{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "eval_steps": 500, "global_step": 506, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0, "forget_loss": 0.14035749435424805, "retain_loss": 2.6653778553009033, "step": 0 }, { "epoch": 0, "forget_loss": 0.19487237930297852, "retain_loss": 2.794126033782959, "step": 0 }, { "epoch": 0, "forget_loss": 0.1652039885520935, "retain_loss": 2.7992446422576904, "step": 0 }, { "epoch": 0, "forget_loss": 0.25970637798309326, "retain_loss": 2.7407283782958984, "step": 0 }, { "epoch": 0.003956478733926805, "grad_norm": 10.1875, "learning_rate": 0.0, "loss": 3.700044631958008, "step": 1 }, { "epoch": 0.03956478733926805, "grad_norm": 2.390625, "learning_rate": 4.5e-05, "loss": 2.9271229637993708, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 0.0006051036179997027, "retain_loss": 2.075515031814575, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 0.0002671501424629241, "retain_loss": 2.2641794681549072, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 0.00017286122601944953, "retain_loss": 2.1814846992492676, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 0.0006279755616560578, "retain_loss": 2.482621669769287, "step": 10 }, { "epoch": 0.0791295746785361, "grad_norm": 2.203125, "learning_rate": 4.909274193548387e-05, "loss": 2.1829532623291015, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 3.1555460736854e-05, "retain_loss": 2.056671380996704, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 1.3120857147441711e-05, "retain_loss": 1.9920682907104492, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 8.181996963685378e-05, "retain_loss": 2.149359703063965, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 4.155315400566906e-05, "retain_loss": 1.797939419746399, "step": 20 }, { "epoch": 0.11869436201780416, "grad_norm": 2.03125, "learning_rate": 4.808467741935484e-05, "loss": 1.9520586013793946, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 2.097734068229329e-06, "retain_loss": 1.9717810153961182, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 9.0424384779908e-07, "retain_loss": 2.1347732543945312, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 1.5195576452242676e-05, "retain_loss": 1.7591031789779663, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 2.9381511922110803e-05, "retain_loss": 2.324976682662964, "step": 30 }, { "epoch": 0.1582591493570722, "grad_norm": 1.9453125, "learning_rate": 4.7076612903225806e-05, "loss": 1.7898395538330079, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 2.347593181184493e-05, "retain_loss": 1.5896490812301636, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 2.7579757443163544e-05, "retain_loss": 1.1212265491485596, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 2.1694726456189528e-05, "retain_loss": 2.051854133605957, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 6.128148925199639e-06, "retain_loss": 1.6619952917099, "step": 40 }, { "epoch": 0.19782393669634027, "grad_norm": 1.96875, "learning_rate": 4.606854838709678e-05, "loss": 1.5459610939025878, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 1.4324338735605124e-05, "retain_loss": 1.7004584074020386, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 5.814884843857726e-06, "retain_loss": 1.3355984687805176, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 1.4618005934607936e-06, "retain_loss": 1.7750041484832764, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 1.6271239928755676e-06, "retain_loss": 1.4064531326293945, "step": 50 }, { "epoch": 0.23738872403560832, "grad_norm": 2.234375, "learning_rate": 4.506048387096775e-05, "loss": 1.4617247581481934, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 9.254661563318223e-06, "retain_loss": 1.8761130571365356, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 9.365790901938453e-06, "retain_loss": 1.1389141082763672, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 6.40576909063384e-05, "retain_loss": 1.0727425813674927, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 3.140630724374205e-06, "retain_loss": 1.2311009168624878, "step": 60 }, { "epoch": 0.27695351137487634, "grad_norm": 2.125, "learning_rate": 4.4052419354838714e-05, "loss": 1.344314193725586, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 2.0407464035088196e-05, "retain_loss": 1.2631032466888428, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 2.1732415916630998e-05, "retain_loss": 1.4336204528808594, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 1.3687983937415993e-06, "retain_loss": 0.9825552105903625, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 1.326372967014322e-06, "retain_loss": 1.3998953104019165, "step": 70 }, { "epoch": 0.3165182987141444, "grad_norm": 2.390625, "learning_rate": 4.3044354838709674e-05, "loss": 1.214775276184082, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 1.3552562450058758e-06, "retain_loss": 1.0332047939300537, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 1.2774065908160992e-05, "retain_loss": 1.007059931755066, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 1.196574157802388e-05, "retain_loss": 0.8872216939926147, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 4.2263846466994437e-07, "retain_loss": 0.8739179372787476, "step": 80 }, { "epoch": 0.3560830860534125, "grad_norm": 2.28125, "learning_rate": 4.203629032258065e-05, "loss": 1.1429722785949707, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 2.522195063647814e-05, "retain_loss": 0.8757652640342712, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 2.2732887373422273e-05, "retain_loss": 0.9618234634399414, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 2.1186065168876667e-06, "retain_loss": 0.82012540102005, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 1.5917968994472176e-05, "retain_loss": 1.2091050148010254, "step": 90 }, { "epoch": 0.39564787339268054, "grad_norm": 2.8125, "learning_rate": 4.1028225806451615e-05, "loss": 0.9440973281860352, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 5.102012892166385e-07, "retain_loss": 0.9886119365692139, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 2.069048832709086e-06, "retain_loss": 1.037408471107483, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 6.717852556903381e-06, "retain_loss": 1.2528358697891235, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 1.7514905266580172e-05, "retain_loss": 0.5935210585594177, "step": 100 }, { "epoch": 0.4352126607319486, "grad_norm": 2.34375, "learning_rate": 4.002016129032258e-05, "loss": 0.8624692916870117, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 1.1634347174549475e-05, "retain_loss": 0.5517681837081909, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 6.400139795914583e-08, "retain_loss": 1.0467280149459839, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 1.9029957911698148e-05, "retain_loss": 1.1268657445907593, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 2.5354455601700465e-07, "retain_loss": 0.8402698040008545, "step": 110 }, { "epoch": 0.47477744807121663, "grad_norm": 2.609375, "learning_rate": 3.901209677419355e-05, "loss": 0.7725035190582276, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 3.8098887671367265e-06, "retain_loss": 0.3457094430923462, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 2.3242062525241636e-05, "retain_loss": 0.6856692433357239, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 8.018694643396884e-06, "retain_loss": 1.2206052541732788, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 3.351060968270758e-06, "retain_loss": 0.7469509840011597, "step": 120 }, { "epoch": 0.5143422354104846, "grad_norm": 2.640625, "learning_rate": 3.800403225806452e-05, "loss": 0.6184093952178955, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 3.902082880813396e-06, "retain_loss": 0.9034799933433533, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 1.8487480701878667e-07, "retain_loss": 0.944042980670929, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 7.273596338563948e-07, "retain_loss": 0.589636504650116, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 6.394767115125433e-05, "retain_loss": 0.8495606780052185, "step": 130 }, { "epoch": 0.5539070227497527, "grad_norm": 2.09375, "learning_rate": 3.6995967741935484e-05, "loss": 0.5851613521575928, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 1.9570716176531278e-06, "retain_loss": 0.31179508566856384, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 4.920673745800741e-06, "retain_loss": 0.7174519300460815, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 9.13152689463459e-07, "retain_loss": 0.44045335054397583, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 7.388467793134623e-07, "retain_loss": 0.35985979437828064, "step": 140 }, { "epoch": 0.5934718100890207, "grad_norm": 2.0625, "learning_rate": 3.598790322580645e-05, "loss": 0.5792353630065918, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 1.258710966567378e-07, "retain_loss": 0.45937851071357727, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 7.370726962108165e-07, "retain_loss": 0.27468791604042053, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 8.536089808330871e-06, "retain_loss": 0.8069593906402588, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 8.370562341042387e-07, "retain_loss": 0.15817895531654358, "step": 150 }, { "epoch": 0.6330365974282888, "grad_norm": 1.703125, "learning_rate": 3.497983870967742e-05, "loss": 0.40374002456665037, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 4.5395900087896734e-06, "retain_loss": 0.6503518223762512, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 1.2475591120164609e-06, "retain_loss": 0.29691439867019653, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 7.69290522839583e-07, "retain_loss": 0.3579648435115814, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 3.610326075431658e-06, "retain_loss": 0.3425873816013336, "step": 160 }, { "epoch": 0.6726013847675568, "grad_norm": 2.171875, "learning_rate": 3.397177419354839e-05, "loss": 0.3935901641845703, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 7.11527832208958e-07, "retain_loss": 0.45746102929115295, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 3.808730753007694e-07, "retain_loss": 0.4325888454914093, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 1.3094180530970334e-06, "retain_loss": 0.1893959939479828, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 5.8092759900318924e-06, "retain_loss": 1.1242655515670776, "step": 170 }, { "epoch": 0.712166172106825, "grad_norm": 2.28125, "learning_rate": 3.296370967741936e-05, "loss": 0.37663323879241944, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 1.8580192318040645e-06, "retain_loss": 0.24432934820652008, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 3.5925964425587154e-07, "retain_loss": 0.2534143328666687, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 5.5626283028686885e-06, "retain_loss": 0.23151986300945282, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 5.4275651564239524e-06, "retain_loss": 0.2856011688709259, "step": 180 }, { "epoch": 0.751730959446093, "grad_norm": 2.265625, "learning_rate": 3.1955645161290326e-05, "loss": 0.3224222183227539, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 2.0351612874947023e-06, "retain_loss": 0.30106958746910095, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 1.137744402512908e-05, "retain_loss": 0.35559219121932983, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 4.769389647663047e-07, "retain_loss": 0.12539999186992645, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 1.536158151793643e-07, "retain_loss": 0.5091521143913269, "step": 190 }, { "epoch": 0.7912957467853611, "grad_norm": 1.984375, "learning_rate": 3.0947580645161286e-05, "loss": 0.32691752910614014, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 4.226240889693145e-06, "retain_loss": 0.13701662421226501, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 1.258511565538356e-05, "retain_loss": 0.10310296714305878, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 1.963040858754539e-06, "retain_loss": 0.08762754499912262, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 1.235385980180581e-06, "retain_loss": 0.2134196013212204, "step": 200 }, { "epoch": 0.8308605341246291, "grad_norm": 1.765625, "learning_rate": 2.9939516129032256e-05, "loss": 0.23421640396118165, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 3.501550236251205e-05, "retain_loss": 0.17837883532047272, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 1.5083353446243564e-06, "retain_loss": 0.1773965209722519, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 1.7007857877615606e-06, "retain_loss": 0.11213231086730957, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 5.62718128094275e-07, "retain_loss": 0.10278806835412979, "step": 210 }, { "epoch": 0.8704253214638972, "grad_norm": 2.53125, "learning_rate": 2.8931451612903227e-05, "loss": 0.20826613903045654, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 5.792419415229233e-07, "retain_loss": 0.06651624292135239, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 1.4317017758003203e-06, "retain_loss": 0.3651600480079651, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 2.790197299873398e-07, "retain_loss": 0.059690482914447784, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 2.358171514060814e-06, "retain_loss": 0.07482323050498962, "step": 220 }, { "epoch": 0.9099901088031652, "grad_norm": 2.015625, "learning_rate": 2.7923387096774194e-05, "loss": 0.19134424924850463, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 1.6788119410193758e-06, "retain_loss": 0.11503209918737411, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 1.3623559880215907e-06, "retain_loss": 0.09323878586292267, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 9.115094599110307e-07, "retain_loss": 0.2667946219444275, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 1.176637056232721e-06, "retain_loss": 0.09373427927494049, "step": 230 }, { "epoch": 0.9495548961424333, "grad_norm": 1.890625, "learning_rate": 2.6915322580645164e-05, "loss": 0.19541234970092775, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 4.323494522395777e-06, "retain_loss": 0.060076676309108734, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 3.29818294630968e-06, "retain_loss": 0.1765175312757492, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 1.1875378902459488e-07, "retain_loss": 0.04277701675891876, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 2.5362205633427948e-06, "retain_loss": 0.05752837285399437, "step": 240 }, { "epoch": 0.9891196834817013, "grad_norm": 1.4453125, "learning_rate": 2.590725806451613e-05, "loss": 0.17172594070434571, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 2.3277580112335272e-06, "retain_loss": 0.08483278006315231, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 5.033873094362207e-06, "retain_loss": 0.24280749261379242, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 4.555405212158803e-06, "retain_loss": 0.4092475473880768, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 4.822932737624797e-07, "retain_loss": 0.062368009239435196, "step": 250 }, { "epoch": 1.0276953511374876, "grad_norm": 2.078125, "learning_rate": 2.48991935483871e-05, "loss": 0.14231110811233522, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 2.3665279513807036e-06, "retain_loss": 0.04678928107023239, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 1.7857011016531033e-06, "retain_loss": 0.15241962671279907, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 3.6310279938334133e-07, "retain_loss": 0.20368625223636627, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 5.83898440709163e-07, "retain_loss": 0.06002001464366913, "step": 260 }, { "epoch": 1.0672601384767557, "grad_norm": 1.4921875, "learning_rate": 2.3891129032258066e-05, "loss": 0.12347257137298584, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 6.095945082051912e-07, "retain_loss": 0.2563520669937134, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 7.611353112224606e-07, "retain_loss": 0.23337779939174652, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 1.0488522548257606e-06, "retain_loss": 0.0406014546751976, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 1.4446818568103481e-06, "retain_loss": 0.03464161604642868, "step": 270 }, { "epoch": 1.1068249258160237, "grad_norm": 0.97265625, "learning_rate": 2.2883064516129033e-05, "loss": 0.12490516901016235, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 2.0972819925191288e-07, "retain_loss": 0.038518667221069336, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 2.309284582224791e-06, "retain_loss": 0.05753574147820473, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 4.084875797616405e-07, "retain_loss": 0.10644430667161942, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 1.1346749317908689e-08, "retain_loss": 0.11500366032123566, "step": 280 }, { "epoch": 1.1463897131552918, "grad_norm": 1.109375, "learning_rate": 2.1875e-05, "loss": 0.09100213646888733, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 4.1612356653786264e-06, "retain_loss": 0.07052873820066452, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 9.997415872931015e-08, "retain_loss": 0.04744183272123337, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 1.2092109500372317e-06, "retain_loss": 0.07089261710643768, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 2.131706423824653e-06, "retain_loss": 0.2522507607936859, "step": 290 }, { "epoch": 1.1859545004945597, "grad_norm": 1.2265625, "learning_rate": 2.086693548387097e-05, "loss": 0.09003463983535767, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 1.1781303328461945e-06, "retain_loss": 0.02843177132308483, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 2.6735065716820827e-07, "retain_loss": 0.04060671105980873, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 3.643542186182458e-06, "retain_loss": 0.0628509372472763, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 4.7639906597396475e-07, "retain_loss": 0.03408809378743172, "step": 300 }, { "epoch": 1.225519287833828, "grad_norm": 0.74609375, "learning_rate": 1.9858870967741934e-05, "loss": 0.07631087303161621, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 1.8324541883885104e-07, "retain_loss": 0.1489202380180359, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 1.505328555140295e-06, "retain_loss": 0.09023259580135345, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 1.2370117019600002e-06, "retain_loss": 0.11567945778369904, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 6.62587012811855e-07, "retain_loss": 0.07907626777887344, "step": 310 }, { "epoch": 1.265084075173096, "grad_norm": 1.5546875, "learning_rate": 1.8850806451612904e-05, "loss": 0.09484220743179321, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 6.137950094853295e-06, "retain_loss": 0.0641757920384407, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 2.837170541170053e-06, "retain_loss": 0.0350993350148201, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 1.2725810449865094e-07, "retain_loss": 0.04542815685272217, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 4.864556899519812e-07, "retain_loss": 0.05786221846938133, "step": 320 }, { "epoch": 1.304648862512364, "grad_norm": 1.328125, "learning_rate": 1.784274193548387e-05, "loss": 0.06601499915122985, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 5.855958420397656e-07, "retain_loss": 0.04558321461081505, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 1.0734013358160155e-06, "retain_loss": 0.061824798583984375, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 3.7222815763016115e-07, "retain_loss": 0.04270363599061966, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 6.617690360144479e-07, "retain_loss": 0.16833367943763733, "step": 330 }, { "epoch": 1.344213649851632, "grad_norm": 0.49609375, "learning_rate": 1.683467741935484e-05, "loss": 0.06903731822967529, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 1.1879722023877548e-06, "retain_loss": 0.02958919107913971, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 4.095278995919216e-07, "retain_loss": 0.04107590764760971, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 7.392562224595167e-07, "retain_loss": 0.404009610414505, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 2.7393613777348946e-07, "retain_loss": 0.02813897468149662, "step": 340 }, { "epoch": 1.3837784371909, "grad_norm": 1.4140625, "learning_rate": 1.5826612903225806e-05, "loss": 0.069232177734375, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 6.865487307550211e-07, "retain_loss": 0.02693028561770916, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 4.187642161923577e-07, "retain_loss": 0.04608120024204254, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 8.139955554042899e-08, "retain_loss": 0.06439628452062607, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 8.785929139776272e-08, "retain_loss": 0.17509031295776367, "step": 350 }, { "epoch": 1.4233432245301683, "grad_norm": 0.7265625, "learning_rate": 1.4818548387096776e-05, "loss": 0.07515097260475159, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 1.5587008874717867e-06, "retain_loss": 0.027441687881946564, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 2.0895599561754352e-07, "retain_loss": 0.022471986711025238, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 6.960410701140063e-07, "retain_loss": 0.0480155348777771, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 1.2851671726821223e-07, "retain_loss": 0.02052563615143299, "step": 360 }, { "epoch": 1.4629080118694362, "grad_norm": 0.75390625, "learning_rate": 1.3810483870967741e-05, "loss": 0.055222952365875246, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 9.040416557581921e-07, "retain_loss": 0.03805703669786453, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 2.56950727361982e-07, "retain_loss": 0.04338216036558151, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 1.555378958073561e-06, "retain_loss": 0.027456099167466164, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 1.5528239600826055e-06, "retain_loss": 0.11473647505044937, "step": 370 }, { "epoch": 1.5024727992087041, "grad_norm": 0.82421875, "learning_rate": 1.280241935483871e-05, "loss": 0.05010308623313904, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 6.783013759559253e-07, "retain_loss": 0.044970110058784485, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 5.374618922360241e-06, "retain_loss": 0.02438696101307869, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 2.9677858037757687e-07, "retain_loss": 0.023668058216571808, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 1.109128611176402e-08, "retain_loss": 0.04492601007223129, "step": 380 }, { "epoch": 1.5420375865479723, "grad_norm": 0.56640625, "learning_rate": 1.1794354838709679e-05, "loss": 0.05652405023574829, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 8.616879654255172e-07, "retain_loss": 0.0335644893348217, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 6.417557187887724e-07, "retain_loss": 0.03396674245595932, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 1.2950030736647022e-07, "retain_loss": 0.04779934883117676, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 3.2295685059580137e-07, "retain_loss": 0.022925786674022675, "step": 390 }, { "epoch": 1.5816023738872405, "grad_norm": 0.92578125, "learning_rate": 1.0786290322580646e-05, "loss": 0.0499935507774353, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 1.4149179605738027e-07, "retain_loss": 0.09424184262752533, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 2.579657198964469e-08, "retain_loss": 0.022774428129196167, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 9.055776899913326e-06, "retain_loss": 0.03912818431854248, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 5.612932341136911e-07, "retain_loss": 0.01830551028251648, "step": 400 }, { "epoch": 1.6211671612265084, "grad_norm": 0.41796875, "learning_rate": 9.778225806451613e-06, "loss": 0.05245124697685242, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 1.0918882935584406e-06, "retain_loss": 0.03899680823087692, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 4.972907845512964e-07, "retain_loss": 0.036254264414310455, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 2.1247342374408618e-06, "retain_loss": 0.03961007297039032, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 4.669371378440701e-08, "retain_loss": 0.14862912893295288, "step": 410 }, { "epoch": 1.6607319485657763, "grad_norm": 0.51953125, "learning_rate": 8.770161290322582e-06, "loss": 0.04853116869926453, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 2.079215306594051e-07, "retain_loss": 0.0473225861787796, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 1.4471823305939324e-05, "retain_loss": 0.04140442609786987, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 4.412034684264654e-07, "retain_loss": 0.022193601354956627, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 2.4310432422680606e-07, "retain_loss": 0.02164550870656967, "step": 420 }, { "epoch": 1.7002967359050445, "grad_norm": 0.5390625, "learning_rate": 7.762096774193549e-06, "loss": 0.04437119960784912, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 7.294415809155907e-07, "retain_loss": 0.04582787677645683, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 1.6502593780387542e-07, "retain_loss": 0.02343440242111683, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 2.4713807533771615e-07, "retain_loss": 0.027170918881893158, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 1.208927358220535e-07, "retain_loss": 0.05825173109769821, "step": 430 }, { "epoch": 1.7398615232443126, "grad_norm": 1.0078125, "learning_rate": 6.754032258064516e-06, "loss": 0.04427404403686523, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 2.307926251887693e-06, "retain_loss": 0.047143761068582535, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 5.785019538961933e-07, "retain_loss": 0.026861390098929405, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 5.3523475429528844e-08, "retain_loss": 0.03771151602268219, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 1.33451692363451e-07, "retain_loss": 0.030054600909352303, "step": 440 }, { "epoch": 1.7794263105835806, "grad_norm": 0.41796875, "learning_rate": 5.745967741935484e-06, "loss": 0.05158230662345886, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 7.282965270860586e-06, "retain_loss": 0.02785342186689377, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 6.616839982598322e-07, "retain_loss": 0.03133387118577957, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 1.6606659869466966e-07, "retain_loss": 0.028537921607494354, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 4.580670065479353e-08, "retain_loss": 0.0259918924421072, "step": 450 }, { "epoch": 1.8189910979228485, "grad_norm": 0.5703125, "learning_rate": 4.737903225806452e-06, "loss": 0.039455512166023256, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 6.603767133128713e-07, "retain_loss": 0.03985912352800369, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 3.943401907235966e-07, "retain_loss": 0.04197068139910698, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 1.5381397133751307e-07, "retain_loss": 0.032300617545843124, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 4.2404886357871874e-07, "retain_loss": 0.05860987678170204, "step": 460 }, { "epoch": 1.8585558852621167, "grad_norm": 0.73046875, "learning_rate": 3.7298387096774197e-06, "loss": 0.03943937718868255, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 1.0874501299440453e-07, "retain_loss": 0.034533917903900146, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 1.2879773692020535e-07, "retain_loss": 0.02758140303194523, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 4.720429387816694e-06, "retain_loss": 0.025139721110463142, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 6.679280772914353e-08, "retain_loss": 0.014747953973710537, "step": 470 }, { "epoch": 1.8981206726013848, "grad_norm": 0.396484375, "learning_rate": 2.721774193548387e-06, "loss": 0.03879976570606232, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 2.390359213677584e-06, "retain_loss": 0.17955298721790314, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 1.1778739690271323e-06, "retain_loss": 0.03140717372298241, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 8.414248213739484e-07, "retain_loss": 0.024419624358415604, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 1.882670517261431e-07, "retain_loss": 0.04787850007414818, "step": 480 }, { "epoch": 1.9376854599406528, "grad_norm": 0.69140625, "learning_rate": 1.7137096774193548e-06, "loss": 0.04431060552597046, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 6.971927746235451e-07, "retain_loss": 0.056132722645998, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 1.3472174487105804e-06, "retain_loss": 0.05203668773174286, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 1.8118841182968026e-07, "retain_loss": 0.02416139468550682, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 9.088048500416335e-07, "retain_loss": 0.04348018392920494, "step": 490 }, { "epoch": 1.9772502472799207, "grad_norm": 0.40625, "learning_rate": 7.056451612903225e-07, "loss": 0.038701888918876645, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 3.1423081736647873e-07, "retain_loss": 0.025279447436332703, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 2.0392130295476818e-07, "retain_loss": 0.08473268151283264, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 6.266716212621759e-08, "retain_loss": 0.03677506372332573, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 3.635840073457075e-07, "retain_loss": 0.02540448121726513, "step": 500 } ], "logging_steps": 10, "max_steps": 506, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 200, "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": 8, "trial_name": null, "trial_params": null }