| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9982631930527722, | |
| "eval_steps": 400, | |
| "global_step": 467, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01068804275217101, | |
| "grad_norm": 7.85619992560556, | |
| "learning_rate": 6.382978723404255e-08, | |
| "logits/chosen": 0.06174152344465256, | |
| "logits/rejected": 0.037658724933862686, | |
| "logps/chosen": -0.2702512741088867, | |
| "logps/rejected": -0.26810693740844727, | |
| "loss": 1.2748, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.2702512741088867, | |
| "rewards/margins": -0.002144329948350787, | |
| "rewards/rejected": -0.26810693740844727, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02137608550434202, | |
| "grad_norm": 5.1085151748010285, | |
| "learning_rate": 1.276595744680851e-07, | |
| "logits/chosen": -0.010784180834889412, | |
| "logits/rejected": -0.012599915266036987, | |
| "logps/chosen": -0.26977622509002686, | |
| "logps/rejected": -0.26753830909729004, | |
| "loss": 1.2725, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.26977622509002686, | |
| "rewards/margins": -0.0022379131987690926, | |
| "rewards/rejected": -0.26753830909729004, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03206412825651302, | |
| "grad_norm": 6.692216762984584, | |
| "learning_rate": 1.9148936170212767e-07, | |
| "logits/chosen": 0.0017719150055199862, | |
| "logits/rejected": 0.004488155245780945, | |
| "logps/chosen": -0.27860361337661743, | |
| "logps/rejected": -0.2837750315666199, | |
| "loss": 1.2793, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -0.27860361337661743, | |
| "rewards/margins": 0.005171425640583038, | |
| "rewards/rejected": -0.2837750315666199, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04275217100868404, | |
| "grad_norm": 8.996125297973581, | |
| "learning_rate": 2.553191489361702e-07, | |
| "logits/chosen": -0.06322917342185974, | |
| "logits/rejected": -0.06904581189155579, | |
| "logps/chosen": -0.2816649079322815, | |
| "logps/rejected": -0.2903934419155121, | |
| "loss": 1.2759, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.2816649079322815, | |
| "rewards/margins": 0.008728553541004658, | |
| "rewards/rejected": -0.2903934419155121, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.053440213760855046, | |
| "grad_norm": 4.855846145207527, | |
| "learning_rate": 3.1914893617021275e-07, | |
| "logits/chosen": -0.06888748705387115, | |
| "logits/rejected": -0.04847611114382744, | |
| "logps/chosen": -0.25571519136428833, | |
| "logps/rejected": -0.2750292420387268, | |
| "loss": 1.2568, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.25571519136428833, | |
| "rewards/margins": 0.01931406930088997, | |
| "rewards/rejected": -0.2750292420387268, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06412825651302605, | |
| "grad_norm": 4.835742771788438, | |
| "learning_rate": 3.8297872340425535e-07, | |
| "logits/chosen": -0.014270897023379803, | |
| "logits/rejected": -0.006606946233659983, | |
| "logps/chosen": -0.27899158000946045, | |
| "logps/rejected": -0.2938278019428253, | |
| "loss": 1.2771, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.27899158000946045, | |
| "rewards/margins": 0.014836207032203674, | |
| "rewards/rejected": -0.2938278019428253, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07481629926519706, | |
| "grad_norm": 6.995787010953034, | |
| "learning_rate": 4.4680851063829783e-07, | |
| "logits/chosen": -0.03009852208197117, | |
| "logits/rejected": -0.009835416451096535, | |
| "logps/chosen": -0.2785848379135132, | |
| "logps/rejected": -0.29589539766311646, | |
| "loss": 1.2753, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.2785848379135132, | |
| "rewards/margins": 0.017310529947280884, | |
| "rewards/rejected": -0.29589539766311646, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08550434201736808, | |
| "grad_norm": 5.347511039842285, | |
| "learning_rate": 5.106382978723404e-07, | |
| "logits/chosen": -0.06675975024700165, | |
| "logits/rejected": -0.07226153463125229, | |
| "logps/chosen": -0.26110929250717163, | |
| "logps/rejected": -0.2702568769454956, | |
| "loss": 1.2672, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.26110929250717163, | |
| "rewards/margins": 0.009147587232291698, | |
| "rewards/rejected": -0.2702568769454956, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09619238476953908, | |
| "grad_norm": 5.575092391618783, | |
| "learning_rate": 5.74468085106383e-07, | |
| "logits/chosen": -0.0533546507358551, | |
| "logits/rejected": -0.010820349678397179, | |
| "logps/chosen": -0.2856101095676422, | |
| "logps/rejected": -0.28608259558677673, | |
| "loss": 1.2739, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.2856101095676422, | |
| "rewards/margins": 0.0004724980390165001, | |
| "rewards/rejected": -0.28608259558677673, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10688042752171009, | |
| "grad_norm": 4.346264213431051, | |
| "learning_rate": 5.999244704827519e-07, | |
| "logits/chosen": -0.016516100615262985, | |
| "logits/rejected": -0.003799914848059416, | |
| "logps/chosen": -0.2856972813606262, | |
| "logps/rejected": -0.30010929703712463, | |
| "loss": 1.2684, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.2856972813606262, | |
| "rewards/margins": 0.014411995187401772, | |
| "rewards/rejected": -0.30010929703712463, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11756847027388109, | |
| "grad_norm": 5.956061544604284, | |
| "learning_rate": 5.994630389303205e-07, | |
| "logits/chosen": 0.017358671873807907, | |
| "logits/rejected": -0.0036045790184289217, | |
| "logps/chosen": -0.2718290686607361, | |
| "logps/rejected": -0.2807454764842987, | |
| "loss": 1.2628, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.2718290686607361, | |
| "rewards/margins": 0.00891642551869154, | |
| "rewards/rejected": -0.2807454764842987, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.1282565130260521, | |
| "grad_norm": 6.117995966956006, | |
| "learning_rate": 5.985827812395378e-07, | |
| "logits/chosen": -0.038348693400621414, | |
| "logits/rejected": -0.0700547844171524, | |
| "logps/chosen": -0.285757839679718, | |
| "logps/rejected": -0.3150267004966736, | |
| "loss": 1.2628, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.285757839679718, | |
| "rewards/margins": 0.029268871992826462, | |
| "rewards/rejected": -0.3150267004966736, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13894455577822312, | |
| "grad_norm": 5.259018808160778, | |
| "learning_rate": 5.972849285303804e-07, | |
| "logits/chosen": -0.023639511317014694, | |
| "logits/rejected": 0.03467974066734314, | |
| "logps/chosen": -0.29160580039024353, | |
| "logps/rejected": -0.31474000215530396, | |
| "loss": 1.2708, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.29160580039024353, | |
| "rewards/margins": 0.02313421666622162, | |
| "rewards/rejected": -0.31474000215530396, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.14963259853039412, | |
| "grad_norm": 5.111432049765508, | |
| "learning_rate": 5.955712959672177e-07, | |
| "logits/chosen": -0.016536986455321312, | |
| "logits/rejected": -0.02064664289355278, | |
| "logps/chosen": -0.29413312673568726, | |
| "logps/rejected": -0.34755489230155945, | |
| "loss": 1.2542, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.29413312673568726, | |
| "rewards/margins": 0.05342177301645279, | |
| "rewards/rejected": -0.34755489230155945, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.16032064128256512, | |
| "grad_norm": 4.471806926829, | |
| "learning_rate": 5.934442802201417e-07, | |
| "logits/chosen": 0.06339865177869797, | |
| "logits/rejected": 0.10371309518814087, | |
| "logps/chosen": -0.30393484234809875, | |
| "logps/rejected": -0.33556467294692993, | |
| "loss": 1.2679, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.30393484234809875, | |
| "rewards/margins": 0.031629838049411774, | |
| "rewards/rejected": -0.33556467294692993, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.17100868403473615, | |
| "grad_norm": 6.2280285157401165, | |
| "learning_rate": 5.909068561130061e-07, | |
| "logits/chosen": -0.013803797774016857, | |
| "logits/rejected": -0.005705256946384907, | |
| "logps/chosen": -0.29255902767181396, | |
| "logps/rejected": -0.32304221391677856, | |
| "loss": 1.2618, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.29255902767181396, | |
| "rewards/margins": 0.030483180657029152, | |
| "rewards/rejected": -0.32304221391677856, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.18169672678690715, | |
| "grad_norm": 4.62670410972183, | |
| "learning_rate": 5.879625724628667e-07, | |
| "logits/chosen": 0.007588694803416729, | |
| "logits/rejected": 0.023027362301945686, | |
| "logps/chosen": -0.29541558027267456, | |
| "logps/rejected": -0.3341410756111145, | |
| "loss": 1.2571, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.29541558027267456, | |
| "rewards/margins": 0.03872550651431084, | |
| "rewards/rejected": -0.3341410756111145, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.19238476953907815, | |
| "grad_norm": 6.280947217475369, | |
| "learning_rate": 5.846155471166399e-07, | |
| "logits/chosen": 0.01748380810022354, | |
| "logits/rejected": 0.038063496351242065, | |
| "logps/chosen": -0.31113117933273315, | |
| "logps/rejected": -0.36816665530204773, | |
| "loss": 1.2529, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.31113117933273315, | |
| "rewards/margins": 0.05703546851873398, | |
| "rewards/rejected": -0.36816665530204773, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20307281229124916, | |
| "grad_norm": 6.33795222726401, | |
| "learning_rate": 5.808704611919212e-07, | |
| "logits/chosen": 0.008561496622860432, | |
| "logits/rejected": -0.009577239863574505, | |
| "logps/chosen": -0.3074243664741516, | |
| "logps/rejected": -0.3172617554664612, | |
| "loss": 1.2598, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.3074243664741516, | |
| "rewards/margins": 0.0098373768851161, | |
| "rewards/rejected": -0.3172617554664612, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.21376085504342018, | |
| "grad_norm": 5.4836519913536, | |
| "learning_rate": 5.767325525300187e-07, | |
| "logits/chosen": 0.012827359139919281, | |
| "logits/rejected": 0.015095492824912071, | |
| "logps/chosen": -0.3060360550880432, | |
| "logps/rejected": -0.3578498363494873, | |
| "loss": 1.257, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.3060360550880432, | |
| "rewards/margins": 0.051813799887895584, | |
| "rewards/rejected": -0.3578498363494873, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22444889779559118, | |
| "grad_norm": 9.106962216409888, | |
| "learning_rate": 5.722076083703594e-07, | |
| "logits/chosen": -0.011002029292285442, | |
| "logits/rejected": -0.01515892706811428, | |
| "logps/chosen": -0.2857062518596649, | |
| "logps/rejected": -0.3439631760120392, | |
| "loss": 1.2453, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.2857062518596649, | |
| "rewards/margins": 0.05825691670179367, | |
| "rewards/rejected": -0.3439631760120392, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.23513694054776219, | |
| "grad_norm": 6.085383517197391, | |
| "learning_rate": 5.673019572565103e-07, | |
| "logits/chosen": -0.023035719990730286, | |
| "logits/rejected": -0.03461029753088951, | |
| "logps/chosen": -0.29488229751586914, | |
| "logps/rejected": -0.3499522805213928, | |
| "loss": 1.2381, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.29488229751586914, | |
| "rewards/margins": 0.05506999418139458, | |
| "rewards/rejected": -0.3499522805213928, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2458249832999332, | |
| "grad_norm": 5.078175131364166, | |
| "learning_rate": 5.620224601851389e-07, | |
| "logits/chosen": 0.005257171578705311, | |
| "logits/rejected": 0.0035833939909934998, | |
| "logps/chosen": -0.3048323094844818, | |
| "logps/rejected": -0.3554520010948181, | |
| "loss": 1.2562, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.3048323094844818, | |
| "rewards/margins": 0.050619661808013916, | |
| "rewards/rejected": -0.3554520010948181, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.2565130260521042, | |
| "grad_norm": 7.17959663924209, | |
| "learning_rate": 5.563765010102885e-07, | |
| "logits/chosen": -0.06386958062648773, | |
| "logits/rejected": -0.04022233560681343, | |
| "logps/chosen": -0.3290977478027344, | |
| "logps/rejected": -0.36753392219543457, | |
| "loss": 1.2551, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.3290977478027344, | |
| "rewards/margins": 0.038436152040958405, | |
| "rewards/rejected": -0.36753392219543457, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.26720106880427524, | |
| "grad_norm": 5.129342143579583, | |
| "learning_rate": 5.503719761163907e-07, | |
| "logits/chosen": -0.10350631177425385, | |
| "logits/rejected": -0.08065585047006607, | |
| "logps/chosen": -0.2961386740207672, | |
| "logps/rejected": -0.35747581720352173, | |
| "loss": 1.2338, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.2961386740207672, | |
| "rewards/margins": 0.06133715435862541, | |
| "rewards/rejected": -0.35747581720352173, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.27788911155644624, | |
| "grad_norm": 5.1404239274385315, | |
| "learning_rate": 5.440172833744582e-07, | |
| "logits/chosen": -0.057831160724163055, | |
| "logits/rejected": -0.018619367852807045, | |
| "logps/chosen": -0.33000677824020386, | |
| "logps/rejected": -0.372509241104126, | |
| "loss": 1.2645, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.33000677824020386, | |
| "rewards/margins": 0.04250246286392212, | |
| "rewards/rejected": -0.372509241104126, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.28857715430861725, | |
| "grad_norm": 8.250677302620405, | |
| "learning_rate": 5.373213103969024e-07, | |
| "logits/chosen": -0.09196853637695312, | |
| "logits/rejected": -0.09740312397480011, | |
| "logps/chosen": -0.3195325434207916, | |
| "logps/rejected": -0.38901036977767944, | |
| "loss": 1.2471, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3195325434207916, | |
| "rewards/margins": 0.06947779655456543, | |
| "rewards/rejected": -0.38901036977767944, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.29926519706078825, | |
| "grad_norm": 7.9172215662895065, | |
| "learning_rate": 5.302934221074033e-07, | |
| "logits/chosen": -0.18326355516910553, | |
| "logits/rejected": -0.18124201893806458, | |
| "logps/chosen": -0.35024067759513855, | |
| "logps/rejected": -0.4191233515739441, | |
| "loss": 1.255, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.35024067759513855, | |
| "rewards/margins": 0.06888268887996674, | |
| "rewards/rejected": -0.4191233515739441, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.30995323981295925, | |
| "grad_norm": 5.5553212274649635, | |
| "learning_rate": 5.229434476432182e-07, | |
| "logits/chosen": -0.043330904096364975, | |
| "logits/rejected": -0.06870276480913162, | |
| "logps/chosen": -0.3102904260158539, | |
| "logps/rejected": -0.36457473039627075, | |
| "loss": 1.2393, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.3102904260158539, | |
| "rewards/margins": 0.05428431183099747, | |
| "rewards/rejected": -0.36457473039627075, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.32064128256513025, | |
| "grad_norm": 13.4755681947572, | |
| "learning_rate": 5.152816666082435e-07, | |
| "logits/chosen": -0.0911487489938736, | |
| "logits/rejected": -0.10429541766643524, | |
| "logps/chosen": -0.32351914048194885, | |
| "logps/rejected": -0.4219169616699219, | |
| "loss": 1.2456, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.32351914048194885, | |
| "rewards/margins": 0.09839782118797302, | |
| "rewards/rejected": -0.4219169616699219, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33132932531730125, | |
| "grad_norm": 7.571668933637504, | |
| "learning_rate": 5.073187946960594e-07, | |
| "logits/chosen": -0.08573289960622787, | |
| "logits/rejected": -0.08545240759849548, | |
| "logps/chosen": -0.3177523910999298, | |
| "logps/rejected": -0.3682125210762024, | |
| "loss": 1.2486, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.3177523910999298, | |
| "rewards/margins": 0.05046012997627258, | |
| "rewards/rejected": -0.3682125210762024, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.3420173680694723, | |
| "grad_norm": 6.943663475388935, | |
| "learning_rate": 4.990659687030634e-07, | |
| "logits/chosen": -0.10656943172216415, | |
| "logits/rejected": -0.078646719455719, | |
| "logps/chosen": -0.31297653913497925, | |
| "logps/rejected": -0.37618112564086914, | |
| "loss": 1.2533, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.31297653913497925, | |
| "rewards/margins": 0.06320458650588989, | |
| "rewards/rejected": -0.37618112564086914, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3527054108216433, | |
| "grad_norm": 8.066884155425667, | |
| "learning_rate": 4.905347309526536e-07, | |
| "logits/chosen": -0.06100498512387276, | |
| "logits/rejected": -0.07344032824039459, | |
| "logps/chosen": -0.29639413952827454, | |
| "logps/rejected": -0.36954885721206665, | |
| "loss": 1.2426, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.29639413952827454, | |
| "rewards/margins": 0.0731547400355339, | |
| "rewards/rejected": -0.36954885721206665, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.3633934535738143, | |
| "grad_norm": 5.59599590252963, | |
| "learning_rate": 4.817370131522459e-07, | |
| "logits/chosen": -0.055537961423397064, | |
| "logits/rejected": -0.029491102322936058, | |
| "logps/chosen": -0.3175775408744812, | |
| "logps/rejected": -0.39036422967910767, | |
| "loss": 1.2452, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.3175775408744812, | |
| "rewards/margins": 0.07278666645288467, | |
| "rewards/rejected": -0.39036422967910767, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3740814963259853, | |
| "grad_norm": 5.434006324330732, | |
| "learning_rate": 4.7268511970570207e-07, | |
| "logits/chosen": -0.08873588591814041, | |
| "logits/rejected": -0.08352749794721603, | |
| "logps/chosen": -0.3087335228919983, | |
| "logps/rejected": -0.3652537763118744, | |
| "loss": 1.2509, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.3087335228919983, | |
| "rewards/margins": 0.056520313024520874, | |
| "rewards/rejected": -0.3652537763118744, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3847695390781563, | |
| "grad_norm": 8.227857461961667, | |
| "learning_rate": 4.6339171050450815e-07, | |
| "logits/chosen": -0.10110144317150116, | |
| "logits/rejected": -0.09192384779453278, | |
| "logps/chosen": -0.31839340925216675, | |
| "logps/rejected": -0.35923346877098083, | |
| "loss": 1.2547, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.31839340925216675, | |
| "rewards/margins": 0.04084008187055588, | |
| "rewards/rejected": -0.35923346877098083, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3954575818303273, | |
| "grad_norm": 5.5156889818151456, | |
| "learning_rate": 4.5386978322177184e-07, | |
| "logits/chosen": -0.05175872892141342, | |
| "logits/rejected": -0.07638035714626312, | |
| "logps/chosen": -0.3135964274406433, | |
| "logps/rejected": -0.35415133833885193, | |
| "loss": 1.2628, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.3135964274406433, | |
| "rewards/margins": 0.040554873645305634, | |
| "rewards/rejected": -0.35415133833885193, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.4061456245824983, | |
| "grad_norm": 8.728192233551633, | |
| "learning_rate": 4.4413265513380134e-07, | |
| "logits/chosen": -0.08628302812576294, | |
| "logits/rejected": -0.0637063980102539, | |
| "logps/chosen": -0.3058716654777527, | |
| "logps/rejected": -0.36986014246940613, | |
| "loss": 1.2462, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.3058716654777527, | |
| "rewards/margins": 0.06398848444223404, | |
| "rewards/rejected": -0.36986014246940613, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4168336673346693, | |
| "grad_norm": 8.179877382911195, | |
| "learning_rate": 4.3419394449468975e-07, | |
| "logits/chosen": -0.06323076784610748, | |
| "logits/rejected": -0.03959321230649948, | |
| "logps/chosen": -0.340233713388443, | |
| "logps/rejected": -0.43143948912620544, | |
| "loss": 1.2398, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.340233713388443, | |
| "rewards/margins": 0.09120576083660126, | |
| "rewards/rejected": -0.43143948912620544, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.42752171008684037, | |
| "grad_norm": 5.7289964774382165, | |
| "learning_rate": 4.2406755148995617e-07, | |
| "logits/chosen": -0.03601834177970886, | |
| "logits/rejected": -0.004852446727454662, | |
| "logps/chosen": -0.3104439675807953, | |
| "logps/rejected": -0.38359013199806213, | |
| "loss": 1.2427, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.3104439675807953, | |
| "rewards/margins": 0.07314613461494446, | |
| "rewards/rejected": -0.38359013199806213, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43820975283901137, | |
| "grad_norm": 6.122259545537742, | |
| "learning_rate": 4.1376763879587855e-07, | |
| "logits/chosen": -0.08502549678087234, | |
| "logits/rejected": -0.12322092056274414, | |
| "logps/chosen": -0.3422417938709259, | |
| "logps/rejected": -0.4025643467903137, | |
| "loss": 1.2568, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.3422417938709259, | |
| "rewards/margins": 0.06032254546880722, | |
| "rewards/rejected": -0.4025643467903137, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.44889779559118237, | |
| "grad_norm": 6.481401331907283, | |
| "learning_rate": 4.0330861177171046e-07, | |
| "logits/chosen": -0.09341476112604141, | |
| "logits/rejected": -0.0774766132235527, | |
| "logps/chosen": -0.3205105662345886, | |
| "logps/rejected": -0.38998597860336304, | |
| "loss": 1.2469, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.3205105662345886, | |
| "rewards/margins": 0.06947537511587143, | |
| "rewards/rejected": -0.38998597860336304, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.45958583834335337, | |
| "grad_norm": 7.252453168693919, | |
| "learning_rate": 3.927050983124842e-07, | |
| "logits/chosen": -0.027269473299384117, | |
| "logits/rejected": -0.07347290217876434, | |
| "logps/chosen": -0.297443687915802, | |
| "logps/rejected": -0.3896985948085785, | |
| "loss": 1.2384, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.297443687915802, | |
| "rewards/margins": 0.09225489944219589, | |
| "rewards/rejected": -0.3896985948085785, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.47027388109552437, | |
| "grad_norm": 6.603881159988962, | |
| "learning_rate": 3.8197192839057603e-07, | |
| "logits/chosen": -0.10676763951778412, | |
| "logits/rejected": -0.12236170470714569, | |
| "logps/chosen": -0.30911481380462646, | |
| "logps/rejected": -0.43346795439720154, | |
| "loss": 1.2387, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.30911481380462646, | |
| "rewards/margins": 0.12435313314199448, | |
| "rewards/rejected": -0.43346795439720154, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.48096192384769537, | |
| "grad_norm": 8.47531573536886, | |
| "learning_rate": 3.7112411331464923e-07, | |
| "logits/chosen": -0.027242982760071754, | |
| "logits/rejected": -0.03159039095044136, | |
| "logps/chosen": -0.3138233721256256, | |
| "logps/rejected": -0.3913508951663971, | |
| "loss": 1.2275, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3138233721256256, | |
| "rewards/margins": 0.07752753794193268, | |
| "rewards/rejected": -0.3913508951663971, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4916499665998664, | |
| "grad_norm": 6.314922973802041, | |
| "learning_rate": 3.601768247349818e-07, | |
| "logits/chosen": -0.03554535657167435, | |
| "logits/rejected": -0.08682241290807724, | |
| "logps/chosen": -0.31549689173698425, | |
| "logps/rejected": -0.38394489884376526, | |
| "loss": 1.2408, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.31549689173698425, | |
| "rewards/margins": 0.0684480220079422, | |
| "rewards/rejected": -0.38394489884376526, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5023380093520374, | |
| "grad_norm": 8.005653894417838, | |
| "learning_rate": 3.491453734245413e-07, | |
| "logits/chosen": -0.06854398548603058, | |
| "logits/rejected": -0.019309626892209053, | |
| "logps/chosen": -0.33874207735061646, | |
| "logps/rejected": -0.4499788284301758, | |
| "loss": 1.2411, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.33874207735061646, | |
| "rewards/margins": 0.11123677343130112, | |
| "rewards/rejected": -0.4499788284301758, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.5130260521042084, | |
| "grad_norm": 6.021580320773287, | |
| "learning_rate": 3.3804518786548455e-07, | |
| "logits/chosen": -0.09851398319005966, | |
| "logits/rejected": -0.08007166534662247, | |
| "logps/chosen": -0.3105488419532776, | |
| "logps/rejected": -0.4278073310852051, | |
| "loss": 1.2457, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -0.3105488419532776, | |
| "rewards/margins": 0.11725848913192749, | |
| "rewards/rejected": -0.4278073310852051, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5237140948563794, | |
| "grad_norm": 5.487038572936299, | |
| "learning_rate": 3.2689179267103006e-07, | |
| "logits/chosen": -0.1326679289340973, | |
| "logits/rejected": -0.12296830117702484, | |
| "logps/chosen": -0.3216492533683777, | |
| "logps/rejected": -0.3619733452796936, | |
| "loss": 1.2301, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.3216492533683777, | |
| "rewards/margins": 0.04032406955957413, | |
| "rewards/rejected": -0.3619733452796936, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.5344021376085505, | |
| "grad_norm": 8.070636702857746, | |
| "learning_rate": 3.1570078687288317e-07, | |
| "logits/chosen": -0.07501702755689621, | |
| "logits/rejected": -0.07981151342391968, | |
| "logps/chosen": -0.3418981432914734, | |
| "logps/rejected": -0.4605533480644226, | |
| "loss": 1.2276, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.3418981432914734, | |
| "rewards/margins": 0.11865530908107758, | |
| "rewards/rejected": -0.4605533480644226, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5450901803607214, | |
| "grad_norm": 8.167728040532104, | |
| "learning_rate": 3.0448782210457906e-07, | |
| "logits/chosen": -0.08226356655359268, | |
| "logits/rejected": -0.07389514893293381, | |
| "logps/chosen": -0.36772793531417847, | |
| "logps/rejected": -0.4519001543521881, | |
| "loss": 1.2468, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.36772793531417847, | |
| "rewards/margins": 0.08417223393917084, | |
| "rewards/rejected": -0.4519001543521881, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.5557782231128925, | |
| "grad_norm": 8.9092828172424, | |
| "learning_rate": 2.932685807112585e-07, | |
| "logits/chosen": -0.1377793848514557, | |
| "logits/rejected": -0.13506443798542023, | |
| "logps/chosen": -0.3222098648548126, | |
| "logps/rejected": -0.41552734375, | |
| "loss": 1.238, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.3222098648548126, | |
| "rewards/margins": 0.09331748634576797, | |
| "rewards/rejected": -0.41552734375, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5664662658650634, | |
| "grad_norm": 13.838548351939439, | |
| "learning_rate": 2.8205875381648974e-07, | |
| "logits/chosen": -0.11301400512456894, | |
| "logits/rejected": -0.11296539008617401, | |
| "logps/chosen": -0.3139273226261139, | |
| "logps/rejected": -0.40139150619506836, | |
| "loss": 1.2453, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.3139273226261139, | |
| "rewards/margins": 0.08746419101953506, | |
| "rewards/rejected": -0.40139150619506836, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.5771543086172345, | |
| "grad_norm": 11.456971077488367, | |
| "learning_rate": 2.708740193768135e-07, | |
| "logits/chosen": -0.08372676372528076, | |
| "logits/rejected": -0.07923340797424316, | |
| "logps/chosen": -0.33234184980392456, | |
| "logps/rejected": -0.4971323013305664, | |
| "loss": 1.2275, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -0.33234184980392456, | |
| "rewards/margins": 0.16479046642780304, | |
| "rewards/rejected": -0.4971323013305664, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5878423513694054, | |
| "grad_norm": 5.570703398185759, | |
| "learning_rate": 2.597300202547034e-07, | |
| "logits/chosen": -0.06883351504802704, | |
| "logits/rejected": -0.0933861956000328, | |
| "logps/chosen": -0.324018269777298, | |
| "logps/rejected": -0.36465924978256226, | |
| "loss": 1.2409, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.324018269777298, | |
| "rewards/margins": 0.04064100235700607, | |
| "rewards/rejected": -0.36465924978256226, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5985303941215765, | |
| "grad_norm": 6.1537667241918355, | |
| "learning_rate": 2.4864234234060747e-07, | |
| "logits/chosen": -0.12475994974374771, | |
| "logits/rejected": -0.10665042698383331, | |
| "logps/chosen": -0.3264842629432678, | |
| "logps/rejected": -0.4078393876552582, | |
| "loss": 1.2345, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -0.3264842629432678, | |
| "rewards/margins": 0.08135511726140976, | |
| "rewards/rejected": -0.4078393876552582, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6092184368737475, | |
| "grad_norm": 8.294056815581182, | |
| "learning_rate": 2.3762649275467223e-07, | |
| "logits/chosen": -0.11794719845056534, | |
| "logits/rejected": -0.13079698383808136, | |
| "logps/chosen": -0.32731741666793823, | |
| "logps/rejected": -0.4040375351905823, | |
| "loss": 1.2497, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.32731741666793823, | |
| "rewards/margins": 0.07672014087438583, | |
| "rewards/rejected": -0.4040375351905823, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.6199064796259185, | |
| "grad_norm": 8.79096353154505, | |
| "learning_rate": 2.2669787815863174e-07, | |
| "logits/chosen": -0.032518744468688965, | |
| "logits/rejected": -0.05465514585375786, | |
| "logps/chosen": -0.32134541869163513, | |
| "logps/rejected": -0.4223794937133789, | |
| "loss": 1.2513, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.32134541869163513, | |
| "rewards/margins": 0.10103406012058258, | |
| "rewards/rejected": -0.4223794937133789, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6305945223780896, | |
| "grad_norm": 5.400467561380045, | |
| "learning_rate": 2.1587178320819919e-07, | |
| "logits/chosen": -0.06230146810412407, | |
| "logits/rejected": -0.0027563839685171843, | |
| "logps/chosen": -0.27156487107276917, | |
| "logps/rejected": -0.3670196831226349, | |
| "loss": 1.2348, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.27156487107276917, | |
| "rewards/margins": 0.09545480459928513, | |
| "rewards/rejected": -0.3670196831226349, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.6412825651302605, | |
| "grad_norm": 12.259873359149967, | |
| "learning_rate": 2.0516334917609277e-07, | |
| "logits/chosen": -0.10260840505361557, | |
| "logits/rejected": -0.04962821304798126, | |
| "logps/chosen": -0.336132675409317, | |
| "logps/rejected": -0.5095348358154297, | |
| "loss": 1.2334, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -0.336132675409317, | |
| "rewards/margins": 0.17340223491191864, | |
| "rewards/rejected": -0.5095348358154297, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6519706078824316, | |
| "grad_norm": 8.217920440778814, | |
| "learning_rate": 1.9458755277559716e-07, | |
| "logits/chosen": -0.11686922609806061, | |
| "logits/rejected": -0.11177687346935272, | |
| "logps/chosen": -0.31434646248817444, | |
| "logps/rejected": -0.40717291831970215, | |
| "loss": 1.24, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.31434646248817444, | |
| "rewards/margins": 0.0928264707326889, | |
| "rewards/rejected": -0.40717291831970215, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.6626586506346025, | |
| "grad_norm": 6.161104077211237, | |
| "learning_rate": 1.8415918521427613e-07, | |
| "logits/chosen": -0.18385252356529236, | |
| "logits/rejected": -0.19196629524230957, | |
| "logps/chosen": -0.31113195419311523, | |
| "logps/rejected": -0.3769488036632538, | |
| "loss": 1.2457, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.31113195419311523, | |
| "rewards/margins": 0.06581689417362213, | |
| "rewards/rejected": -0.3769488036632538, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6733466933867736, | |
| "grad_norm": 10.34689683004134, | |
| "learning_rate": 1.7389283150713038e-07, | |
| "logits/chosen": -0.1292671114206314, | |
| "logits/rejected": -0.1229550838470459, | |
| "logps/chosen": -0.35926195979118347, | |
| "logps/rejected": -0.42332392930984497, | |
| "loss": 1.2549, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.35926195979118347, | |
| "rewards/margins": 0.06406201422214508, | |
| "rewards/rejected": -0.42332392930984497, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.6840347361389446, | |
| "grad_norm": 9.698239508672282, | |
| "learning_rate": 1.6380285007813597e-07, | |
| "logits/chosen": -0.11564113199710846, | |
| "logits/rejected": -0.12854574620723724, | |
| "logps/chosen": -0.3144187033176422, | |
| "logps/rejected": -0.3402511179447174, | |
| "loss": 1.2605, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.3144187033176422, | |
| "rewards/margins": 0.02583237923681736, | |
| "rewards/rejected": -0.3402511179447174, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6947227788911156, | |
| "grad_norm": 6.070464304081896, | |
| "learning_rate": 1.539033526786898e-07, | |
| "logits/chosen": -0.1405663937330246, | |
| "logits/rejected": -0.11777285486459732, | |
| "logps/chosen": -0.32500436902046204, | |
| "logps/rejected": -0.4695354402065277, | |
| "loss": 1.2421, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.32500436902046204, | |
| "rewards/margins": 0.14453105628490448, | |
| "rewards/rejected": -0.4695354402065277, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.7054108216432866, | |
| "grad_norm": 5.549203189727758, | |
| "learning_rate": 1.4420818465104924e-07, | |
| "logits/chosen": -0.18301475048065186, | |
| "logits/rejected": -0.18062211573123932, | |
| "logps/chosen": -0.3069472312927246, | |
| "logps/rejected": -0.3610253930091858, | |
| "loss": 1.2325, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3069472312927246, | |
| "rewards/margins": 0.05407816916704178, | |
| "rewards/rejected": -0.3610253930091858, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7160988643954576, | |
| "grad_norm": 6.763068049142248, | |
| "learning_rate": 1.3473090556436928e-07, | |
| "logits/chosen": -0.10250736773014069, | |
| "logits/rejected": -0.11988594383001328, | |
| "logps/chosen": -0.32310950756073, | |
| "logps/rejected": -0.4247487485408783, | |
| "loss": 1.2479, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.32310950756073, | |
| "rewards/margins": 0.10163921117782593, | |
| "rewards/rejected": -0.4247487485408783, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.7267869071476286, | |
| "grad_norm": 6.312494529658719, | |
| "learning_rate": 1.2548477025041833e-07, | |
| "logits/chosen": -0.1759723722934723, | |
| "logits/rejected": -0.15970100462436676, | |
| "logps/chosen": -0.31687361001968384, | |
| "logps/rejected": -0.4269244074821472, | |
| "loss": 1.2473, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -0.31687361001968384, | |
| "rewards/margins": 0.11005084216594696, | |
| "rewards/rejected": -0.4269244074821472, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7374749498997996, | |
| "grad_norm": 7.012548386778781, | |
| "learning_rate": 1.1648271026549805e-07, | |
| "logits/chosen": -0.16853009164333344, | |
| "logits/rejected": -0.1741519272327423, | |
| "logps/chosen": -0.3043145537376404, | |
| "logps/rejected": -0.41543588042259216, | |
| "loss": 1.2355, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.3043145537376404, | |
| "rewards/margins": 0.11112137138843536, | |
| "rewards/rejected": -0.41543588042259216, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.7481629926519706, | |
| "grad_norm": 8.592898323037762, | |
| "learning_rate": 1.0773731580449275e-07, | |
| "logits/chosen": -0.0907156690955162, | |
| "logits/rejected": -0.10430169105529785, | |
| "logps/chosen": -0.33002233505249023, | |
| "logps/rejected": -0.432032972574234, | |
| "loss": 1.2274, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.33002233505249023, | |
| "rewards/margins": 0.10201063007116318, | |
| "rewards/rejected": -0.432032972574234, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7588510354041417, | |
| "grad_norm": 5.346851949829031, | |
| "learning_rate": 9.926081809234262e-08, | |
| "logits/chosen": -0.15248560905456543, | |
| "logits/rejected": -0.14991840720176697, | |
| "logps/chosen": -0.3535235822200775, | |
| "logps/rejected": -0.5069695115089417, | |
| "loss": 1.2334, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3535235822200775, | |
| "rewards/margins": 0.15344594419002533, | |
| "rewards/rejected": -0.5069695115089417, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.7695390781563126, | |
| "grad_norm": 17.058589802109246, | |
| "learning_rate": 9.106507227756998e-08, | |
| "logits/chosen": -0.11075520515441895, | |
| "logits/rejected": -0.11937098205089569, | |
| "logps/chosen": -0.3546099066734314, | |
| "logps/rejected": -0.4121836721897125, | |
| "loss": 1.2538, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.3546099066734314, | |
| "rewards/margins": 0.05757373571395874, | |
| "rewards/rejected": -0.4121836721897125, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7802271209084837, | |
| "grad_norm": 8.71836207461451, | |
| "learning_rate": 8.316154085178256e-08, | |
| "logits/chosen": -0.16542118787765503, | |
| "logits/rejected": -0.17096641659736633, | |
| "logps/chosen": -0.35373011231422424, | |
| "logps/rejected": -0.4567010998725891, | |
| "loss": 1.2417, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.35373011231422424, | |
| "rewards/margins": 0.10297106206417084, | |
| "rewards/rejected": -0.4567010998725891, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.7909151636606546, | |
| "grad_norm": 7.987346992327182, | |
| "learning_rate": 7.55612776183419e-08, | |
| "logits/chosen": -0.1105569452047348, | |
| "logits/rejected": -0.09565238654613495, | |
| "logps/chosen": -0.34828633069992065, | |
| "logps/rejected": -0.4095218777656555, | |
| "loss": 1.2353, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.34828633069992065, | |
| "rewards/margins": 0.06123558431863785, | |
| "rewards/rejected": -0.4095218777656555, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8016032064128257, | |
| "grad_norm": 8.515114879561368, | |
| "learning_rate": 6.827491223262017e-08, | |
| "logits/chosen": -0.1510094702243805, | |
| "logits/rejected": -0.13519130647182465, | |
| "logps/chosen": -0.3320384621620178, | |
| "logps/rejected": -0.39452531933784485, | |
| "loss": 1.2311, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.3320384621620178, | |
| "rewards/margins": 0.06248681992292404, | |
| "rewards/rejected": -0.39452531933784485, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.8122912491649966, | |
| "grad_norm": 7.03157225265326, | |
| "learning_rate": 6.131263533546572e-08, | |
| "logits/chosen": -0.1356309950351715, | |
| "logits/rejected": -0.13989830017089844, | |
| "logps/chosen": -0.3249013423919678, | |
| "logps/rejected": -0.4495007395744324, | |
| "loss": 1.2322, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -0.3249013423919678, | |
| "rewards/margins": 0.12459937483072281, | |
| "rewards/rejected": -0.4495007395744324, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8229792919171677, | |
| "grad_norm": 8.670843552810656, | |
| "learning_rate": 5.468418430067059e-08, | |
| "logits/chosen": -0.13802111148834229, | |
| "logits/rejected": -0.12261234223842621, | |
| "logps/chosen": -0.3400149345397949, | |
| "logps/rejected": -0.4104432165622711, | |
| "loss": 1.2468, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3400149345397949, | |
| "rewards/margins": 0.07042822986841202, | |
| "rewards/rejected": -0.4104432165622711, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.8336673346693386, | |
| "grad_norm": 6.105321944303861, | |
| "learning_rate": 4.839882961637282e-08, | |
| "logits/chosen": -0.1445506513118744, | |
| "logits/rejected": -0.12124904245138168, | |
| "logps/chosen": -0.3267291188240051, | |
| "logps/rejected": -0.3966960906982422, | |
| "loss": 1.2457, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.3267291188240051, | |
| "rewards/margins": 0.06996697932481766, | |
| "rewards/rejected": -0.3966960906982422, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8443553774215097, | |
| "grad_norm": 7.42690529749409, | |
| "learning_rate": 4.2465361919440165e-08, | |
| "logits/chosen": -0.17786213755607605, | |
| "logits/rejected": -0.1693142205476761, | |
| "logps/chosen": -0.3282647728919983, | |
| "logps/rejected": -0.3676055073738098, | |
| "loss": 1.2416, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.3282647728919983, | |
| "rewards/margins": 0.03934071958065033, | |
| "rewards/rejected": -0.3676055073738098, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.8550434201736807, | |
| "grad_norm": 6.448291334596692, | |
| "learning_rate": 3.6892079700970036e-08, | |
| "logits/chosen": -0.19666707515716553, | |
| "logits/rejected": -0.1866413801908493, | |
| "logps/chosen": -0.3443647027015686, | |
| "logps/rejected": -0.3764795660972595, | |
| "loss": 1.2443, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3443647027015686, | |
| "rewards/margins": 0.032114893198013306, | |
| "rewards/rejected": -0.3764795660972595, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8550434201736807, | |
| "eval_logits/chosen": 0.01234325673431158, | |
| "eval_logits/rejected": 0.003141747321933508, | |
| "eval_logps/chosen": -0.33608412742614746, | |
| "eval_logps/rejected": -0.401297003030777, | |
| "eval_loss": 1.241620659828186, | |
| "eval_rewards/accuracies": 0.5914633870124817, | |
| "eval_rewards/chosen": -0.33608412742614746, | |
| "eval_rewards/margins": 0.06521284580230713, | |
| "eval_rewards/rejected": -0.401297003030777, | |
| "eval_runtime": 422.8878, | |
| "eval_samples_per_second": 4.637, | |
| "eval_steps_per_second": 0.291, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8657314629258517, | |
| "grad_norm": 8.756154081277105, | |
| "learning_rate": 3.1686777700099e-08, | |
| "logits/chosen": -0.12114688009023666, | |
| "logits/rejected": -0.17152926325798035, | |
| "logps/chosen": -0.3494204878807068, | |
| "logps/rejected": -0.3755631446838379, | |
| "loss": 1.245, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.3494204878807068, | |
| "rewards/margins": 0.026142627000808716, | |
| "rewards/rejected": -0.3755631446838379, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.8764195056780227, | |
| "grad_norm": 7.899265779506385, | |
| "learning_rate": 2.685673600235524e-08, | |
| "logits/chosen": -0.11135631799697876, | |
| "logits/rejected": -0.13768556714057922, | |
| "logps/chosen": -0.3498903512954712, | |
| "logps/rejected": -0.3871268332004547, | |
| "loss": 1.261, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.3498903512954712, | |
| "rewards/margins": 0.03723648935556412, | |
| "rewards/rejected": -0.3871268332004547, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.8871075484301937, | |
| "grad_norm": 14.01441787078892, | |
| "learning_rate": 2.2408709857800988e-08, | |
| "logits/chosen": -0.12833549082279205, | |
| "logits/rejected": -0.10664300620555878, | |
| "logps/chosen": -0.2865287661552429, | |
| "logps/rejected": -0.38992565870285034, | |
| "loss": 1.2338, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.2865287661552429, | |
| "rewards/margins": 0.10339690744876862, | |
| "rewards/rejected": -0.38992565870285034, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.8977955911823647, | |
| "grad_norm": 5.705006993624562, | |
| "learning_rate": 1.8348920233204167e-08, | |
| "logits/chosen": -0.08621132373809814, | |
| "logits/rejected": -0.06882130354642868, | |
| "logps/chosen": -0.3323257565498352, | |
| "logps/rejected": -0.4798430800437927, | |
| "loss": 1.2399, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.3323257565498352, | |
| "rewards/margins": 0.1475173979997635, | |
| "rewards/rejected": -0.4798430800437927, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9084836339345357, | |
| "grad_norm": 5.012398062093744, | |
| "learning_rate": 1.468304511145394e-08, | |
| "logits/chosen": -0.028029289096593857, | |
| "logits/rejected": -0.07289622724056244, | |
| "logps/chosen": -0.31065744161605835, | |
| "logps/rejected": -0.4257968068122864, | |
| "loss": 1.2356, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -0.31065744161605835, | |
| "rewards/margins": 0.11513932794332504, | |
| "rewards/rejected": -0.4257968068122864, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.9191716766867067, | |
| "grad_norm": 6.051006778161325, | |
| "learning_rate": 1.1416211550388222e-08, | |
| "logits/chosen": -0.11329293251037598, | |
| "logits/rejected": -0.09477666765451431, | |
| "logps/chosen": -0.29777267575263977, | |
| "logps/rejected": -0.3625491261482239, | |
| "loss": 1.2347, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.29777267575263977, | |
| "rewards/margins": 0.0647764801979065, | |
| "rewards/rejected": -0.3625491261482239, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9298597194388778, | |
| "grad_norm": 7.54223521221758, | |
| "learning_rate": 8.552988512139748e-09, | |
| "logits/chosen": -0.10169041156768799, | |
| "logits/rejected": -0.11717716604471207, | |
| "logps/chosen": -0.3231045603752136, | |
| "logps/rejected": -0.44011014699935913, | |
| "loss": 1.2366, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.3231045603752136, | |
| "rewards/margins": 0.11700558662414551, | |
| "rewards/rejected": -0.44011014699935913, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.9405477621910487, | |
| "grad_norm": 17.447935307423492, | |
| "learning_rate": 6.097380473029356e-09, | |
| "logits/chosen": -0.13727042078971863, | |
| "logits/rejected": -0.1495351493358612, | |
| "logps/chosen": -0.3348698019981384, | |
| "logps/rejected": -0.39052897691726685, | |
| "loss": 1.2477, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.3348698019981384, | |
| "rewards/margins": 0.05565913766622543, | |
| "rewards/rejected": -0.39052897691726685, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9512358049432198, | |
| "grad_norm": 7.32756997364182, | |
| "learning_rate": 4.052821822943597e-09, | |
| "logits/chosen": -0.05230356380343437, | |
| "logits/rejected": -0.05104394629597664, | |
| "logps/chosen": -0.31396013498306274, | |
| "logps/rejected": -0.38287192583084106, | |
| "loss": 1.2578, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.31396013498306274, | |
| "rewards/margins": 0.06891177594661713, | |
| "rewards/rejected": -0.38287192583084106, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.9619238476953907, | |
| "grad_norm": 7.458324039594406, | |
| "learning_rate": 2.4221720620301368e-09, | |
| "logits/chosen": -0.11121572554111481, | |
| "logits/rejected": -0.10439705848693848, | |
| "logps/chosen": -0.3217839300632477, | |
| "logps/rejected": -0.4048451781272888, | |
| "loss": 1.2277, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -0.3217839300632477, | |
| "rewards/margins": 0.08306124061346054, | |
| "rewards/rejected": -0.4048451781272888, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9726118904475618, | |
| "grad_norm": 7.12208845758839, | |
| "learning_rate": 1.2077118014282794e-09, | |
| "logits/chosen": -0.06711649894714355, | |
| "logits/rejected": -0.03383985906839371, | |
| "logps/chosen": -0.34118419885635376, | |
| "logps/rejected": -0.4141947329044342, | |
| "loss": 1.2497, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.34118419885635376, | |
| "rewards/margins": 0.07301049679517746, | |
| "rewards/rejected": -0.4141947329044342, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.9832999331997327, | |
| "grad_norm": 6.594294682889427, | |
| "learning_rate": 4.1113957362785e-10, | |
| "logits/chosen": -0.06302393972873688, | |
| "logits/rejected": -0.08739855140447617, | |
| "logps/chosen": -0.3296561539173126, | |
| "logps/rejected": -0.37389543652534485, | |
| "loss": 1.2492, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.3296561539173126, | |
| "rewards/margins": 0.04423931986093521, | |
| "rewards/rejected": -0.37389543652534485, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.9939879759519038, | |
| "grad_norm": 8.269794544806404, | |
| "learning_rate": 3.3569456917970085e-11, | |
| "logits/chosen": -0.036546867340803146, | |
| "logits/rejected": -0.05820956081151962, | |
| "logps/chosen": -0.31292372941970825, | |
| "logps/rejected": -0.4097159504890442, | |
| "loss": 1.2414, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.31292372941970825, | |
| "rewards/margins": 0.09679219126701355, | |
| "rewards/rejected": -0.4097159504890442, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.9982631930527722, | |
| "step": 467, | |
| "total_flos": 0.0, | |
| "train_loss": 1.248079635857004, | |
| "train_runtime": 20917.8016, | |
| "train_samples_per_second": 2.862, | |
| "train_steps_per_second": 0.022 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 467, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |