -404.3973693847656, | |
| "logps/rejected": -389.05389404296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.1794585234303995e-06, | |
| "logits/chosen": 0.16628649830818176, | |
| "logits/rejected": 0.27538853883743286, | |
| "logps/chosen": -418.27752685546875, | |
| "logps/rejected": -411.8797302246094, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.0873044111206407e-06, | |
| "logits/chosen": 0.25115758180618286, | |
| "logits/rejected": 0.27927032113075256, | |
| "logps/chosen": -378.09954833984375, | |
| "logps/rejected": -408.1378173828125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.9957233905648293e-06, | |
| "logits/chosen": 0.15070781111717224, | |
| "logits/rejected": 0.18165943026542664, | |
| "logps/chosen": -366.8854064941406, | |
| "logps/rejected": -366.2696533203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.904842636196402e-06, | |
| "logits/chosen": 0.2222457379102707, | |
| "logits/rejected": 0.27251940965652466, | |
| "logps/chosen": -410.501708984375, | |
| "logps/rejected": -397.37176513671875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.814788350020726e-06, | |
| "logits/chosen": 0.20956209301948547, | |
| "logits/rejected": 0.33177804946899414, | |
| "logps/chosen": -430.733154296875, | |
| "logps/rejected": -419.57171630859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.725685586364051e-06, | |
| "logits/chosen": 0.18513911962509155, | |
| "logits/rejected": 0.23731224238872528, | |
| "logps/chosen": -357.6490173339844, | |
| "logps/rejected": -381.7027893066406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.6376580782162172e-06, | |
| "logits/chosen": 0.15748406946659088, | |
| "logits/rejected": 0.15097612142562866, | |
| "logps/chosen": -408.1110534667969, | |
| "logps/rejected": -388.00592041015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.550828065408227e-06, | |
| "logits/chosen": 0.2554689049720764, | |
| "logits/rejected": 0.17269960045814514, | |
| "logps/chosen": -394.124755859375, | |
| "logps/rejected": -425.56640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.4653161248633053e-06, | |
| "logits/chosen": 0.17127978801727295, | |
| "logits/rejected": 0.19732491672039032, | |
| "logps/chosen": -372.42608642578125, | |
| "logps/rejected": -354.8905944824219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.381241003157162e-06, | |
| "logits/chosen": 0.15929067134857178, | |
| "logits/rejected": 0.21373197436332703, | |
| "logps/chosen": -389.2451477050781, | |
| "logps/rejected": -373.00384521484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.298719451619979e-06, | |
| "logits/chosen": 0.16366329789161682, | |
| "logits/rejected": 0.35690051317214966, | |
| "logps/chosen": -423.4112854003906, | |
| "logps/rejected": -407.52838134765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.2178660642091036e-06, | |
| "logits/chosen": 0.1489204317331314, | |
| "logits/rejected": 0.2767201364040375, | |
| "logps/chosen": -428.436279296875, | |
| "logps/rejected": -410.24310302734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.1387931183775821e-06, | |
| "logits/chosen": 0.2932317554950714, | |
| "logits/rejected": 0.2427135407924652, | |
| "logps/chosen": -389.24945068359375, | |
| "logps/rejected": -385.98431396484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.061610419159532e-06, | |
| "logits/chosen": 0.2365073710680008, | |
| "logits/rejected": 0.2310241162776947, | |
| "logps/chosen": -370.1266174316406, | |
| "logps/rejected": -356.8659973144531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.864251466888364e-07, | |
| "logits/chosen": 0.15916849672794342, | |
| "logits/rejected": 0.27135321497917175, | |
| "logps/chosen": -395.91595458984375, | |
| "logps/rejected": -390.37420654296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 9.133417073629288e-07, | |
| "logits/chosen": 0.15306909382343292, | |
| "logits/rejected": 0.2204822301864624, | |
| "logps/chosen": -395.4036865234375, | |
| "logps/rejected": -399.2796325683594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.424615888583332e-07, | |
| "logits/chosen": 0.17490191757678986, | |
| "logits/rejected": 0.27996110916137695, | |
| "logps/chosen": -409.8934631347656, | |
| "logps/rejected": -411.23480224609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.738832191993092e-07, | |
| "logits/chosen": 0.20883989334106445, | |
| "logits/rejected": 0.22538049519062042, | |
| "logps/chosen": -420.298095703125, | |
| "logps/rejected": -409.4375305175781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.077018300752917e-07, | |
| "logits/chosen": 0.2503129839897156, | |
| "logits/rejected": 0.21712422370910645, | |
| "logps/chosen": -424.7351989746094, | |
| "logps/rejected": -402.7471618652344, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.440093245969342e-07, | |
| "logits/chosen": 0.18410465121269226, | |
| "logits/rejected": 0.2769954800605774, | |
| "logps/chosen": -381.27532958984375, | |
| "logps/rejected": -350.406494140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.828941496744075e-07, | |
| "logits/chosen": 0.19894076883792877, | |
| "logits/rejected": 0.2793256342411041, | |
| "logps/chosen": -423.1390075683594, | |
| "logps/rejected": -429.28472900390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.244411731951671e-07, | |
| "logits/chosen": 0.14793696999549866, | |
| "logits/rejected": 0.34965780377388, | |
| "logps/chosen": -386.9423522949219, | |
| "logps/rejected": -361.16180419921875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.6873156617173594e-07, | |
| "logits/chosen": 0.1257942169904709, | |
| "logits/rejected": 0.19840510189533234, | |
| "logps/chosen": -433.75634765625, | |
| "logps/rejected": -401.2398376464844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.1584269002318653e-07, | |
| "logits/chosen": 0.1670645773410797, | |
| "logits/rejected": 0.28423744440078735, | |
| "logps/chosen": -436.582763671875, | |
| "logps/rejected": -441.0210876464844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.658479891468258e-07, | |
| "logits/chosen": 0.2583349943161011, | |
| "logits/rejected": 0.29037925601005554, | |
| "logps/chosen": -402.0462951660156, | |
| "logps/rejected": -401.1708068847656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.18816888929272e-07, | |
| "logits/chosen": 0.15391795337200165, | |
| "logits/rejected": 0.2551635205745697, | |
| "logps/chosen": -418.2384338378906, | |
| "logps/rejected": -386.5793762207031, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.748146993385484e-07, | |
| "logits/chosen": 0.24473810195922852, | |
| "logits/rejected": 0.2084934264421463, | |
| "logps/chosen": -445.95849609375, | |
| "logps/rejected": -444.5816345214844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.3390252423108077e-07, | |
| "logits/chosen": 0.14975102245807648, | |
| "logits/rejected": 0.33251649141311646, | |
| "logps/chosen": -438.55206298828125, | |
| "logps/rejected": -396.28961181640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.961371764995243e-07, | |
| "logits/chosen": 0.28815943002700806, | |
| "logits/rejected": 0.2893601059913635, | |
| "logps/chosen": -418.80413818359375, | |
| "logps/rejected": -398.5482177734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.61571099179261e-07, | |
| "logits/chosen": 0.14490856230258942, | |
| "logits/rejected": 0.22375395894050598, | |
| "logps/chosen": -368.81512451171875, | |
| "logps/rejected": -351.7200012207031, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.3025229262312367e-07, | |
| "logits/chosen": 0.1880328208208084, | |
| "logits/rejected": 0.25640177726745605, | |
| "logps/chosen": -419.2107849121094, | |
| "logps/rejected": -408.9618835449219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.0222424784546853e-07, | |
| "logits/chosen": 0.34192991256713867, | |
| "logits/rejected": 0.2116626501083374, | |
| "logps/chosen": -400.17901611328125, | |
| "logps/rejected": -410.69036865234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.752588612816553e-08, | |
| "logits/chosen": 0.2738000750541687, | |
| "logits/rejected": 0.19856445491313934, | |
| "logps/chosen": -449.3203125, | |
| "logps/rejected": -424.33392333984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 5.619150497236991e-08, | |
| "logits/chosen": 0.1780010461807251, | |
| "logits/rejected": 0.23323968052864075, | |
| "logps/chosen": -403.2125244140625, | |
| "logps/rejected": -399.965087890625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.825073047112743e-08, | |
| "logits/chosen": 0.14893004298210144, | |
| "logits/rejected": 0.25546133518218994, | |
| "logps/chosen": -415.7379455566406, | |
| "logps/rejected": -399.75103759765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.372847616895685e-08, | |
| "logits/chosen": 0.2190527468919754, | |
| "logits/rejected": 0.18029525876045227, | |
| "logps/chosen": -382.27874755859375, | |
| "logps/rejected": -406.80657958984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.264490846553279e-08, | |
| "logits/chosen": 0.22972765564918518, | |
| "logits/rejected": 0.21643073856830597, | |
| "logps/chosen": -430.3443298339844, | |
| "logps/rejected": -428.8854064941406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 5.015418611516165e-09, | |
| "logits/chosen": 0.1947575956583023, | |
| "logits/rejected": 0.18881197273731232, | |
| "logps/chosen": -383.63995361328125, | |
| "logps/rejected": -366.2297668457031, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 8.506013354186993e-10, | |
| "logits/chosen": 0.22291317582130432, | |
| "logits/rejected": 0.2307729721069336, | |
| "logps/chosen": -384.4226379394531, | |
| "logps/rejected": -388.9614562988281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 937, | |
| "total_flos": 0.0, | |
| "train_loss": 0.0, | |
| "train_runtime": 6824.5604, | |
| "train_samples_per_second": 4.396, | |
| "train_steps_per_second": 0.137 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 937, | |
| "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 | |
| } | |