diff --git "a/checkpoint-1600/trainer_state.json" "b/checkpoint-1600/trainer_state.json" new file mode 100644--- /dev/null +++ "b/checkpoint-1600/trainer_state.json" @@ -0,0 +1,5001 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.8327605956471937, + "eval_steps": 500, + "global_step": 1600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0, + "forget_loss": 3.44928662343591e-06, + "retain_loss": 6.609266757965088, + "step": 0 + }, + { + "epoch": 0, + "forget_loss": 0.038371894508600235, + "retain_loss": 6.274988651275635, + "step": 0 + }, + { + "epoch": 0, + "forget_loss": 0.001581120421178639, + "retain_loss": 6.232261657714844, + "step": 0 + }, + { + "epoch": 0, + "forget_loss": 8.447099389741197e-06, + "retain_loss": 6.230651378631592, + "step": 0 + }, + { + "epoch": 0.001145475372279496, + "grad_norm": 282.0, + "learning_rate": 0.0, + "loss": 6.386748313903809, + "step": 1 + }, + { + "epoch": 0.011454753722794959, + "grad_norm": 9.4375, + "learning_rate": 4.5e-05, + "loss": 3.339502970377604, + "step": 10 + }, + { + "epoch": 0.011454753722794959, + "forget_loss": 2.795343243633397e-05, + "retain_loss": 1.8692482709884644, + "step": 10 + }, + { + "epoch": 0.011454753722794959, + "forget_loss": 5.640117888106033e-05, + "retain_loss": 2.053846597671509, + "step": 10 + }, + { + "epoch": 0.011454753722794959, + "forget_loss": 3.110460966126993e-05, + "retain_loss": 2.0232813358306885, + "step": 10 + }, + { + "epoch": 0.011454753722794959, + "forget_loss": 4.845933744945796e-06, + "retain_loss": 2.4105024337768555, + "step": 10 + }, + { + "epoch": 0.022909507445589918, + "grad_norm": 38.5, + "learning_rate": 4.9740783410138256e-05, + "loss": 2.1140087127685545, + "step": 20 + }, + { + "epoch": 0.022909507445589918, + "forget_loss": 1.082867765944684e-05, + "retain_loss": 2.002471446990967, + "step": 20 + }, + { + "epoch": 0.022909507445589918, + "forget_loss": 3.829125489573926e-06, + "retain_loss": 2.0777769088745117, + "step": 20 + }, + { + "epoch": 0.022909507445589918, + "forget_loss": 2.8280664992053062e-05, + "retain_loss": 2.1336162090301514, + "step": 20 + }, + { + "epoch": 0.022909507445589918, + "forget_loss": 1.5830282791284844e-05, + "retain_loss": 1.7852996587753296, + "step": 20 + }, + { + "epoch": 0.03436426116838488, + "grad_norm": 5.6875, + "learning_rate": 4.945276497695853e-05, + "loss": 1.8668533325195313, + "step": 30 + }, + { + "epoch": 0.03436426116838488, + "forget_loss": 1.3028584362473339e-05, + "retain_loss": 1.870590090751648, + "step": 30 + }, + { + "epoch": 0.03436426116838488, + "forget_loss": 9.773620149644557e-06, + "retain_loss": 2.1787960529327393, + "step": 30 + }, + { + "epoch": 0.03436426116838488, + "forget_loss": 2.0277722796890885e-05, + "retain_loss": 1.7472785711288452, + "step": 30 + }, + { + "epoch": 0.03436426116838488, + "forget_loss": 1.0735096111602616e-05, + "retain_loss": 2.227604866027832, + "step": 30 + }, + { + "epoch": 0.045819014891179836, + "grad_norm": 5.4375, + "learning_rate": 4.916474654377881e-05, + "loss": 1.695387840270996, + "step": 40 + }, + { + "epoch": 0.045819014891179836, + "forget_loss": 5.423281436378602e-06, + "retain_loss": 1.2835420370101929, + "step": 40 + }, + { + "epoch": 0.045819014891179836, + "forget_loss": 9.37562845138018e-07, + "retain_loss": 0.8797236680984497, + "step": 40 + }, + { + "epoch": 0.045819014891179836, + "forget_loss": 4.068289854330942e-06, + "retain_loss": 1.9588595628738403, + "step": 40 + }, + { + "epoch": 0.045819014891179836, + "forget_loss": 2.925765329564456e-06, + "retain_loss": 1.7313635349273682, + "step": 40 + }, + { + "epoch": 0.0572737686139748, + "grad_norm": 24.5, + "learning_rate": 4.887672811059908e-05, + "loss": 1.3742843627929688, + "step": 50 + }, + { + "epoch": 0.0572737686139748, + "forget_loss": 1.3801378372590989e-05, + "retain_loss": 1.4735946655273438, + "step": 50 + }, + { + "epoch": 0.0572737686139748, + "forget_loss": 8.779801646596752e-06, + "retain_loss": 1.0075594186782837, + "step": 50 + }, + { + "epoch": 0.0572737686139748, + "forget_loss": 1.5394653019029647e-05, + "retain_loss": 1.8036051988601685, + "step": 50 + }, + { + "epoch": 0.0572737686139748, + "forget_loss": 8.239669114118442e-06, + "retain_loss": 1.379029393196106, + "step": 50 + }, + { + "epoch": 0.06872852233676977, + "grad_norm": 4.375, + "learning_rate": 4.858870967741936e-05, + "loss": 1.3121252059936523, + "step": 60 + }, + { + "epoch": 0.06872852233676977, + "forget_loss": 3.8279208638414275e-06, + "retain_loss": 1.7327319383621216, + "step": 60 + }, + { + "epoch": 0.06872852233676977, + "forget_loss": 3.3966937280638376e-06, + "retain_loss": 1.0556788444519043, + "step": 60 + }, + { + "epoch": 0.06872852233676977, + "forget_loss": 3.698090495163342e-06, + "retain_loss": 0.9602264761924744, + "step": 60 + }, + { + "epoch": 0.06872852233676977, + "forget_loss": 4.131216337555088e-05, + "retain_loss": 1.1526304483413696, + "step": 60 + }, + { + "epoch": 0.08018327605956473, + "grad_norm": 4.5, + "learning_rate": 4.830069124423963e-05, + "loss": 1.1931480407714843, + "step": 70 + }, + { + "epoch": 0.08018327605956473, + "forget_loss": 3.9989358810998965e-06, + "retain_loss": 1.1984384059906006, + "step": 70 + }, + { + "epoch": 0.08018327605956473, + "forget_loss": 2.2779448045184836e-05, + "retain_loss": 1.3697820901870728, + "step": 70 + }, + { + "epoch": 0.08018327605956473, + "forget_loss": 7.756965715088882e-06, + "retain_loss": 0.8010083436965942, + "step": 70 + }, + { + "epoch": 0.08018327605956473, + "forget_loss": 1.9700873963302e-06, + "retain_loss": 1.1254682540893555, + "step": 70 + }, + { + "epoch": 0.09163802978235967, + "grad_norm": 12.625, + "learning_rate": 4.801267281105991e-05, + "loss": 1.0133052825927735, + "step": 80 + }, + { + "epoch": 0.09163802978235967, + "forget_loss": 7.642759555892553e-06, + "retain_loss": 0.8252509236335754, + "step": 80 + }, + { + "epoch": 0.09163802978235967, + "forget_loss": 8.59803822095273e-06, + "retain_loss": 0.7506107687950134, + "step": 80 + }, + { + "epoch": 0.09163802978235967, + "forget_loss": 4.804069703823188e-06, + "retain_loss": 0.7315255403518677, + "step": 80 + }, + { + "epoch": 0.09163802978235967, + "forget_loss": 4.970052486896748e-06, + "retain_loss": 0.7012558579444885, + "step": 80 + }, + { + "epoch": 0.10309278350515463, + "grad_norm": 4.0625, + "learning_rate": 4.772465437788018e-05, + "loss": 0.9705428123474121, + "step": 90 + }, + { + "epoch": 0.10309278350515463, + "forget_loss": 1.4775166619074298e-06, + "retain_loss": 0.7782251834869385, + "step": 90 + }, + { + "epoch": 0.10309278350515463, + "forget_loss": 1.5886946584942052e-06, + "retain_loss": 0.8455419540405273, + "step": 90 + }, + { + "epoch": 0.10309278350515463, + "forget_loss": 2.0495062926784158e-06, + "retain_loss": 0.6329978704452515, + "step": 90 + }, + { + "epoch": 0.10309278350515463, + "forget_loss": 3.828020635410212e-05, + "retain_loss": 0.8053420186042786, + "step": 90 + }, + { + "epoch": 0.1145475372279496, + "grad_norm": 11.0, + "learning_rate": 4.743663594470046e-05, + "loss": 0.8036625862121582, + "step": 100 + }, + { + "epoch": 0.1145475372279496, + "forget_loss": 4.424137387104565e-06, + "retain_loss": 0.949962317943573, + "step": 100 + }, + { + "epoch": 0.1145475372279496, + "forget_loss": 2.453501565469196e-06, + "retain_loss": 0.8932625651359558, + "step": 100 + }, + { + "epoch": 0.1145475372279496, + "forget_loss": 1.6591504845564486e-06, + "retain_loss": 1.121445894241333, + "step": 100 + }, + { + "epoch": 0.1145475372279496, + "forget_loss": 4.8642609726812225e-06, + "retain_loss": 0.37143388390541077, + "step": 100 + }, + { + "epoch": 0.12600229095074456, + "grad_norm": 3.453125, + "learning_rate": 4.7148617511520735e-05, + "loss": 0.6938334941864014, + "step": 110 + }, + { + "epoch": 0.12600229095074456, + "forget_loss": 4.921608706354164e-06, + "retain_loss": 0.3767385482788086, + "step": 110 + }, + { + "epoch": 0.12600229095074456, + "forget_loss": 4.79454456581152e-07, + "retain_loss": 0.6782515048980713, + "step": 110 + }, + { + "epoch": 0.12600229095074456, + "forget_loss": 3.918839865946211e-06, + "retain_loss": 0.8204460144042969, + "step": 110 + }, + { + "epoch": 0.12600229095074456, + "forget_loss": 4.693280516221421e-06, + "retain_loss": 0.841111958026886, + "step": 110 + }, + { + "epoch": 0.13745704467353953, + "grad_norm": 3.4375, + "learning_rate": 4.6860599078341014e-05, + "loss": 0.6417326927185059, + "step": 120 + }, + { + "epoch": 0.13745704467353953, + "forget_loss": 1.713265555736143e-05, + "retain_loss": 0.3679219186306, + "step": 120 + }, + { + "epoch": 0.13745704467353953, + "forget_loss": 8.306015843118075e-06, + "retain_loss": 0.6136802434921265, + "step": 120 + }, + { + "epoch": 0.13745704467353953, + "forget_loss": 2.449452495056903e-06, + "retain_loss": 1.2731798887252808, + "step": 120 + }, + { + "epoch": 0.13745704467353953, + "forget_loss": 4.083998192072613e-06, + "retain_loss": 0.5807819962501526, + "step": 120 + }, + { + "epoch": 0.14891179839633448, + "grad_norm": 4.03125, + "learning_rate": 4.657258064516129e-05, + "loss": 0.5295979499816894, + "step": 130 + }, + { + "epoch": 0.14891179839633448, + "forget_loss": 8.430975867668167e-06, + "retain_loss": 0.7869871854782104, + "step": 130 + }, + { + "epoch": 0.14891179839633448, + "forget_loss": 9.652731023379602e-06, + "retain_loss": 0.9147917628288269, + "step": 130 + }, + { + "epoch": 0.14891179839633448, + "forget_loss": 1.1524152796482667e-06, + "retain_loss": 0.48021697998046875, + "step": 130 + }, + { + "epoch": 0.14891179839633448, + "forget_loss": 2.4387713892792817e-06, + "retain_loss": 0.8574126362800598, + "step": 130 + }, + { + "epoch": 0.16036655211912945, + "grad_norm": 4.5625, + "learning_rate": 4.628456221198157e-05, + "loss": 0.5009561061859131, + "step": 140 + }, + { + "epoch": 0.16036655211912945, + "forget_loss": 5.176518243388273e-06, + "retain_loss": 0.2740374803543091, + "step": 140 + }, + { + "epoch": 0.16036655211912945, + "forget_loss": 4.7920007091306616e-06, + "retain_loss": 0.6662377119064331, + "step": 140 + }, + { + "epoch": 0.16036655211912945, + "forget_loss": 2.156430582544999e-06, + "retain_loss": 0.3841518461704254, + "step": 140 + }, + { + "epoch": 0.16036655211912945, + "forget_loss": 1.6250938642770052e-05, + "retain_loss": 0.3435481786727905, + "step": 140 + }, + { + "epoch": 0.1718213058419244, + "grad_norm": 2.9375, + "learning_rate": 4.599654377880185e-05, + "loss": 0.5061944961547852, + "step": 150 + }, + { + "epoch": 0.1718213058419244, + "forget_loss": 2.413504262221977e-06, + "retain_loss": 0.4674696624279022, + "step": 150 + }, + { + "epoch": 0.1718213058419244, + "forget_loss": 9.788258466869593e-06, + "retain_loss": 0.32738229632377625, + "step": 150 + }, + { + "epoch": 0.1718213058419244, + "forget_loss": 1.659399458731059e-06, + "retain_loss": 0.4860354959964752, + "step": 150 + }, + { + "epoch": 0.1718213058419244, + "forget_loss": 6.014466293891019e-07, + "retain_loss": 0.25174200534820557, + "step": 150 + }, + { + "epoch": 0.18327605956471935, + "grad_norm": 3.140625, + "learning_rate": 4.5708525345622124e-05, + "loss": 0.33492887020111084, + "step": 160 + }, + { + "epoch": 0.18327605956471935, + "forget_loss": 4.615132638718933e-06, + "retain_loss": 0.30095958709716797, + "step": 160 + }, + { + "epoch": 0.18327605956471935, + "forget_loss": 1.7731171055856976e-06, + "retain_loss": 0.19373854994773865, + "step": 160 + }, + { + "epoch": 0.18327605956471935, + "forget_loss": 2.2210085717233596e-06, + "retain_loss": 0.23821769654750824, + "step": 160 + }, + { + "epoch": 0.18327605956471935, + "forget_loss": 1.3112939996062778e-06, + "retain_loss": 0.3393084406852722, + "step": 160 + }, + { + "epoch": 0.19473081328751432, + "grad_norm": 2.671875, + "learning_rate": 4.54205069124424e-05, + "loss": 0.3298675775527954, + "step": 170 + }, + { + "epoch": 0.19473081328751432, + "forget_loss": 4.927760528516956e-06, + "retain_loss": 0.44287964701652527, + "step": 170 + }, + { + "epoch": 0.19473081328751432, + "forget_loss": 4.030034688184969e-06, + "retain_loss": 0.4538065195083618, + "step": 170 + }, + { + "epoch": 0.19473081328751432, + "forget_loss": 2.3950892682478298e-06, + "retain_loss": 0.31112024188041687, + "step": 170 + }, + { + "epoch": 0.19473081328751432, + "forget_loss": 2.15224758903787e-06, + "retain_loss": 0.9548900127410889, + "step": 170 + }, + { + "epoch": 0.20618556701030927, + "grad_norm": 2.953125, + "learning_rate": 4.5132488479262676e-05, + "loss": 0.3629490852355957, + "step": 180 + }, + { + "epoch": 0.20618556701030927, + "forget_loss": 7.814773198333569e-06, + "retain_loss": 0.2744402289390564, + "step": 180 + }, + { + "epoch": 0.20618556701030927, + "forget_loss": 9.177201150123437e-07, + "retain_loss": 0.16577409207820892, + "step": 180 + }, + { + "epoch": 0.20618556701030927, + "forget_loss": 2.789876816677861e-06, + "retain_loss": 0.26117604970932007, + "step": 180 + }, + { + "epoch": 0.20618556701030927, + "forget_loss": 3.266710336902179e-06, + "retain_loss": 0.2821800708770752, + "step": 180 + }, + { + "epoch": 0.21764032073310424, + "grad_norm": 3.125, + "learning_rate": 4.4844470046082955e-05, + "loss": 0.3086943864822388, + "step": 190 + }, + { + "epoch": 0.21764032073310424, + "forget_loss": 2.877425231417874e-06, + "retain_loss": 0.4397953450679779, + "step": 190 + }, + { + "epoch": 0.21764032073310424, + "forget_loss": 5.625155608868226e-06, + "retain_loss": 0.40479227900505066, + "step": 190 + }, + { + "epoch": 0.21764032073310424, + "forget_loss": 7.261318387463689e-06, + "retain_loss": 0.08186227083206177, + "step": 190 + }, + { + "epoch": 0.21764032073310424, + "forget_loss": 1.0345914915887988e-06, + "retain_loss": 0.5473820567131042, + "step": 190 + }, + { + "epoch": 0.2290950744558992, + "grad_norm": 2.59375, + "learning_rate": 4.455645161290323e-05, + "loss": 0.30754408836364744, + "step": 200 + }, + { + "epoch": 0.2290950744558992, + "forget_loss": 1.8709645246417494e-06, + "retain_loss": 0.15519601106643677, + "step": 200 + }, + { + "epoch": 0.2290950744558992, + "forget_loss": 5.221798460297578e-07, + "retain_loss": 0.0977967157959938, + "step": 200 + }, + { + "epoch": 0.2290950744558992, + "forget_loss": 4.533095307124313e-06, + "retain_loss": 0.12215632200241089, + "step": 200 + }, + { + "epoch": 0.2290950744558992, + "forget_loss": 2.3582092580909375e-06, + "retain_loss": 0.14002011716365814, + "step": 200 + }, + { + "epoch": 0.24054982817869416, + "grad_norm": 3.140625, + "learning_rate": 4.4268433179723506e-05, + "loss": 0.25385417938232424, + "step": 210 + }, + { + "epoch": 0.24054982817869416, + "forget_loss": 4.970596364728408e-06, + "retain_loss": 0.17109914124011993, + "step": 210 + }, + { + "epoch": 0.24054982817869416, + "forget_loss": 1.6977497580228373e-05, + "retain_loss": 0.2036435753107071, + "step": 210 + }, + { + "epoch": 0.24054982817869416, + "forget_loss": 3.1074764592631254e-06, + "retain_loss": 0.13067637383937836, + "step": 210 + }, + { + "epoch": 0.24054982817869416, + "forget_loss": 2.0813706669287058e-06, + "retain_loss": 0.1853303760290146, + "step": 210 + }, + { + "epoch": 0.2520045819014891, + "grad_norm": 2.5, + "learning_rate": 4.398041474654378e-05, + "loss": 0.2122504711151123, + "step": 220 + }, + { + "epoch": 0.2520045819014891, + "forget_loss": 1.3193698578106705e-06, + "retain_loss": 0.14352917671203613, + "step": 220 + }, + { + "epoch": 0.2520045819014891, + "forget_loss": 1.1001748134731315e-06, + "retain_loss": 0.30384859442710876, + "step": 220 + }, + { + "epoch": 0.2520045819014891, + "forget_loss": 1.4734857813891722e-06, + "retain_loss": 0.06934046000242233, + "step": 220 + }, + { + "epoch": 0.2520045819014891, + "forget_loss": 2.7343555757397553e-06, + "retain_loss": 0.1081199198961258, + "step": 220 + }, + { + "epoch": 0.2634593356242841, + "grad_norm": 3.09375, + "learning_rate": 4.369239631336406e-05, + "loss": 0.20068085193634033, + "step": 230 + }, + { + "epoch": 0.2634593356242841, + "forget_loss": 2.154215962946182e-06, + "retain_loss": 0.10300299525260925, + "step": 230 + }, + { + "epoch": 0.2634593356242841, + "forget_loss": 9.424740710528567e-06, + "retain_loss": 0.13396301865577698, + "step": 230 + }, + { + "epoch": 0.2634593356242841, + "forget_loss": 3.0522624001605436e-06, + "retain_loss": 0.33339884877204895, + "step": 230 + }, + { + "epoch": 0.2634593356242841, + "forget_loss": 8.897523912310135e-06, + "retain_loss": 0.1534104347229004, + "step": 230 + }, + { + "epoch": 0.27491408934707906, + "grad_norm": 2.671875, + "learning_rate": 4.340437788018433e-05, + "loss": 0.1963863492012024, + "step": 240 + }, + { + "epoch": 0.27491408934707906, + "forget_loss": 1.7897947373057832e-06, + "retain_loss": 0.11809217184782028, + "step": 240 + }, + { + "epoch": 0.27491408934707906, + "forget_loss": 5.25018833741342e-07, + "retain_loss": 0.15055039525032043, + "step": 240 + }, + { + "epoch": 0.27491408934707906, + "forget_loss": 6.8833096520393156e-06, + "retain_loss": 0.16997316479682922, + "step": 240 + }, + { + "epoch": 0.27491408934707906, + "forget_loss": 2.313455297553446e-06, + "retain_loss": 0.10752896964550018, + "step": 240 + }, + { + "epoch": 0.286368843069874, + "grad_norm": 2.59375, + "learning_rate": 4.311635944700461e-05, + "loss": 0.20680031776428223, + "step": 250 + }, + { + "epoch": 0.286368843069874, + "forget_loss": 2.1803207346238196e-06, + "retain_loss": 0.07074207067489624, + "step": 250 + }, + { + "epoch": 0.286368843069874, + "forget_loss": 1.8291285641680588e-06, + "retain_loss": 0.09072300791740417, + "step": 250 + }, + { + "epoch": 0.286368843069874, + "forget_loss": 1.0191804449277697e-06, + "retain_loss": 0.4518914818763733, + "step": 250 + }, + { + "epoch": 0.286368843069874, + "forget_loss": 1.1188744792889338e-06, + "retain_loss": 0.09803809970617294, + "step": 250 + }, + { + "epoch": 0.29782359679266895, + "grad_norm": 2.703125, + "learning_rate": 4.282834101382488e-05, + "loss": 0.16371082067489623, + "step": 260 + }, + { + "epoch": 0.29782359679266895, + "forget_loss": 3.8669927562295925e-06, + "retain_loss": 0.17645332217216492, + "step": 260 + }, + { + "epoch": 0.29782359679266895, + "forget_loss": 5.920769581280183e-06, + "retain_loss": 0.18589472770690918, + "step": 260 + }, + { + "epoch": 0.29782359679266895, + "forget_loss": 5.846962267241906e-06, + "retain_loss": 0.1563558578491211, + "step": 260 + }, + { + "epoch": 0.29782359679266895, + "forget_loss": 2.4713845050428063e-06, + "retain_loss": 0.08506585657596588, + "step": 260 + }, + { + "epoch": 0.30927835051546393, + "grad_norm": 1.984375, + "learning_rate": 4.254032258064516e-05, + "loss": 0.1610316514968872, + "step": 270 + }, + { + "epoch": 0.30927835051546393, + "forget_loss": 2.003027475439012e-06, + "retain_loss": 0.1364717036485672, + "step": 270 + }, + { + "epoch": 0.30927835051546393, + "forget_loss": 2.4025737275223946e-06, + "retain_loss": 0.06489767134189606, + "step": 270 + }, + { + "epoch": 0.30927835051546393, + "forget_loss": 1.9498058918543393e-06, + "retain_loss": 0.15014496445655823, + "step": 270 + }, + { + "epoch": 0.30927835051546393, + "forget_loss": 1.4814335145274526e-06, + "retain_loss": 0.10888711363077164, + "step": 270 + }, + { + "epoch": 0.3207331042382589, + "grad_norm": 2.03125, + "learning_rate": 4.225230414746544e-05, + "loss": 0.1513701558113098, + "step": 280 + }, + { + "epoch": 0.3207331042382589, + "forget_loss": 3.564718099369202e-06, + "retain_loss": 0.18767109513282776, + "step": 280 + }, + { + "epoch": 0.3207331042382589, + "forget_loss": 1.5525797607551794e-06, + "retain_loss": 0.1132984533905983, + "step": 280 + }, + { + "epoch": 0.3207331042382589, + "forget_loss": 2.530495294195134e-06, + "retain_loss": 0.14477676153182983, + "step": 280 + }, + { + "epoch": 0.3207331042382589, + "forget_loss": 1.2278740086912876e-06, + "retain_loss": 0.22871914505958557, + "step": 280 + }, + { + "epoch": 0.3321878579610538, + "grad_norm": 1.9375, + "learning_rate": 4.196428571428572e-05, + "loss": 0.13684289455413817, + "step": 290 + }, + { + "epoch": 0.3321878579610538, + "forget_loss": 6.087240080887568e-07, + "retain_loss": 0.12418590486049652, + "step": 290 + }, + { + "epoch": 0.3321878579610538, + "forget_loss": 1.3446937145999982e-06, + "retain_loss": 0.07405225187540054, + "step": 290 + }, + { + "epoch": 0.3321878579610538, + "forget_loss": 5.7531210586603265e-06, + "retain_loss": 0.18760758638381958, + "step": 290 + }, + { + "epoch": 0.3321878579610538, + "forget_loss": 3.2198163353314158e-06, + "retain_loss": 0.06644970923662186, + "step": 290 + }, + { + "epoch": 0.3436426116838488, + "grad_norm": 1.734375, + "learning_rate": 4.167626728110599e-05, + "loss": 0.12020657062530518, + "step": 300 + }, + { + "epoch": 0.3436426116838488, + "forget_loss": 3.8907467114768224e-07, + "retain_loss": 0.09106294065713882, + "step": 300 + }, + { + "epoch": 0.3436426116838488, + "forget_loss": 1.2785067156073637e-06, + "retain_loss": 0.07731538265943527, + "step": 300 + }, + { + "epoch": 0.3436426116838488, + "forget_loss": 7.883311923251313e-07, + "retain_loss": 0.0779929980635643, + "step": 300 + }, + { + "epoch": 0.3436426116838488, + "forget_loss": 1.6741282706789207e-06, + "retain_loss": 0.11640097200870514, + "step": 300 + }, + { + "epoch": 0.3550973654066438, + "grad_norm": 2.09375, + "learning_rate": 4.138824884792627e-05, + "loss": 0.11634387969970703, + "step": 310 + }, + { + "epoch": 0.3550973654066438, + "forget_loss": 1.1909932027265313e-06, + "retain_loss": 0.09301010519266129, + "step": 310 + }, + { + "epoch": 0.3550973654066438, + "forget_loss": 5.385973622651363e-07, + "retain_loss": 0.13168799877166748, + "step": 310 + }, + { + "epoch": 0.3550973654066438, + "forget_loss": 4.287514457246289e-06, + "retain_loss": 0.07441989332437515, + "step": 310 + }, + { + "epoch": 0.3550973654066438, + "forget_loss": 5.496154244610807e-06, + "retain_loss": 0.0625639408826828, + "step": 310 + }, + { + "epoch": 0.3665521191294387, + "grad_norm": 2.09375, + "learning_rate": 4.1100230414746544e-05, + "loss": 0.13164939880371093, + "step": 320 + }, + { + "epoch": 0.3665521191294387, + "forget_loss": 1.5266854234141647e-06, + "retain_loss": 0.07643962651491165, + "step": 320 + }, + { + "epoch": 0.3665521191294387, + "forget_loss": 5.235110052126402e-07, + "retain_loss": 0.10404697060585022, + "step": 320 + }, + { + "epoch": 0.3665521191294387, + "forget_loss": 3.643459649538272e-06, + "retain_loss": 0.11882194131612778, + "step": 320 + }, + { + "epoch": 0.3665521191294387, + "forget_loss": 9.903519639919978e-07, + "retain_loss": 0.07760447263717651, + "step": 320 + }, + { + "epoch": 0.37800687285223367, + "grad_norm": 1.7265625, + "learning_rate": 4.081221198156682e-05, + "loss": 0.11806095838546753, + "step": 330 + }, + { + "epoch": 0.37800687285223367, + "forget_loss": 2.186560550399008e-06, + "retain_loss": 0.05915055423974991, + "step": 330 + }, + { + "epoch": 0.37800687285223367, + "forget_loss": 2.5664467102615163e-06, + "retain_loss": 0.06395165622234344, + "step": 330 + }, + { + "epoch": 0.37800687285223367, + "forget_loss": 1.2531793345260667e-06, + "retain_loss": 0.1056695505976677, + "step": 330 + }, + { + "epoch": 0.37800687285223367, + "forget_loss": 7.011165507719852e-06, + "retain_loss": 0.10078335553407669, + "step": 330 + }, + { + "epoch": 0.38946162657502864, + "grad_norm": 1.734375, + "learning_rate": 4.0524193548387096e-05, + "loss": 0.10047016143798829, + "step": 340 + }, + { + "epoch": 0.38946162657502864, + "forget_loss": 1.4236219385566073e-06, + "retain_loss": 0.5226035714149475, + "step": 340 + }, + { + "epoch": 0.38946162657502864, + "forget_loss": 1.951645344888675e-06, + "retain_loss": 0.10346213728189468, + "step": 340 + }, + { + "epoch": 0.38946162657502864, + "forget_loss": 2.398945980530698e-06, + "retain_loss": 0.03990686312317848, + "step": 340 + }, + { + "epoch": 0.38946162657502864, + "forget_loss": 4.940415237797424e-06, + "retain_loss": 0.08106988668441772, + "step": 340 + }, + { + "epoch": 0.4009163802978236, + "grad_norm": 1.9296875, + "learning_rate": 4.0236175115207375e-05, + "loss": 0.10264077186584472, + "step": 350 + }, + { + "epoch": 0.4009163802978236, + "forget_loss": 1.1442448339948896e-06, + "retain_loss": 0.07710669934749603, + "step": 350 + }, + { + "epoch": 0.4009163802978236, + "forget_loss": 1.1809590887423838e-06, + "retain_loss": 0.384318470954895, + "step": 350 + }, + { + "epoch": 0.4009163802978236, + "forget_loss": 1.8739415281743277e-06, + "retain_loss": 0.1440838724374771, + "step": 350 + }, + { + "epoch": 0.4009163802978236, + "forget_loss": 1.596398078618222e-06, + "retain_loss": 0.08889831602573395, + "step": 350 + }, + { + "epoch": 0.41237113402061853, + "grad_norm": 1.40625, + "learning_rate": 3.9948156682027654e-05, + "loss": 0.09905218482017517, + "step": 360 + }, + { + "epoch": 0.41237113402061853, + "forget_loss": 6.401254722732119e-06, + "retain_loss": 0.06261590123176575, + "step": 360 + }, + { + "epoch": 0.41237113402061853, + "forget_loss": 3.825871317530982e-07, + "retain_loss": 0.15589363873004913, + "step": 360 + }, + { + "epoch": 0.41237113402061853, + "forget_loss": 1.1340009677951457e-06, + "retain_loss": 0.026375912129878998, + "step": 360 + }, + { + "epoch": 0.41237113402061853, + "forget_loss": 1.092132379199029e-06, + "retain_loss": 0.07016332447528839, + "step": 360 + }, + { + "epoch": 0.4238258877434135, + "grad_norm": 2.28125, + "learning_rate": 3.9660138248847926e-05, + "loss": 0.10010771751403809, + "step": 370 + }, + { + "epoch": 0.4238258877434135, + "forget_loss": 5.463415959638951e-07, + "retain_loss": 0.09104490280151367, + "step": 370 + }, + { + "epoch": 0.4238258877434135, + "forget_loss": 7.711361149631557e-07, + "retain_loss": 0.07780595868825912, + "step": 370 + }, + { + "epoch": 0.4238258877434135, + "forget_loss": 8.610635973127501e-07, + "retain_loss": 0.0850876048207283, + "step": 370 + }, + { + "epoch": 0.4238258877434135, + "forget_loss": 2.193863338106894e-06, + "retain_loss": 0.02873000130057335, + "step": 370 + }, + { + "epoch": 0.4352806414662085, + "grad_norm": 1.8125, + "learning_rate": 3.9372119815668206e-05, + "loss": 0.07582432627677918, + "step": 380 + }, + { + "epoch": 0.4352806414662085, + "forget_loss": 5.40978362550959e-07, + "retain_loss": 0.03535467013716698, + "step": 380 + }, + { + "epoch": 0.4352806414662085, + "forget_loss": 8.922065717342775e-07, + "retain_loss": 0.057049620896577835, + "step": 380 + }, + { + "epoch": 0.4352806414662085, + "forget_loss": 5.560720524044882e-07, + "retain_loss": 0.1527683138847351, + "step": 380 + }, + { + "epoch": 0.4352806414662085, + "forget_loss": 4.7091802457543963e-07, + "retain_loss": 0.1109684631228447, + "step": 380 + }, + { + "epoch": 0.44673539518900346, + "grad_norm": 1.7734375, + "learning_rate": 3.908410138248848e-05, + "loss": 0.09133011102676392, + "step": 390 + }, + { + "epoch": 0.44673539518900346, + "forget_loss": 8.056795763877744e-07, + "retain_loss": 0.09070051461458206, + "step": 390 + }, + { + "epoch": 0.44673539518900346, + "forget_loss": 1.1127366406071815e-06, + "retain_loss": 0.0616828016936779, + "step": 390 + }, + { + "epoch": 0.44673539518900346, + "forget_loss": 3.5524210488802055e-06, + "retain_loss": 0.06886927038431168, + "step": 390 + }, + { + "epoch": 0.44673539518900346, + "forget_loss": 2.0374077394080814e-06, + "retain_loss": 0.10105489939451218, + "step": 390 + }, + { + "epoch": 0.4581901489117984, + "grad_norm": 1.59375, + "learning_rate": 3.879608294930876e-05, + "loss": 0.07695338726043702, + "step": 400 + }, + { + "epoch": 0.4581901489117984, + "forget_loss": 1.342506493529072e-06, + "retain_loss": 0.03703423589468002, + "step": 400 + }, + { + "epoch": 0.4581901489117984, + "forget_loss": 4.241804276716721e-07, + "retain_loss": 0.07702656090259552, + "step": 400 + }, + { + "epoch": 0.4581901489117984, + "forget_loss": 2.735993405167392e-07, + "retain_loss": 0.04419003427028656, + "step": 400 + }, + { + "epoch": 0.4581901489117984, + "forget_loss": 3.817524429905461e-07, + "retain_loss": 0.040069010108709335, + "step": 400 + }, + { + "epoch": 0.46964490263459335, + "grad_norm": 1.671875, + "learning_rate": 3.8508064516129036e-05, + "loss": 0.07269055247306824, + "step": 410 + }, + { + "epoch": 0.46964490263459335, + "forget_loss": 2.8240863230166724e-06, + "retain_loss": 0.04700532928109169, + "step": 410 + }, + { + "epoch": 0.46964490263459335, + "forget_loss": 1.747673792351634e-07, + "retain_loss": 0.033581238240003586, + "step": 410 + }, + { + "epoch": 0.46964490263459335, + "forget_loss": 5.92311096170306e-07, + "retain_loss": 0.09985388815402985, + "step": 410 + }, + { + "epoch": 0.46964490263459335, + "forget_loss": 1.7826670273279888e-07, + "retain_loss": 0.07349472492933273, + "step": 410 + }, + { + "epoch": 0.48109965635738833, + "grad_norm": 1.8046875, + "learning_rate": 3.8220046082949316e-05, + "loss": 0.07912165522575379, + "step": 420 + }, + { + "epoch": 0.48109965635738833, + "forget_loss": 1.2650438918626605e-07, + "retain_loss": 0.06736782938241959, + "step": 420 + }, + { + "epoch": 0.48109965635738833, + "forget_loss": 1.5735973875052878e-06, + "retain_loss": 0.03057463839650154, + "step": 420 + }, + { + "epoch": 0.48109965635738833, + "forget_loss": 4.024736313112953e-07, + "retain_loss": 0.05045014247298241, + "step": 420 + }, + { + "epoch": 0.48109965635738833, + "forget_loss": 2.5390670543856686e-06, + "retain_loss": 0.07917766273021698, + "step": 420 + }, + { + "epoch": 0.4925544100801833, + "grad_norm": 1.1953125, + "learning_rate": 3.793202764976959e-05, + "loss": 0.0656080961227417, + "step": 430 + }, + { + "epoch": 0.4925544100801833, + "forget_loss": 5.468404538078175e-07, + "retain_loss": 0.02683027647435665, + "step": 430 + }, + { + "epoch": 0.4925544100801833, + "forget_loss": 5.585835083365964e-07, + "retain_loss": 0.0377357117831707, + "step": 430 + }, + { + "epoch": 0.4925544100801833, + "forget_loss": 3.003877679930156e-07, + "retain_loss": 0.052259303629398346, + "step": 430 + }, + { + "epoch": 0.4925544100801833, + "forget_loss": 1.2751459053106373e-06, + "retain_loss": 0.10837093740701675, + "step": 430 + }, + { + "epoch": 0.5040091638029782, + "grad_norm": 1.8046875, + "learning_rate": 3.764400921658987e-05, + "loss": 0.06297182440757751, + "step": 440 + }, + { + "epoch": 0.5040091638029782, + "forget_loss": 2.636994622662314e-07, + "retain_loss": 0.04880654811859131, + "step": 440 + }, + { + "epoch": 0.5040091638029782, + "forget_loss": 2.0067361674591666e-06, + "retain_loss": 0.027269020676612854, + "step": 440 + }, + { + "epoch": 0.5040091638029782, + "forget_loss": 1.8749557284536422e-06, + "retain_loss": 0.04770921915769577, + "step": 440 + }, + { + "epoch": 0.5040091638029782, + "forget_loss": 7.214839001790097e-07, + "retain_loss": 0.07100628316402435, + "step": 440 + }, + { + "epoch": 0.5154639175257731, + "grad_norm": 1.0859375, + "learning_rate": 3.735599078341014e-05, + "loss": 0.0665643870830536, + "step": 450 + }, + { + "epoch": 0.5154639175257731, + "forget_loss": 8.713016796946249e-08, + "retain_loss": 0.042148686945438385, + "step": 450 + }, + { + "epoch": 0.5154639175257731, + "forget_loss": 6.566732508872519e-07, + "retain_loss": 0.05762806907296181, + "step": 450 + }, + { + "epoch": 0.5154639175257731, + "forget_loss": 2.520494888358371e-07, + "retain_loss": 0.0422804020345211, + "step": 450 + }, + { + "epoch": 0.5154639175257731, + "forget_loss": 2.0052689251315314e-06, + "retain_loss": 0.05088704079389572, + "step": 450 + }, + { + "epoch": 0.5269186712485682, + "grad_norm": 1.140625, + "learning_rate": 3.706797235023042e-05, + "loss": 0.06394548416137695, + "step": 460 + }, + { + "epoch": 0.5269186712485682, + "forget_loss": 4.606073389368248e-07, + "retain_loss": 0.0739588811993599, + "step": 460 + }, + { + "epoch": 0.5269186712485682, + "forget_loss": 4.4586619196707034e-07, + "retain_loss": 0.059775300323963165, + "step": 460 + }, + { + "epoch": 0.5269186712485682, + "forget_loss": 4.324638211983256e-07, + "retain_loss": 0.08108869194984436, + "step": 460 + }, + { + "epoch": 0.5269186712485682, + "forget_loss": 5.558591738008545e-07, + "retain_loss": 0.027880918234586716, + "step": 460 + }, + { + "epoch": 0.5383734249713631, + "grad_norm": 1.171875, + "learning_rate": 3.677995391705069e-05, + "loss": 0.05814751386642456, + "step": 470 + }, + { + "epoch": 0.5383734249713631, + "forget_loss": 3.032934046132141e-07, + "retain_loss": 0.024084806442260742, + "step": 470 + }, + { + "epoch": 0.5383734249713631, + "forget_loss": 8.804982485344226e-07, + "retain_loss": 0.03435460478067398, + "step": 470 + }, + { + "epoch": 0.5383734249713631, + "forget_loss": 1.1724860087269917e-06, + "retain_loss": 0.048975635319948196, + "step": 470 + }, + { + "epoch": 0.5383734249713631, + "forget_loss": 3.8122900036796636e-07, + "retain_loss": 0.0677463635802269, + "step": 470 + }, + { + "epoch": 0.5498281786941581, + "grad_norm": 1.734375, + "learning_rate": 3.649193548387097e-05, + "loss": 0.060336172580718994, + "step": 480 + }, + { + "epoch": 0.5498281786941581, + "forget_loss": 9.458409522267175e-07, + "retain_loss": 0.07034104317426682, + "step": 480 + }, + { + "epoch": 0.5498281786941581, + "forget_loss": 6.857035828033986e-07, + "retain_loss": 0.03801558166742325, + "step": 480 + }, + { + "epoch": 0.5498281786941581, + "forget_loss": 1.3080871212878264e-06, + "retain_loss": 0.07545764744281769, + "step": 480 + }, + { + "epoch": 0.5498281786941581, + "forget_loss": 3.9946274910107604e-07, + "retain_loss": 0.03869559243321419, + "step": 480 + }, + { + "epoch": 0.561282932416953, + "grad_norm": 1.0546875, + "learning_rate": 3.620391705069124e-05, + "loss": 0.05845548510551453, + "step": 490 + }, + { + "epoch": 0.561282932416953, + "forget_loss": 9.76757291937247e-07, + "retain_loss": 0.04530367627739906, + "step": 490 + }, + { + "epoch": 0.561282932416953, + "forget_loss": 5.240474933998485e-07, + "retain_loss": 0.03550443798303604, + "step": 490 + }, + { + "epoch": 0.561282932416953, + "forget_loss": 7.904294534455403e-07, + "retain_loss": 0.08905749022960663, + "step": 490 + }, + { + "epoch": 0.561282932416953, + "forget_loss": 3.536492840794381e-07, + "retain_loss": 0.05929813161492348, + "step": 490 + }, + { + "epoch": 0.572737686139748, + "grad_norm": 1.1640625, + "learning_rate": 3.591589861751152e-05, + "loss": 0.06306434273719788, + "step": 500 + }, + { + "epoch": 0.572737686139748, + "forget_loss": 4.770619170813006e-07, + "retain_loss": 0.0518912598490715, + "step": 500 + }, + { + "epoch": 0.572737686139748, + "forget_loss": 7.711427940648719e-08, + "retain_loss": 0.06265456974506378, + "step": 500 + }, + { + "epoch": 0.572737686139748, + "forget_loss": 1.8670939425646793e-06, + "retain_loss": 0.07513444125652313, + "step": 500 + }, + { + "epoch": 0.572737686139748, + "forget_loss": 4.7316902396232763e-07, + "retain_loss": 0.08540341258049011, + "step": 500 + }, + { + "epoch": 0.584192439862543, + "grad_norm": 0.984375, + "learning_rate": 3.5627880184331795e-05, + "loss": 0.05606219172477722, + "step": 510 + }, + { + "epoch": 0.584192439862543, + "forget_loss": 8.002095910342177e-07, + "retain_loss": 0.04273177683353424, + "step": 510 + }, + { + "epoch": 0.584192439862543, + "forget_loss": 1.8704803039781837e-07, + "retain_loss": 0.053921207785606384, + "step": 510 + }, + { + "epoch": 0.584192439862543, + "forget_loss": 3.775904247049766e-07, + "retain_loss": 0.0322711281478405, + "step": 510 + }, + { + "epoch": 0.584192439862543, + "forget_loss": 6.053579681974952e-07, + "retain_loss": 0.05659232288599014, + "step": 510 + }, + { + "epoch": 0.5956471935853379, + "grad_norm": 1.2265625, + "learning_rate": 3.5339861751152074e-05, + "loss": 0.049942076206207275, + "step": 520 + }, + { + "epoch": 0.5956471935853379, + "forget_loss": 2.9679642921109917e-07, + "retain_loss": 0.08346150070428848, + "step": 520 + }, + { + "epoch": 0.5956471935853379, + "forget_loss": 2.005875501254195e-07, + "retain_loss": 0.030936188995838165, + "step": 520 + }, + { + "epoch": 0.5956471935853379, + "forget_loss": 1.0443299061080324e-06, + "retain_loss": 0.09520356357097626, + "step": 520 + }, + { + "epoch": 0.5956471935853379, + "forget_loss": 2.0819126689275436e-07, + "retain_loss": 0.0474083349108696, + "step": 520 + }, + { + "epoch": 0.6071019473081328, + "grad_norm": 1.0625, + "learning_rate": 3.5051843317972346e-05, + "loss": 0.05396345257759094, + "step": 530 + }, + { + "epoch": 0.6071019473081328, + "forget_loss": 8.294673534692265e-07, + "retain_loss": 0.042203839868307114, + "step": 530 + }, + { + "epoch": 0.6071019473081328, + "forget_loss": 7.164442195062293e-08, + "retain_loss": 0.030783096328377724, + "step": 530 + }, + { + "epoch": 0.6071019473081328, + "forget_loss": 1.3861478691978846e-05, + "retain_loss": 0.04198005795478821, + "step": 530 + }, + { + "epoch": 0.6071019473081328, + "forget_loss": 4.225911993671616e-07, + "retain_loss": 0.06045946478843689, + "step": 530 + }, + { + "epoch": 0.6185567010309279, + "grad_norm": 1.2578125, + "learning_rate": 3.4763824884792625e-05, + "loss": 0.05722784399986267, + "step": 540 + }, + { + "epoch": 0.6185567010309279, + "forget_loss": 3.0147746201691916e-06, + "retain_loss": 0.0411076694726944, + "step": 540 + }, + { + "epoch": 0.6185567010309279, + "forget_loss": 2.582411582352506e-07, + "retain_loss": 0.023279590532183647, + "step": 540 + }, + { + "epoch": 0.6185567010309279, + "forget_loss": 2.0758065488735156e-07, + "retain_loss": 0.04550226032733917, + "step": 540 + }, + { + "epoch": 0.6185567010309279, + "forget_loss": 9.63341904025583e-07, + "retain_loss": 0.0418274812400341, + "step": 540 + }, + { + "epoch": 0.6300114547537228, + "grad_norm": 1.3359375, + "learning_rate": 3.4475806451612905e-05, + "loss": 0.04913765788078308, + "step": 550 + }, + { + "epoch": 0.6300114547537228, + "forget_loss": 2.0537379441520898e-07, + "retain_loss": 0.046085719019174576, + "step": 550 + }, + { + "epoch": 0.6300114547537228, + "forget_loss": 4.5911775714557734e-07, + "retain_loss": 0.041781309992074966, + "step": 550 + }, + { + "epoch": 0.6300114547537228, + "forget_loss": 1.873733936008648e-06, + "retain_loss": 0.030963057652115822, + "step": 550 + }, + { + "epoch": 0.6300114547537228, + "forget_loss": 2.3070414556514152e-07, + "retain_loss": 0.036035358905792236, + "step": 550 + }, + { + "epoch": 0.6414662084765178, + "grad_norm": 1.2109375, + "learning_rate": 3.4187788018433184e-05, + "loss": 0.046972605586051944, + "step": 560 + }, + { + "epoch": 0.6414662084765178, + "forget_loss": 1.888665224214492e-07, + "retain_loss": 0.04609245806932449, + "step": 560 + }, + { + "epoch": 0.6414662084765178, + "forget_loss": 1.1302165603410685e-06, + "retain_loss": 0.04581566900014877, + "step": 560 + }, + { + "epoch": 0.6414662084765178, + "forget_loss": 3.713363128099445e-07, + "retain_loss": 0.05701315402984619, + "step": 560 + }, + { + "epoch": 0.6414662084765178, + "forget_loss": 1.0438467512585703e-07, + "retain_loss": 0.04488621652126312, + "step": 560 + }, + { + "epoch": 0.6529209621993127, + "grad_norm": 1.640625, + "learning_rate": 3.389976958525346e-05, + "loss": 0.04600127041339874, + "step": 570 + }, + { + "epoch": 0.6529209621993127, + "forget_loss": 1.2589470088641974e-07, + "retain_loss": 0.04046102985739708, + "step": 570 + }, + { + "epoch": 0.6529209621993127, + "forget_loss": 2.068235062324675e-06, + "retain_loss": 0.06527687609195709, + "step": 570 + }, + { + "epoch": 0.6529209621993127, + "forget_loss": 4.6239173911999387e-07, + "retain_loss": 0.02964375726878643, + "step": 570 + }, + { + "epoch": 0.6529209621993127, + "forget_loss": 1.7551890323375119e-06, + "retain_loss": 0.04939418286085129, + "step": 570 + }, + { + "epoch": 0.6643757159221076, + "grad_norm": 1.4921875, + "learning_rate": 3.3611751152073736e-05, + "loss": 0.04693392515182495, + "step": 580 + }, + { + "epoch": 0.6643757159221076, + "forget_loss": 1.162212441840893e-07, + "retain_loss": 0.01956232823431492, + "step": 580 + }, + { + "epoch": 0.6643757159221076, + "forget_loss": 5.034058858655044e-07, + "retain_loss": 0.029308225959539413, + "step": 580 + }, + { + "epoch": 0.6643757159221076, + "forget_loss": 1.8678618118883605e-07, + "retain_loss": 0.057615552097558975, + "step": 580 + }, + { + "epoch": 0.6643757159221076, + "forget_loss": 5.4416375405708095e-08, + "retain_loss": 0.06805437058210373, + "step": 580 + }, + { + "epoch": 0.6758304696449027, + "grad_norm": 0.85546875, + "learning_rate": 3.3323732718894015e-05, + "loss": 0.044681668281555176, + "step": 590 + }, + { + "epoch": 0.6758304696449027, + "forget_loss": 6.884723688926897e-07, + "retain_loss": 0.032636139541864395, + "step": 590 + }, + { + "epoch": 0.6758304696449027, + "forget_loss": 2.1161808660963288e-07, + "retain_loss": 0.03656921535730362, + "step": 590 + }, + { + "epoch": 0.6758304696449027, + "forget_loss": 3.788201468069019e-07, + "retain_loss": 0.0688297301530838, + "step": 590 + }, + { + "epoch": 0.6758304696449027, + "forget_loss": 8.874022228155809e-08, + "retain_loss": 0.021696442738175392, + "step": 590 + }, + { + "epoch": 0.6872852233676976, + "grad_norm": 0.94140625, + "learning_rate": 3.303571428571429e-05, + "loss": 0.04318673312664032, + "step": 600 + }, + { + "epoch": 0.6872852233676976, + "forget_loss": 3.341031913350889e-07, + "retain_loss": 0.03312600031495094, + "step": 600 + }, + { + "epoch": 0.6872852233676976, + "forget_loss": 1.59767705554259e-06, + "retain_loss": 0.037248943001031876, + "step": 600 + }, + { + "epoch": 0.6872852233676976, + "forget_loss": 1.2864059328876465e-07, + "retain_loss": 0.051617469638586044, + "step": 600 + }, + { + "epoch": 0.6872852233676976, + "forget_loss": 1.3946912247320142e-07, + "retain_loss": 0.07654328644275665, + "step": 600 + }, + { + "epoch": 0.6987399770904925, + "grad_norm": 0.6484375, + "learning_rate": 3.2747695852534566e-05, + "loss": 0.043347522616386414, + "step": 610 + }, + { + "epoch": 0.6987399770904925, + "forget_loss": 1.6136347369410942e-07, + "retain_loss": 0.06624381989240646, + "step": 610 + }, + { + "epoch": 0.6987399770904925, + "forget_loss": 1.5072893688738986e-07, + "retain_loss": 0.025636199861764908, + "step": 610 + }, + { + "epoch": 0.6987399770904925, + "forget_loss": 3.572060336409777e-07, + "retain_loss": 0.029844949021935463, + "step": 610 + }, + { + "epoch": 0.6987399770904925, + "forget_loss": 3.3666503895801725e-07, + "retain_loss": 0.08041810244321823, + "step": 610 + }, + { + "epoch": 0.7101947308132875, + "grad_norm": 0.76953125, + "learning_rate": 3.245967741935484e-05, + "loss": 0.04078985154628754, + "step": 620 + }, + { + "epoch": 0.7101947308132875, + "forget_loss": 2.2686955958306498e-07, + "retain_loss": 0.03770332783460617, + "step": 620 + }, + { + "epoch": 0.7101947308132875, + "forget_loss": 1.2471460308915994e-07, + "retain_loss": 0.04125113785266876, + "step": 620 + }, + { + "epoch": 0.7101947308132875, + "forget_loss": 5.980033392916084e-07, + "retain_loss": 0.028695637360215187, + "step": 620 + }, + { + "epoch": 0.7101947308132875, + "forget_loss": 2.1789082893519662e-07, + "retain_loss": 0.021218575537204742, + "step": 620 + }, + { + "epoch": 0.7216494845360825, + "grad_norm": 0.72265625, + "learning_rate": 3.217165898617512e-05, + "loss": 0.03691001832485199, + "step": 630 + }, + { + "epoch": 0.7216494845360825, + "forget_loss": 5.882121172362531e-07, + "retain_loss": 0.028744207695126534, + "step": 630 + }, + { + "epoch": 0.7216494845360825, + "forget_loss": 3.706446420892462e-07, + "retain_loss": 0.037669021636247635, + "step": 630 + }, + { + "epoch": 0.7216494845360825, + "forget_loss": 9.750963272381341e-07, + "retain_loss": 0.055296238511800766, + "step": 630 + }, + { + "epoch": 0.7216494845360825, + "forget_loss": 2.988974188156135e-07, + "retain_loss": 0.04488551616668701, + "step": 630 + }, + { + "epoch": 0.7331042382588774, + "grad_norm": 0.8671875, + "learning_rate": 3.188364055299539e-05, + "loss": 0.038949227333068846, + "step": 640 + }, + { + "epoch": 0.7331042382588774, + "forget_loss": 1.2395511816976068e-07, + "retain_loss": 0.027795039117336273, + "step": 640 + }, + { + "epoch": 0.7331042382588774, + "forget_loss": 5.673878149536904e-07, + "retain_loss": 0.06736742705106735, + "step": 640 + }, + { + "epoch": 0.7331042382588774, + "forget_loss": 1.1004679834059061e-07, + "retain_loss": 0.0364450179040432, + "step": 640 + }, + { + "epoch": 0.7331042382588774, + "forget_loss": 1.2864478549090563e-07, + "retain_loss": 0.04605640098452568, + "step": 640 + }, + { + "epoch": 0.7445589919816724, + "grad_norm": 1.2578125, + "learning_rate": 3.159562211981567e-05, + "loss": 0.03986696600914001, + "step": 650 + }, + { + "epoch": 0.7445589919816724, + "forget_loss": 5.023429139328073e-07, + "retain_loss": 0.035111404955387115, + "step": 650 + }, + { + "epoch": 0.7445589919816724, + "forget_loss": 4.809768370250822e-07, + "retain_loss": 0.02187221683561802, + "step": 650 + }, + { + "epoch": 0.7445589919816724, + "forget_loss": 4.244728188496083e-06, + "retain_loss": 0.08197719603776932, + "step": 650 + }, + { + "epoch": 0.7445589919816724, + "forget_loss": 1.1673816402435477e-07, + "retain_loss": 0.08870132267475128, + "step": 650 + }, + { + "epoch": 0.7560137457044673, + "grad_norm": 0.6640625, + "learning_rate": 3.130760368663594e-05, + "loss": 0.039774364233016966, + "step": 660 + }, + { + "epoch": 0.7560137457044673, + "forget_loss": 1.1249169062921283e-07, + "retain_loss": 0.038397908210754395, + "step": 660 + }, + { + "epoch": 0.7560137457044673, + "forget_loss": 5.782552534583374e-07, + "retain_loss": 0.030945003032684326, + "step": 660 + }, + { + "epoch": 0.7560137457044673, + "forget_loss": 2.596835315671342e-07, + "retain_loss": 0.04453842341899872, + "step": 660 + }, + { + "epoch": 0.7560137457044673, + "forget_loss": 8.396180533054576e-07, + "retain_loss": 0.04271215945482254, + "step": 660 + }, + { + "epoch": 0.7674684994272624, + "grad_norm": 0.89453125, + "learning_rate": 3.101958525345622e-05, + "loss": 0.03887006938457489, + "step": 670 + }, + { + "epoch": 0.7674684994272624, + "forget_loss": 8.253943661884477e-08, + "retain_loss": 0.03154648467898369, + "step": 670 + }, + { + "epoch": 0.7674684994272624, + "forget_loss": 4.430251010489883e-07, + "retain_loss": 0.047183044254779816, + "step": 670 + }, + { + "epoch": 0.7674684994272624, + "forget_loss": 5.662621447299898e-07, + "retain_loss": 0.0315515510737896, + "step": 670 + }, + { + "epoch": 0.7674684994272624, + "forget_loss": 2.8115960049035493e-07, + "retain_loss": 0.024359095841646194, + "step": 670 + }, + { + "epoch": 0.7789232531500573, + "grad_norm": 0.890625, + "learning_rate": 3.07315668202765e-05, + "loss": 0.03558169901371002, + "step": 680 + }, + { + "epoch": 0.7789232531500573, + "forget_loss": 2.95549085649327e-07, + "retain_loss": 0.03642899915575981, + "step": 680 + }, + { + "epoch": 0.7789232531500573, + "forget_loss": 7.595866691190167e-08, + "retain_loss": 0.026957420632243156, + "step": 680 + }, + { + "epoch": 0.7789232531500573, + "forget_loss": 3.971479429765168e-07, + "retain_loss": 0.026554634794592857, + "step": 680 + }, + { + "epoch": 0.7789232531500573, + "forget_loss": 1.0031957572209649e-05, + "retain_loss": 0.055837519466876984, + "step": 680 + }, + { + "epoch": 0.7903780068728522, + "grad_norm": 0.72265625, + "learning_rate": 3.0443548387096776e-05, + "loss": 0.032911685109138486, + "step": 690 + }, + { + "epoch": 0.7903780068728522, + "forget_loss": 2.3226121470543148e-07, + "retain_loss": 0.028471192345023155, + "step": 690 + }, + { + "epoch": 0.7903780068728522, + "forget_loss": 2.1240731484795106e-07, + "retain_loss": 0.061532407999038696, + "step": 690 + }, + { + "epoch": 0.7903780068728522, + "forget_loss": 3.4258829373357e-07, + "retain_loss": 0.045654185116291046, + "step": 690 + }, + { + "epoch": 0.7903780068728522, + "forget_loss": 1.3982165114612144e-07, + "retain_loss": 0.04419733211398125, + "step": 690 + }, + { + "epoch": 0.8018327605956472, + "grad_norm": 0.369140625, + "learning_rate": 3.015552995391705e-05, + "loss": 0.03598496913909912, + "step": 700 + }, + { + "epoch": 0.8018327605956472, + "forget_loss": 3.2066935773400473e-07, + "retain_loss": 0.020428789779543877, + "step": 700 + }, + { + "epoch": 0.8018327605956472, + "forget_loss": 9.008678603095177e-07, + "retain_loss": 0.03619084134697914, + "step": 700 + }, + { + "epoch": 0.8018327605956472, + "forget_loss": 3.202090113063605e-07, + "retain_loss": 0.04894010350108147, + "step": 700 + }, + { + "epoch": 0.8018327605956472, + "forget_loss": 3.2703263741495903e-07, + "retain_loss": 0.039299145340919495, + "step": 700 + }, + { + "epoch": 0.8132875143184422, + "grad_norm": 0.48046875, + "learning_rate": 2.9867511520737328e-05, + "loss": 0.03154398202896118, + "step": 710 + }, + { + "epoch": 0.8132875143184422, + "forget_loss": 1.6021775195440568e-07, + "retain_loss": 0.05344302952289581, + "step": 710 + }, + { + "epoch": 0.8132875143184422, + "forget_loss": 3.327061222080374e-07, + "retain_loss": 0.018441231921315193, + "step": 710 + }, + { + "epoch": 0.8132875143184422, + "forget_loss": 1.2161015661149577e-07, + "retain_loss": 0.01777707226574421, + "step": 710 + }, + { + "epoch": 0.8132875143184422, + "forget_loss": 2.601665585189039e-07, + "retain_loss": 0.021317142993211746, + "step": 710 + }, + { + "epoch": 0.8247422680412371, + "grad_norm": 0.71484375, + "learning_rate": 2.9579493087557604e-05, + "loss": 0.031114420294761656, + "step": 720 + }, + { + "epoch": 0.8247422680412371, + "forget_loss": 3.656723990275168e-08, + "retain_loss": 0.02041560783982277, + "step": 720 + }, + { + "epoch": 0.8247422680412371, + "forget_loss": 7.133304791295814e-08, + "retain_loss": 0.0420437790453434, + "step": 720 + }, + { + "epoch": 0.8247422680412371, + "forget_loss": 1.2686764705449605e-07, + "retain_loss": 0.03707875683903694, + "step": 720 + }, + { + "epoch": 0.8247422680412371, + "forget_loss": 3.098398337897379e-07, + "retain_loss": 0.03388817980885506, + "step": 720 + }, + { + "epoch": 0.8361970217640321, + "grad_norm": 0.62109375, + "learning_rate": 2.9291474654377883e-05, + "loss": 0.031438523530960084, + "step": 730 + }, + { + "epoch": 0.8361970217640321, + "forget_loss": 6.143314408291189e-07, + "retain_loss": 0.037241023033857346, + "step": 730 + }, + { + "epoch": 0.8361970217640321, + "forget_loss": 1.567052549944492e-07, + "retain_loss": 0.029914623126387596, + "step": 730 + }, + { + "epoch": 0.8361970217640321, + "forget_loss": 1.7319891298939183e-07, + "retain_loss": 0.026813646778464317, + "step": 730 + }, + { + "epoch": 0.8361970217640321, + "forget_loss": 5.76246179662121e-08, + "retain_loss": 0.036752790212631226, + "step": 730 + }, + { + "epoch": 0.847651775486827, + "grad_norm": 0.447265625, + "learning_rate": 2.9003456221198155e-05, + "loss": 0.028349018096923827, + "step": 740 + }, + { + "epoch": 0.847651775486827, + "forget_loss": 1.0741296136984602e-06, + "retain_loss": 0.028334422037005424, + "step": 740 + }, + { + "epoch": 0.847651775486827, + "forget_loss": 5.47009335605253e-07, + "retain_loss": 0.03212451562285423, + "step": 740 + }, + { + "epoch": 0.847651775486827, + "forget_loss": 7.226554998851498e-08, + "retain_loss": 0.038880933076143265, + "step": 740 + }, + { + "epoch": 0.847651775486827, + "forget_loss": 1.0503694625185744e-07, + "retain_loss": 0.020103227347135544, + "step": 740 + }, + { + "epoch": 0.8591065292096219, + "grad_norm": 0.62109375, + "learning_rate": 2.8715437788018435e-05, + "loss": 0.027798715233802795, + "step": 750 + }, + { + "epoch": 0.8591065292096219, + "forget_loss": 9.740627149312786e-08, + "retain_loss": 0.03145676106214523, + "step": 750 + }, + { + "epoch": 0.8591065292096219, + "forget_loss": 8.817877272804253e-08, + "retain_loss": 0.0294309351593256, + "step": 750 + }, + { + "epoch": 0.8591065292096219, + "forget_loss": 2.1280260398270912e-07, + "retain_loss": 0.01636165753006935, + "step": 750 + }, + { + "epoch": 0.8591065292096219, + "forget_loss": 2.8950364594493294e-07, + "retain_loss": 0.03548318147659302, + "step": 750 + }, + { + "epoch": 0.870561282932417, + "grad_norm": 0.26171875, + "learning_rate": 2.8427419354838714e-05, + "loss": 0.027926307916641236, + "step": 760 + }, + { + "epoch": 0.870561282932417, + "forget_loss": 1.0444756526339916e-07, + "retain_loss": 0.035280242562294006, + "step": 760 + }, + { + "epoch": 0.870561282932417, + "forget_loss": 1.4455953589731507e-07, + "retain_loss": 0.03736939653754234, + "step": 760 + }, + { + "epoch": 0.870561282932417, + "forget_loss": 8.093392267483068e-08, + "retain_loss": 0.02323060669004917, + "step": 760 + }, + { + "epoch": 0.870561282932417, + "forget_loss": 4.825432711186295e-07, + "retain_loss": 0.028354808688163757, + "step": 760 + }, + { + "epoch": 0.8820160366552119, + "grad_norm": 0.419921875, + "learning_rate": 2.8139400921658986e-05, + "loss": 0.02746644616127014, + "step": 770 + }, + { + "epoch": 0.8820160366552119, + "forget_loss": 6.515568173881547e-08, + "retain_loss": 0.022168956696987152, + "step": 770 + }, + { + "epoch": 0.8820160366552119, + "forget_loss": 4.0877694118535146e-08, + "retain_loss": 0.02454095333814621, + "step": 770 + }, + { + "epoch": 0.8820160366552119, + "forget_loss": 4.9038700211667674e-08, + "retain_loss": 0.009829733520746231, + "step": 770 + }, + { + "epoch": 0.8820160366552119, + "forget_loss": 2.664446583366953e-07, + "retain_loss": 0.01737787388265133, + "step": 770 + }, + { + "epoch": 0.8934707903780069, + "grad_norm": 0.482421875, + "learning_rate": 2.7851382488479265e-05, + "loss": 0.025814527273178102, + "step": 780 + }, + { + "epoch": 0.8934707903780069, + "forget_loss": 3.281249050246515e-08, + "retain_loss": 0.020622696727514267, + "step": 780 + }, + { + "epoch": 0.8934707903780069, + "forget_loss": 5.478144515791428e-08, + "retain_loss": 0.020266670733690262, + "step": 780 + }, + { + "epoch": 0.8934707903780069, + "forget_loss": 6.129197061000013e-08, + "retain_loss": 0.018652314320206642, + "step": 780 + }, + { + "epoch": 0.8934707903780069, + "forget_loss": 2.284050850676067e-07, + "retain_loss": 0.013459251262247562, + "step": 780 + }, + { + "epoch": 0.9049255441008018, + "grad_norm": 0.51953125, + "learning_rate": 2.756336405529954e-05, + "loss": 0.027006852626800536, + "step": 790 + }, + { + "epoch": 0.9049255441008018, + "forget_loss": 2.6094051364111692e-08, + "retain_loss": 0.018663065508008003, + "step": 790 + }, + { + "epoch": 0.9049255441008018, + "forget_loss": 2.7085914666713506e-07, + "retain_loss": 0.009133810177445412, + "step": 790 + }, + { + "epoch": 0.9049255441008018, + "forget_loss": 8.054676214896972e-08, + "retain_loss": 0.02572845108807087, + "step": 790 + }, + { + "epoch": 0.9049255441008018, + "forget_loss": 1.0726026289376023e-07, + "retain_loss": 0.024219941347837448, + "step": 790 + }, + { + "epoch": 0.9163802978235968, + "grad_norm": 0.328125, + "learning_rate": 2.727534562211982e-05, + "loss": 0.02423024773597717, + "step": 800 + }, + { + "epoch": 0.9163802978235968, + "forget_loss": 7.722936601339825e-08, + "retain_loss": 0.04125501587986946, + "step": 800 + }, + { + "epoch": 0.9163802978235968, + "forget_loss": 9.053938043734888e-08, + "retain_loss": 0.02275395393371582, + "step": 800 + }, + { + "epoch": 0.9163802978235968, + "forget_loss": 1.4687527993828553e-07, + "retain_loss": 0.026991432532668114, + "step": 800 + }, + { + "epoch": 0.9163802978235968, + "forget_loss": 3.380711177669582e-07, + "retain_loss": 0.025429554283618927, + "step": 800 + }, + { + "epoch": 0.9278350515463918, + "grad_norm": 0.74609375, + "learning_rate": 2.6987327188940093e-05, + "loss": 0.027974823117256166, + "step": 810 + }, + { + "epoch": 0.9278350515463918, + "forget_loss": 2.2884126593680776e-08, + "retain_loss": 0.034118637442588806, + "step": 810 + }, + { + "epoch": 0.9278350515463918, + "forget_loss": 2.2747957473256974e-07, + "retain_loss": 0.016463886946439743, + "step": 810 + }, + { + "epoch": 0.9278350515463918, + "forget_loss": 2.8712694444266162e-08, + "retain_loss": 0.05120648816227913, + "step": 810 + }, + { + "epoch": 0.9278350515463918, + "forget_loss": 3.9451157363146194e-07, + "retain_loss": 0.015402361750602722, + "step": 810 + }, + { + "epoch": 0.9392898052691867, + "grad_norm": 0.232421875, + "learning_rate": 2.6699308755760372e-05, + "loss": 0.025883400440216066, + "step": 820 + }, + { + "epoch": 0.9392898052691867, + "forget_loss": 6.696375720594006e-08, + "retain_loss": 0.03956631198525429, + "step": 820 + }, + { + "epoch": 0.9392898052691867, + "forget_loss": 3.5951167376424564e-08, + "retain_loss": 0.024411948397755623, + "step": 820 + }, + { + "epoch": 0.9392898052691867, + "forget_loss": 1.2076303335106786e-07, + "retain_loss": 0.03519268333911896, + "step": 820 + }, + { + "epoch": 0.9392898052691867, + "forget_loss": 3.650887236972267e-08, + "retain_loss": 0.022809229791164398, + "step": 820 + }, + { + "epoch": 0.9507445589919816, + "grad_norm": 0.181640625, + "learning_rate": 2.6411290322580645e-05, + "loss": 0.026767241954803466, + "step": 830 + }, + { + "epoch": 0.9507445589919816, + "forget_loss": 1.1937875399326003e-07, + "retain_loss": 0.009928344748914242, + "step": 830 + }, + { + "epoch": 0.9507445589919816, + "forget_loss": 4.2366110619695974e-07, + "retain_loss": 0.010477675125002861, + "step": 830 + }, + { + "epoch": 0.9507445589919816, + "forget_loss": 9.538210576920392e-08, + "retain_loss": 0.030232908204197884, + "step": 830 + }, + { + "epoch": 0.9507445589919816, + "forget_loss": 4.3475404254422756e-07, + "retain_loss": 0.027472183108329773, + "step": 830 + }, + { + "epoch": 0.9621993127147767, + "grad_norm": 0.38671875, + "learning_rate": 2.6123271889400924e-05, + "loss": 0.026079732179641723, + "step": 840 + }, + { + "epoch": 0.9621993127147767, + "forget_loss": 9.560527303165145e-08, + "retain_loss": 0.035694848746061325, + "step": 840 + }, + { + "epoch": 0.9621993127147767, + "forget_loss": 9.010035739720479e-08, + "retain_loss": 0.02770412340760231, + "step": 840 + }, + { + "epoch": 0.9621993127147767, + "forget_loss": 1.2952777979080565e-06, + "retain_loss": 0.025091225281357765, + "step": 840 + }, + { + "epoch": 0.9621993127147767, + "forget_loss": 1.8988541228281974e-07, + "retain_loss": 0.02703693136572838, + "step": 840 + }, + { + "epoch": 0.9736540664375716, + "grad_norm": 0.62109375, + "learning_rate": 2.5835253456221196e-05, + "loss": 0.024774086475372315, + "step": 850 + }, + { + "epoch": 0.9736540664375716, + "forget_loss": 2.6555783350090678e-08, + "retain_loss": 0.033134713768959045, + "step": 850 + }, + { + "epoch": 0.9736540664375716, + "forget_loss": 2.7541537406250427e-07, + "retain_loss": 0.018386946991086006, + "step": 850 + }, + { + "epoch": 0.9736540664375716, + "forget_loss": 2.7093907561948072e-08, + "retain_loss": 0.022385219112038612, + "step": 850 + }, + { + "epoch": 0.9736540664375716, + "forget_loss": 6.318795442439296e-08, + "retain_loss": 0.03880416601896286, + "step": 850 + }, + { + "epoch": 0.9851088201603666, + "grad_norm": 0.57421875, + "learning_rate": 2.5547235023041475e-05, + "loss": 0.02385828346014023, + "step": 860 + }, + { + "epoch": 0.9851088201603666, + "forget_loss": 3.6007147485861424e-08, + "retain_loss": 0.03307569772005081, + "step": 860 + }, + { + "epoch": 0.9851088201603666, + "forget_loss": 8.178755450671815e-08, + "retain_loss": 0.02462468110024929, + "step": 860 + }, + { + "epoch": 0.9851088201603666, + "forget_loss": 6.628091853144724e-08, + "retain_loss": 0.020657110959291458, + "step": 860 + }, + { + "epoch": 0.9851088201603666, + "forget_loss": 6.628161486332829e-08, + "retain_loss": 0.032617855817079544, + "step": 860 + }, + { + "epoch": 0.9965635738831615, + "grad_norm": 0.236328125, + "learning_rate": 2.525921658986175e-05, + "loss": 0.024864019453525545, + "step": 870 + }, + { + "epoch": 0.9965635738831615, + "forget_loss": 7.505889243475394e-08, + "retain_loss": 0.01775062270462513, + "step": 870 + }, + { + "epoch": 0.9965635738831615, + "forget_loss": 1.0310161258075823e-07, + "retain_loss": 0.02414717711508274, + "step": 870 + }, + { + "epoch": 0.9965635738831615, + "forget_loss": 2.3729118936444138e-07, + "retain_loss": 0.027325890958309174, + "step": 870 + }, + { + "epoch": 0.9965635738831615, + "forget_loss": 2.6968228539203665e-08, + "retain_loss": 0.022969461977481842, + "step": 870 + }, + { + "epoch": 1.0080183276059564, + "grad_norm": 0.2041015625, + "learning_rate": 2.497119815668203e-05, + "loss": 0.025278234481811525, + "step": 880 + }, + { + "epoch": 1.0080183276059564, + "forget_loss": 1.2440519014944584e-07, + "retain_loss": 0.028113704174757004, + "step": 880 + }, + { + "epoch": 1.0080183276059564, + "forget_loss": 6.923085038579302e-08, + "retain_loss": 0.035193175077438354, + "step": 880 + }, + { + "epoch": 1.0080183276059564, + "forget_loss": 3.75761644022532e-08, + "retain_loss": 0.018518565222620964, + "step": 880 + }, + { + "epoch": 1.0080183276059564, + "forget_loss": 2.242090957338405e-08, + "retain_loss": 0.017261430621147156, + "step": 880 + }, + { + "epoch": 1.0194730813287514, + "grad_norm": 0.37109375, + "learning_rate": 2.4683179723502306e-05, + "loss": 0.024524253606796265, + "step": 890 + }, + { + "epoch": 1.0194730813287514, + "forget_loss": 3.599013496113912e-07, + "retain_loss": 0.034332528710365295, + "step": 890 + }, + { + "epoch": 1.0194730813287514, + "forget_loss": 1.7287271703025908e-07, + "retain_loss": 0.01914825290441513, + "step": 890 + }, + { + "epoch": 1.0194730813287514, + "forget_loss": 9.782493748389243e-08, + "retain_loss": 0.023516574874520302, + "step": 890 + }, + { + "epoch": 1.0194730813287514, + "forget_loss": 1.5106013506738236e-07, + "retain_loss": 0.019009225070476532, + "step": 890 + }, + { + "epoch": 1.0309278350515463, + "grad_norm": 0.1787109375, + "learning_rate": 2.4395161290322582e-05, + "loss": 0.024021840095520018, + "step": 900 + }, + { + "epoch": 1.0309278350515463, + "forget_loss": 2.6426999966133735e-07, + "retain_loss": 0.029771650210022926, + "step": 900 + }, + { + "epoch": 1.0309278350515463, + "forget_loss": 5.678279535459296e-08, + "retain_loss": 0.02570934407413006, + "step": 900 + }, + { + "epoch": 1.0309278350515463, + "forget_loss": 2.8987988898165895e-08, + "retain_loss": 0.02868933044373989, + "step": 900 + }, + { + "epoch": 1.0309278350515463, + "forget_loss": 1.6226587717937946e-07, + "retain_loss": 0.024329733103513718, + "step": 900 + }, + { + "epoch": 1.0423825887743414, + "grad_norm": 0.259765625, + "learning_rate": 2.4107142857142858e-05, + "loss": 0.02456314116716385, + "step": 910 + }, + { + "epoch": 1.0423825887743414, + "forget_loss": 3.358028379807365e-07, + "retain_loss": 0.014405038207769394, + "step": 910 + }, + { + "epoch": 1.0423825887743414, + "forget_loss": 8.003183893379173e-08, + "retain_loss": 0.012783446349203587, + "step": 910 + }, + { + "epoch": 1.0423825887743414, + "forget_loss": 3.088370874593238e-07, + "retain_loss": 0.04180136322975159, + "step": 910 + }, + { + "epoch": 1.0423825887743414, + "forget_loss": 1.6710265526853618e-07, + "retain_loss": 0.015366771258413792, + "step": 910 + }, + { + "epoch": 1.0538373424971363, + "grad_norm": 0.3359375, + "learning_rate": 2.3819124423963134e-05, + "loss": 0.023695509135723113, + "step": 920 + }, + { + "epoch": 1.0538373424971363, + "forget_loss": 1.9374034820884845e-07, + "retain_loss": 0.0453634150326252, + "step": 920 + }, + { + "epoch": 1.0538373424971363, + "forget_loss": 4.01810851258233e-08, + "retain_loss": 0.027023399248719215, + "step": 920 + }, + { + "epoch": 1.0538373424971363, + "forget_loss": 3.384187508004288e-08, + "retain_loss": 0.011590844951570034, + "step": 920 + }, + { + "epoch": 1.0538373424971363, + "forget_loss": 2.813683330771255e-08, + "retain_loss": 0.02153298258781433, + "step": 920 + }, + { + "epoch": 1.0652920962199313, + "grad_norm": 0.228515625, + "learning_rate": 2.3531105990783413e-05, + "loss": 0.023248623311519622, + "step": 930 + }, + { + "epoch": 1.0652920962199313, + "forget_loss": 3.62256820096718e-08, + "retain_loss": 0.04065975919365883, + "step": 930 + }, + { + "epoch": 1.0652920962199313, + "forget_loss": 1.747634712501167e-07, + "retain_loss": 0.022159142419695854, + "step": 930 + }, + { + "epoch": 1.0652920962199313, + "forget_loss": 8.5983149489266e-09, + "retain_loss": 0.011597687378525734, + "step": 930 + }, + { + "epoch": 1.0652920962199313, + "forget_loss": 2.4163021805634344e-08, + "retain_loss": 0.015060486271977425, + "step": 930 + }, + { + "epoch": 1.0767468499427262, + "grad_norm": 0.216796875, + "learning_rate": 2.324308755760369e-05, + "loss": 0.022622266411781312, + "step": 940 + }, + { + "epoch": 1.0767468499427262, + "forget_loss": 2.2849988567941182e-08, + "retain_loss": 0.014040598645806313, + "step": 940 + }, + { + "epoch": 1.0767468499427262, + "forget_loss": 1.0765910829491077e-08, + "retain_loss": 0.021398184821009636, + "step": 940 + }, + { + "epoch": 1.0767468499427262, + "forget_loss": 3.679618885144009e-07, + "retain_loss": 0.01932062953710556, + "step": 940 + }, + { + "epoch": 1.0767468499427262, + "forget_loss": 4.7389018931198734e-08, + "retain_loss": 0.03855086490511894, + "step": 940 + }, + { + "epoch": 1.088201603665521, + "grad_norm": 0.1669921875, + "learning_rate": 2.2955069124423965e-05, + "loss": 0.022816181182861328, + "step": 950 + }, + { + "epoch": 1.088201603665521, + "forget_loss": 5.09719377816964e-08, + "retain_loss": 0.01380919013172388, + "step": 950 + }, + { + "epoch": 1.088201603665521, + "forget_loss": 2.481048966274102e-08, + "retain_loss": 0.02061321772634983, + "step": 950 + }, + { + "epoch": 1.088201603665521, + "forget_loss": 3.024410943908151e-08, + "retain_loss": 0.021798167377710342, + "step": 950 + }, + { + "epoch": 1.088201603665521, + "forget_loss": 6.814799036192198e-08, + "retain_loss": 0.02376674860715866, + "step": 950 + }, + { + "epoch": 1.0996563573883162, + "grad_norm": 0.2099609375, + "learning_rate": 2.266705069124424e-05, + "loss": 0.023587432503700257, + "step": 960 + }, + { + "epoch": 1.0996563573883162, + "forget_loss": 7.863557982545899e-08, + "retain_loss": 0.034126508980989456, + "step": 960 + }, + { + "epoch": 1.0996563573883162, + "forget_loss": 1.2243476454898428e-08, + "retain_loss": 0.025233590975403786, + "step": 960 + }, + { + "epoch": 1.0996563573883162, + "forget_loss": 2.9529656941917892e-08, + "retain_loss": 0.026653651148080826, + "step": 960 + }, + { + "epoch": 1.0996563573883162, + "forget_loss": 1.9638031290014624e-07, + "retain_loss": 0.021439678966999054, + "step": 960 + }, + { + "epoch": 1.1111111111111112, + "grad_norm": 0.1318359375, + "learning_rate": 2.2379032258064516e-05, + "loss": 0.02087740898132324, + "step": 970 + }, + { + "epoch": 1.1111111111111112, + "forget_loss": 5.724940876916662e-08, + "retain_loss": 0.031759753823280334, + "step": 970 + }, + { + "epoch": 1.1111111111111112, + "forget_loss": 2.719896485814388e-08, + "retain_loss": 0.02280360274016857, + "step": 970 + }, + { + "epoch": 1.1111111111111112, + "forget_loss": 1.3328020997960266e-07, + "retain_loss": 0.034389056265354156, + "step": 970 + }, + { + "epoch": 1.1111111111111112, + "forget_loss": 2.331986337367198e-08, + "retain_loss": 0.027264058589935303, + "step": 970 + }, + { + "epoch": 1.122565864833906, + "grad_norm": 0.2255859375, + "learning_rate": 2.2091013824884792e-05, + "loss": 0.02323000878095627, + "step": 980 + }, + { + "epoch": 1.122565864833906, + "forget_loss": 6.373924179570167e-07, + "retain_loss": 0.022899720817804337, + "step": 980 + }, + { + "epoch": 1.122565864833906, + "forget_loss": 8.226674452771476e-08, + "retain_loss": 0.017953844740986824, + "step": 980 + }, + { + "epoch": 1.122565864833906, + "forget_loss": 3.995974395820667e-08, + "retain_loss": 0.023313166573643684, + "step": 980 + }, + { + "epoch": 1.122565864833906, + "forget_loss": 5.536628577829106e-08, + "retain_loss": 0.02542729862034321, + "step": 980 + }, + { + "epoch": 1.134020618556701, + "grad_norm": 0.142578125, + "learning_rate": 2.1802995391705068e-05, + "loss": 0.02163739502429962, + "step": 990 + }, + { + "epoch": 1.134020618556701, + "forget_loss": 4.036724732259245e-08, + "retain_loss": 0.03384210169315338, + "step": 990 + }, + { + "epoch": 1.134020618556701, + "forget_loss": 1.0333982913834916e-07, + "retain_loss": 0.021868150681257248, + "step": 990 + }, + { + "epoch": 1.134020618556701, + "forget_loss": 5.9504564831058815e-08, + "retain_loss": 0.010602345690131187, + "step": 990 + }, + { + "epoch": 1.134020618556701, + "forget_loss": 4.604233794225365e-08, + "retain_loss": 0.03545906022191048, + "step": 990 + }, + { + "epoch": 1.145475372279496, + "grad_norm": 0.1796875, + "learning_rate": 2.1514976958525347e-05, + "loss": 0.023172636330127717, + "step": 1000 + }, + { + "epoch": 1.145475372279496, + "forget_loss": 5.493423032021383e-07, + "retain_loss": 0.032305389642715454, + "step": 1000 + }, + { + "epoch": 1.145475372279496, + "forget_loss": 2.8481522917900293e-07, + "retain_loss": 0.028948910534381866, + "step": 1000 + }, + { + "epoch": 1.145475372279496, + "forget_loss": 1.4940823689357785e-08, + "retain_loss": 0.051252782344818115, + "step": 1000 + }, + { + "epoch": 1.145475372279496, + "forget_loss": 2.7136126234950098e-08, + "retain_loss": 0.013740372844040394, + "step": 1000 + }, + { + "epoch": 1.1569301260022908, + "grad_norm": 0.1748046875, + "learning_rate": 2.1226958525345623e-05, + "loss": 0.023918163776397706, + "step": 1010 + }, + { + "epoch": 1.1569301260022908, + "forget_loss": 5.731934393793381e-08, + "retain_loss": 0.022277191281318665, + "step": 1010 + }, + { + "epoch": 1.1569301260022908, + "forget_loss": 6.457765522327463e-08, + "retain_loss": 0.01483561284840107, + "step": 1010 + }, + { + "epoch": 1.1569301260022908, + "forget_loss": 2.3290189332669797e-08, + "retain_loss": 0.011415845714509487, + "step": 1010 + }, + { + "epoch": 1.1569301260022908, + "forget_loss": 8.277501706288604e-08, + "retain_loss": 0.0220047514885664, + "step": 1010 + }, + { + "epoch": 1.168384879725086, + "grad_norm": 0.1982421875, + "learning_rate": 2.09389400921659e-05, + "loss": 0.024459759891033172, + "step": 1020 + }, + { + "epoch": 1.168384879725086, + "forget_loss": 2.648368280233626e-08, + "retain_loss": 0.017724836245179176, + "step": 1020 + }, + { + "epoch": 1.168384879725086, + "forget_loss": 7.936857571166911e-08, + "retain_loss": 0.016395514830946922, + "step": 1020 + }, + { + "epoch": 1.168384879725086, + "forget_loss": 3.1398002420246485e-08, + "retain_loss": 0.024634411558508873, + "step": 1020 + }, + { + "epoch": 1.168384879725086, + "forget_loss": 4.1696470276519904e-08, + "retain_loss": 0.02067175693809986, + "step": 1020 + }, + { + "epoch": 1.179839633447881, + "grad_norm": 0.2109375, + "learning_rate": 2.0650921658986175e-05, + "loss": 0.021289190649986266, + "step": 1030 + }, + { + "epoch": 1.179839633447881, + "forget_loss": 7.644550947816242e-08, + "retain_loss": 0.018215566873550415, + "step": 1030 + }, + { + "epoch": 1.179839633447881, + "forget_loss": 9.573097869974845e-09, + "retain_loss": 0.02367250621318817, + "step": 1030 + }, + { + "epoch": 1.179839633447881, + "forget_loss": 1.727073062340878e-08, + "retain_loss": 0.029460415244102478, + "step": 1030 + }, + { + "epoch": 1.179839633447881, + "forget_loss": 2.794078568513214e-08, + "retain_loss": 0.015188030898571014, + "step": 1030 + }, + { + "epoch": 1.1912943871706758, + "grad_norm": 0.1923828125, + "learning_rate": 2.0362903225806454e-05, + "loss": 0.021110835671424865, + "step": 1040 + }, + { + "epoch": 1.1912943871706758, + "forget_loss": 2.4121279196265277e-08, + "retain_loss": 0.017713716253638268, + "step": 1040 + }, + { + "epoch": 1.1912943871706758, + "forget_loss": 9.602725015156466e-08, + "retain_loss": 0.025713184848427773, + "step": 1040 + }, + { + "epoch": 1.1912943871706758, + "forget_loss": 2.1239122816041345e-07, + "retain_loss": 0.017294256016612053, + "step": 1040 + }, + { + "epoch": 1.1912943871706758, + "forget_loss": 2.7992754780825635e-08, + "retain_loss": 0.01869981549680233, + "step": 1040 + }, + { + "epoch": 1.2027491408934707, + "grad_norm": 0.1611328125, + "learning_rate": 2.007488479262673e-05, + "loss": 0.02240833044052124, + "step": 1050 + }, + { + "epoch": 1.2027491408934707, + "forget_loss": 5.0892207781316756e-08, + "retain_loss": 0.02070891670882702, + "step": 1050 + }, + { + "epoch": 1.2027491408934707, + "forget_loss": 2.7021894055678786e-08, + "retain_loss": 0.016385093331336975, + "step": 1050 + }, + { + "epoch": 1.2027491408934707, + "forget_loss": 1.3184821057166118e-07, + "retain_loss": 0.012673438526690006, + "step": 1050 + }, + { + "epoch": 1.2027491408934707, + "forget_loss": 2.6013424303528154e-07, + "retain_loss": 0.048716556280851364, + "step": 1050 + }, + { + "epoch": 1.2142038946162657, + "grad_norm": 0.1552734375, + "learning_rate": 1.9786866359447005e-05, + "loss": 0.02106960415840149, + "step": 1060 + }, + { + "epoch": 1.2142038946162657, + "forget_loss": 6.823061937666353e-08, + "retain_loss": 0.0206641536206007, + "step": 1060 + }, + { + "epoch": 1.2142038946162657, + "forget_loss": 7.832964854515012e-08, + "retain_loss": 0.017997639253735542, + "step": 1060 + }, + { + "epoch": 1.2142038946162657, + "forget_loss": 1.1000237165603721e-08, + "retain_loss": 0.026349617168307304, + "step": 1060 + }, + { + "epoch": 1.2142038946162657, + "forget_loss": 2.8729951750960936e-08, + "retain_loss": 0.03591274842619896, + "step": 1060 + }, + { + "epoch": 1.2256586483390608, + "grad_norm": 0.1796875, + "learning_rate": 1.9498847926267285e-05, + "loss": 0.023325277864933013, + "step": 1070 + }, + { + "epoch": 1.2256586483390608, + "forget_loss": 5.677220116240278e-08, + "retain_loss": 0.017587844282388687, + "step": 1070 + }, + { + "epoch": 1.2256586483390608, + "forget_loss": 1.0990573429126016e-07, + "retain_loss": 0.014925125055015087, + "step": 1070 + }, + { + "epoch": 1.2256586483390608, + "forget_loss": 6.773029781470541e-08, + "retain_loss": 0.023775167763233185, + "step": 1070 + }, + { + "epoch": 1.2256586483390608, + "forget_loss": 2.3923700354089306e-08, + "retain_loss": 0.028247609734535217, + "step": 1070 + }, + { + "epoch": 1.2371134020618557, + "grad_norm": 0.189453125, + "learning_rate": 1.921082949308756e-05, + "loss": 0.022065652906894682, + "step": 1080 + }, + { + "epoch": 1.2371134020618557, + "forget_loss": 6.123426654625064e-09, + "retain_loss": 0.011591405607759953, + "step": 1080 + }, + { + "epoch": 1.2371134020618557, + "forget_loss": 3.527255643120952e-08, + "retain_loss": 0.015910839661955833, + "step": 1080 + }, + { + "epoch": 1.2371134020618557, + "forget_loss": 9.751467189289542e-08, + "retain_loss": 0.007821759209036827, + "step": 1080 + }, + { + "epoch": 1.2371134020618557, + "forget_loss": 2.560214170443942e-06, + "retain_loss": 0.021734777837991714, + "step": 1080 + }, + { + "epoch": 1.2485681557846506, + "grad_norm": 0.1357421875, + "learning_rate": 1.8922811059907836e-05, + "loss": 0.021877190470695494, + "step": 1090 + }, + { + "epoch": 1.2485681557846506, + "forget_loss": 8.14992162645467e-08, + "retain_loss": 0.024061294272542, + "step": 1090 + }, + { + "epoch": 1.2485681557846506, + "forget_loss": 5.5814986410496203e-08, + "retain_loss": 0.027457626536488533, + "step": 1090 + }, + { + "epoch": 1.2485681557846506, + "forget_loss": 7.33262709218252e-07, + "retain_loss": 0.016484102234244347, + "step": 1090 + }, + { + "epoch": 1.2485681557846506, + "forget_loss": 1.206726665259339e-07, + "retain_loss": 0.018026303499937057, + "step": 1090 + }, + { + "epoch": 1.2600229095074456, + "grad_norm": 0.1630859375, + "learning_rate": 1.8634792626728112e-05, + "loss": 0.023145848512649538, + "step": 1100 + }, + { + "epoch": 1.2600229095074456, + "forget_loss": 4.808517672927337e-08, + "retain_loss": 0.02503185160458088, + "step": 1100 + }, + { + "epoch": 1.2600229095074456, + "forget_loss": 6.350063586069155e-08, + "retain_loss": 0.01136860717087984, + "step": 1100 + }, + { + "epoch": 1.2600229095074456, + "forget_loss": 7.402125845601404e-08, + "retain_loss": 0.023490875959396362, + "step": 1100 + }, + { + "epoch": 1.2600229095074456, + "forget_loss": 1.3122317454872245e-08, + "retain_loss": 0.009062117896974087, + "step": 1100 + }, + { + "epoch": 1.2714776632302405, + "grad_norm": 0.201171875, + "learning_rate": 1.8346774193548388e-05, + "loss": 0.025184395909309387, + "step": 1110 + }, + { + "epoch": 1.2714776632302405, + "forget_loss": 6.530651130987053e-09, + "retain_loss": 0.010407781228423119, + "step": 1110 + }, + { + "epoch": 1.2714776632302405, + "forget_loss": 9.545535561983343e-08, + "retain_loss": 0.01965162716805935, + "step": 1110 + }, + { + "epoch": 1.2714776632302405, + "forget_loss": 8.817356622614625e-09, + "retain_loss": 0.03341147303581238, + "step": 1110 + }, + { + "epoch": 1.2714776632302405, + "forget_loss": 3.525239478108233e-08, + "retain_loss": 0.025235410779714584, + "step": 1110 + }, + { + "epoch": 1.2829324169530354, + "grad_norm": 0.181640625, + "learning_rate": 1.8058755760368664e-05, + "loss": 0.021285700798034667, + "step": 1120 + }, + { + "epoch": 1.2829324169530354, + "forget_loss": 2.0350250906631118e-07, + "retain_loss": 0.02628685161471367, + "step": 1120 + }, + { + "epoch": 1.2829324169530354, + "forget_loss": 5.7581591050848147e-08, + "retain_loss": 0.03228295221924782, + "step": 1120 + }, + { + "epoch": 1.2829324169530354, + "forget_loss": 2.3273013738389636e-08, + "retain_loss": 0.02587534859776497, + "step": 1120 + }, + { + "epoch": 1.2829324169530354, + "forget_loss": 1.0459762478376433e-07, + "retain_loss": 0.020380979403853416, + "step": 1120 + }, + { + "epoch": 1.2943871706758305, + "grad_norm": 0.203125, + "learning_rate": 1.777073732718894e-05, + "loss": 0.02414794862270355, + "step": 1130 + }, + { + "epoch": 1.2943871706758305, + "forget_loss": 2.0749947537979097e-08, + "retain_loss": 0.027595985680818558, + "step": 1130 + }, + { + "epoch": 1.2943871706758305, + "forget_loss": 1.0538060735143517e-07, + "retain_loss": 0.03342067450284958, + "step": 1130 + }, + { + "epoch": 1.2943871706758305, + "forget_loss": 1.0734373745435732e-07, + "retain_loss": 0.0166862141340971, + "step": 1130 + }, + { + "epoch": 1.2943871706758305, + "forget_loss": 1.4255629565695926e-08, + "retain_loss": 0.013674532994627953, + "step": 1130 + }, + { + "epoch": 1.3058419243986255, + "grad_norm": 0.1845703125, + "learning_rate": 1.748271889400922e-05, + "loss": 0.022767890989780427, + "step": 1140 + }, + { + "epoch": 1.3058419243986255, + "forget_loss": 1.1125802501510407e-08, + "retain_loss": 0.021995795890688896, + "step": 1140 + }, + { + "epoch": 1.3058419243986255, + "forget_loss": 2.5488265720241543e-08, + "retain_loss": 0.01751900650560856, + "step": 1140 + }, + { + "epoch": 1.3058419243986255, + "forget_loss": 1.843478969476564e-08, + "retain_loss": 0.026673439890146255, + "step": 1140 + }, + { + "epoch": 1.3058419243986255, + "forget_loss": 1.5914393713956088e-07, + "retain_loss": 0.016024751588702202, + "step": 1140 + }, + { + "epoch": 1.3172966781214204, + "grad_norm": 0.2158203125, + "learning_rate": 1.7194700460829495e-05, + "loss": 0.0235839381814003, + "step": 1150 + }, + { + "epoch": 1.3172966781214204, + "forget_loss": 1.8218237585188035e-08, + "retain_loss": 0.014317602850496769, + "step": 1150 + }, + { + "epoch": 1.3172966781214204, + "forget_loss": 1.9948769747202277e-08, + "retain_loss": 0.021403903141617775, + "step": 1150 + }, + { + "epoch": 1.3172966781214204, + "forget_loss": 7.099181242864461e-09, + "retain_loss": 0.030618898570537567, + "step": 1150 + }, + { + "epoch": 1.3172966781214204, + "forget_loss": 5.6788714175581845e-08, + "retain_loss": 0.01884060725569725, + "step": 1150 + }, + { + "epoch": 1.3287514318442153, + "grad_norm": 0.203125, + "learning_rate": 1.690668202764977e-05, + "loss": 0.02227586954832077, + "step": 1160 + }, + { + "epoch": 1.3287514318442153, + "forget_loss": 1.0105923209380308e-08, + "retain_loss": 0.028436832129955292, + "step": 1160 + }, + { + "epoch": 1.3287514318442153, + "forget_loss": 5.427875180430419e-07, + "retain_loss": 0.011484701186418533, + "step": 1160 + }, + { + "epoch": 1.3287514318442153, + "forget_loss": 6.005623021110296e-08, + "retain_loss": 0.007445945404469967, + "step": 1160 + }, + { + "epoch": 1.3287514318442153, + "forget_loss": 3.227989253673513e-08, + "retain_loss": 0.031919676810503006, + "step": 1160 + }, + { + "epoch": 1.3402061855670104, + "grad_norm": 0.197265625, + "learning_rate": 1.6618663594470046e-05, + "loss": 0.023313036561012267, + "step": 1170 + }, + { + "epoch": 1.3402061855670104, + "forget_loss": 2.445302449416431e-08, + "retain_loss": 0.018606863915920258, + "step": 1170 + }, + { + "epoch": 1.3402061855670104, + "forget_loss": 3.023370425125904e-07, + "retain_loss": 0.0139198899269104, + "step": 1170 + }, + { + "epoch": 1.3402061855670104, + "forget_loss": 4.2964636293163494e-08, + "retain_loss": 0.020331142470240593, + "step": 1170 + }, + { + "epoch": 1.3402061855670104, + "forget_loss": 1.9158818531650468e-08, + "retain_loss": 0.029496945440769196, + "step": 1170 + }, + { + "epoch": 1.3516609392898054, + "grad_norm": 0.1708984375, + "learning_rate": 1.6330645161290322e-05, + "loss": 0.021104805171489716, + "step": 1180 + }, + { + "epoch": 1.3516609392898054, + "forget_loss": 4.3303703733954535e-08, + "retain_loss": 0.014237130992114544, + "step": 1180 + }, + { + "epoch": 1.3516609392898054, + "forget_loss": 7.990769823607025e-09, + "retain_loss": 0.04001491144299507, + "step": 1180 + }, + { + "epoch": 1.3516609392898054, + "forget_loss": 9.303267489713107e-09, + "retain_loss": 0.027049237862229347, + "step": 1180 + }, + { + "epoch": 1.3516609392898054, + "forget_loss": 5.501159971998959e-09, + "retain_loss": 0.005112822633236647, + "step": 1180 + }, + { + "epoch": 1.3631156930126003, + "grad_norm": 0.1884765625, + "learning_rate": 1.6042626728110598e-05, + "loss": 0.023045459389686586, + "step": 1190 + }, + { + "epoch": 1.3631156930126003, + "forget_loss": 2.7066937136055458e-08, + "retain_loss": 0.020094262436032295, + "step": 1190 + }, + { + "epoch": 1.3631156930126003, + "forget_loss": 1.7727275647416718e-08, + "retain_loss": 0.041554100811481476, + "step": 1190 + }, + { + "epoch": 1.3631156930126003, + "forget_loss": 3.883479759991815e-09, + "retain_loss": 0.01996440999209881, + "step": 1190 + }, + { + "epoch": 1.3631156930126003, + "forget_loss": 8.269919860026675e-09, + "retain_loss": 0.025666773319244385, + "step": 1190 + }, + { + "epoch": 1.3745704467353952, + "grad_norm": 0.1923828125, + "learning_rate": 1.5754608294930877e-05, + "loss": 0.023423218727111818, + "step": 1200 + }, + { + "epoch": 1.3745704467353952, + "forget_loss": 1.748937350498636e-08, + "retain_loss": 0.018628757447004318, + "step": 1200 + }, + { + "epoch": 1.3745704467353952, + "forget_loss": 1.6336418440232592e-08, + "retain_loss": 0.03357892856001854, + "step": 1200 + }, + { + "epoch": 1.3745704467353952, + "forget_loss": 3.25022995184554e-08, + "retain_loss": 0.01277889497578144, + "step": 1200 + }, + { + "epoch": 1.3745704467353952, + "forget_loss": 7.0995285206265635e-09, + "retain_loss": 0.02935495600104332, + "step": 1200 + }, + { + "epoch": 1.38602520045819, + "grad_norm": 0.16796875, + "learning_rate": 1.5466589861751153e-05, + "loss": 0.023562929034233092, + "step": 1210 + }, + { + "epoch": 1.38602520045819, + "forget_loss": 7.78369457776762e-09, + "retain_loss": 0.021808171644806862, + "step": 1210 + }, + { + "epoch": 1.38602520045819, + "forget_loss": 2.3151198735860135e-09, + "retain_loss": 0.01690870337188244, + "step": 1210 + }, + { + "epoch": 1.38602520045819, + "forget_loss": 9.253830479671876e-10, + "retain_loss": 0.01985970325767994, + "step": 1210 + }, + { + "epoch": 1.38602520045819, + "forget_loss": 6.011931397154058e-09, + "retain_loss": 0.021763188764452934, + "step": 1210 + }, + { + "epoch": 1.397479954180985, + "grad_norm": 0.1572265625, + "learning_rate": 1.5178571428571429e-05, + "loss": 0.020696282386779785, + "step": 1220 + }, + { + "epoch": 1.397479954180985, + "forget_loss": 2.3503828217030787e-08, + "retain_loss": 0.02087138406932354, + "step": 1220 + }, + { + "epoch": 1.397479954180985, + "forget_loss": 2.5618970056484613e-08, + "retain_loss": 0.026184234768152237, + "step": 1220 + }, + { + "epoch": 1.397479954180985, + "forget_loss": 8.21958678898227e-09, + "retain_loss": 0.0283048078417778, + "step": 1220 + }, + { + "epoch": 1.397479954180985, + "forget_loss": 6.894857573769286e-09, + "retain_loss": 0.02001921460032463, + "step": 1220 + }, + { + "epoch": 1.40893470790378, + "grad_norm": 0.203125, + "learning_rate": 1.4890552995391705e-05, + "loss": 0.022527270019054413, + "step": 1230 + }, + { + "epoch": 1.40893470790378, + "forget_loss": 2.1096978741752537e-07, + "retain_loss": 0.012916534207761288, + "step": 1230 + }, + { + "epoch": 1.40893470790378, + "forget_loss": 1.9281825913708417e-08, + "retain_loss": 0.022061077877879143, + "step": 1230 + }, + { + "epoch": 1.40893470790378, + "forget_loss": 6.622970971648101e-08, + "retain_loss": 0.02101759985089302, + "step": 1230 + }, + { + "epoch": 1.40893470790378, + "forget_loss": 1.26990897797441e-08, + "retain_loss": 0.023891810327768326, + "step": 1230 + }, + { + "epoch": 1.420389461626575, + "grad_norm": 0.2119140625, + "learning_rate": 1.4602534562211984e-05, + "loss": 0.023477652668952943, + "step": 1240 + }, + { + "epoch": 1.420389461626575, + "forget_loss": 4.966549838059109e-08, + "retain_loss": 0.03463604301214218, + "step": 1240 + }, + { + "epoch": 1.420389461626575, + "forget_loss": 4.284724752778857e-08, + "retain_loss": 0.02400204725563526, + "step": 1240 + }, + { + "epoch": 1.420389461626575, + "forget_loss": 2.6644977779710644e-08, + "retain_loss": 0.024855146184563637, + "step": 1240 + }, + { + "epoch": 1.420389461626575, + "forget_loss": 1.1419896850384248e-07, + "retain_loss": 0.02515013888478279, + "step": 1240 + }, + { + "epoch": 1.43184421534937, + "grad_norm": 0.2138671875, + "learning_rate": 1.431451612903226e-05, + "loss": 0.02142278999090195, + "step": 1250 + }, + { + "epoch": 1.43184421534937, + "forget_loss": 6.727535151185293e-08, + "retain_loss": 0.03439854457974434, + "step": 1250 + }, + { + "epoch": 1.43184421534937, + "forget_loss": 2.731752779538965e-08, + "retain_loss": 0.018644528463482857, + "step": 1250 + }, + { + "epoch": 1.43184421534937, + "forget_loss": 2.9547951641006875e-08, + "retain_loss": 0.016844170168042183, + "step": 1250 + }, + { + "epoch": 1.43184421534937, + "forget_loss": 2.0531512490151727e-08, + "retain_loss": 0.030812717974185944, + "step": 1250 + }, + { + "epoch": 1.443298969072165, + "grad_norm": 0.1826171875, + "learning_rate": 1.4026497695852537e-05, + "loss": 0.02201966047286987, + "step": 1260 + }, + { + "epoch": 1.443298969072165, + "forget_loss": 3.202220000275702e-08, + "retain_loss": 0.028422845527529716, + "step": 1260 + }, + { + "epoch": 1.443298969072165, + "forget_loss": 1.8043699867575924e-08, + "retain_loss": 0.030724935233592987, + "step": 1260 + }, + { + "epoch": 1.443298969072165, + "forget_loss": 9.87688295595035e-08, + "retain_loss": 0.026053862646222115, + "step": 1260 + }, + { + "epoch": 1.443298969072165, + "forget_loss": 3.2861024124031246e-08, + "retain_loss": 0.025916006416082382, + "step": 1260 + }, + { + "epoch": 1.4547537227949598, + "grad_norm": 0.1826171875, + "learning_rate": 1.3738479262672813e-05, + "loss": 0.02156946063041687, + "step": 1270 + }, + { + "epoch": 1.4547537227949598, + "forget_loss": 2.1402931338343478e-07, + "retain_loss": 0.023167060688138008, + "step": 1270 + }, + { + "epoch": 1.4547537227949598, + "forget_loss": 4.80714135164817e-08, + "retain_loss": 0.012924721464514732, + "step": 1270 + }, + { + "epoch": 1.4547537227949598, + "forget_loss": 8.835430165277103e-09, + "retain_loss": 0.020214807242155075, + "step": 1270 + }, + { + "epoch": 1.4547537227949598, + "forget_loss": 3.223247801997786e-08, + "retain_loss": 0.02204856276512146, + "step": 1270 + }, + { + "epoch": 1.466208476517755, + "grad_norm": 0.1875, + "learning_rate": 1.3450460829493089e-05, + "loss": 0.021540743112564088, + "step": 1280 + }, + { + "epoch": 1.466208476517755, + "forget_loss": 6.666966356760895e-08, + "retain_loss": 0.028293240815401077, + "step": 1280 + }, + { + "epoch": 1.466208476517755, + "forget_loss": 3.187368591284212e-08, + "retain_loss": 0.03067215532064438, + "step": 1280 + }, + { + "epoch": 1.466208476517755, + "forget_loss": 8.677272234081101e-08, + "retain_loss": 0.01857157237827778, + "step": 1280 + }, + { + "epoch": 1.466208476517755, + "forget_loss": 3.4804223503215326e-08, + "retain_loss": 0.01711919903755188, + "step": 1280 + }, + { + "epoch": 1.47766323024055, + "grad_norm": 0.1748046875, + "learning_rate": 1.3162442396313365e-05, + "loss": 0.023854960501194, + "step": 1290 + }, + { + "epoch": 1.47766323024055, + "forget_loss": 1.4682684934541612e-07, + "retain_loss": 0.013613694347441196, + "step": 1290 + }, + { + "epoch": 1.47766323024055, + "forget_loss": 2.4446702440172885e-08, + "retain_loss": 0.03469227999448776, + "step": 1290 + }, + { + "epoch": 1.47766323024055, + "forget_loss": 1.4488199084894404e-08, + "retain_loss": 0.016956379637122154, + "step": 1290 + }, + { + "epoch": 1.47766323024055, + "forget_loss": 5.782320400271601e-08, + "retain_loss": 0.01388214249163866, + "step": 1290 + }, + { + "epoch": 1.4891179839633448, + "grad_norm": 0.19921875, + "learning_rate": 1.2874423963133642e-05, + "loss": 0.022388391196727753, + "step": 1300 + }, + { + "epoch": 1.4891179839633448, + "forget_loss": 1.731160281792654e-08, + "retain_loss": 0.020429806783795357, + "step": 1300 + }, + { + "epoch": 1.4891179839633448, + "forget_loss": 4.9649994338096803e-08, + "retain_loss": 0.01567276380956173, + "step": 1300 + }, + { + "epoch": 1.4891179839633448, + "forget_loss": 1.094394903589091e-07, + "retain_loss": 0.016468234360218048, + "step": 1300 + }, + { + "epoch": 1.4891179839633448, + "forget_loss": 2.912297247803508e-08, + "retain_loss": 0.022291528061032295, + "step": 1300 + }, + { + "epoch": 1.5005727376861397, + "grad_norm": 0.205078125, + "learning_rate": 1.2586405529953918e-05, + "loss": 0.022077234089374544, + "step": 1310 + }, + { + "epoch": 1.5005727376861397, + "forget_loss": 1.0040228914931504e-07, + "retain_loss": 0.017616135999560356, + "step": 1310 + }, + { + "epoch": 1.5005727376861397, + "forget_loss": 4.605796277701302e-08, + "retain_loss": 0.03136283531785011, + "step": 1310 + }, + { + "epoch": 1.5005727376861397, + "forget_loss": 5.563991933854595e-08, + "retain_loss": 0.0219328124076128, + "step": 1310 + }, + { + "epoch": 1.5005727376861397, + "forget_loss": 1.8549467739603642e-08, + "retain_loss": 0.0195622481405735, + "step": 1310 + }, + { + "epoch": 1.5120274914089347, + "grad_norm": 0.23046875, + "learning_rate": 1.2298387096774194e-05, + "loss": 0.02453242689371109, + "step": 1320 + }, + { + "epoch": 1.5120274914089347, + "forget_loss": 1.5807668063416713e-08, + "retain_loss": 0.01336264330893755, + "step": 1320 + }, + { + "epoch": 1.5120274914089347, + "forget_loss": 3.320761265968031e-08, + "retain_loss": 0.01914050243794918, + "step": 1320 + }, + { + "epoch": 1.5120274914089347, + "forget_loss": 3.8064783325353346e-08, + "retain_loss": 0.02273029275238514, + "step": 1320 + }, + { + "epoch": 1.5120274914089347, + "forget_loss": 1.188167964016884e-08, + "retain_loss": 0.018610704690217972, + "step": 1320 + }, + { + "epoch": 1.5234822451317296, + "grad_norm": 0.1484375, + "learning_rate": 1.2010368663594471e-05, + "loss": 0.02155406177043915, + "step": 1330 + }, + { + "epoch": 1.5234822451317296, + "forget_loss": 4.999679390493839e-07, + "retain_loss": 0.03529933840036392, + "step": 1330 + }, + { + "epoch": 1.5234822451317296, + "forget_loss": 1.7155528553303157e-08, + "retain_loss": 0.0193784236907959, + "step": 1330 + }, + { + "epoch": 1.5234822451317296, + "forget_loss": 4.2701351787854946e-08, + "retain_loss": 0.023776818066835403, + "step": 1330 + }, + { + "epoch": 1.5234822451317296, + "forget_loss": 4.939164099937443e-08, + "retain_loss": 0.025353359058499336, + "step": 1330 + }, + { + "epoch": 1.5349369988545245, + "grad_norm": 0.216796875, + "learning_rate": 1.1722350230414747e-05, + "loss": 0.02174515575170517, + "step": 1340 + }, + { + "epoch": 1.5349369988545245, + "forget_loss": 1.5110476425661545e-08, + "retain_loss": 0.027764471247792244, + "step": 1340 + }, + { + "epoch": 1.5349369988545245, + "forget_loss": 1.2457677556199087e-08, + "retain_loss": 0.022536301985383034, + "step": 1340 + }, + { + "epoch": 1.5349369988545245, + "forget_loss": 9.526234379109155e-08, + "retain_loss": 0.022036096081137657, + "step": 1340 + }, + { + "epoch": 1.5349369988545245, + "forget_loss": 4.588682145367784e-08, + "retain_loss": 0.026010531932115555, + "step": 1340 + }, + { + "epoch": 1.5463917525773194, + "grad_norm": 0.169921875, + "learning_rate": 1.1434331797235023e-05, + "loss": 0.02197621315717697, + "step": 1350 + }, + { + "epoch": 1.5463917525773194, + "forget_loss": 4.101625705743572e-08, + "retain_loss": 0.020112277939915657, + "step": 1350 + }, + { + "epoch": 1.5463917525773194, + "forget_loss": 2.926245024070795e-09, + "retain_loss": 0.0265397559851408, + "step": 1350 + }, + { + "epoch": 1.5463917525773194, + "forget_loss": 2.3485831945890823e-08, + "retain_loss": 0.014467987231910229, + "step": 1350 + }, + { + "epoch": 1.5463917525773194, + "forget_loss": 9.590316096819151e-08, + "retain_loss": 0.01878652721643448, + "step": 1350 + }, + { + "epoch": 1.5578465063001146, + "grad_norm": 0.166015625, + "learning_rate": 1.11463133640553e-05, + "loss": 0.020632724463939666, + "step": 1360 + }, + { + "epoch": 1.5578465063001146, + "forget_loss": 1.4095779654610396e-08, + "retain_loss": 0.02716868370771408, + "step": 1360 + }, + { + "epoch": 1.5578465063001146, + "forget_loss": 3.688577621119293e-08, + "retain_loss": 0.02675730548799038, + "step": 1360 + }, + { + "epoch": 1.5578465063001146, + "forget_loss": 1.3284892297349415e-08, + "retain_loss": 0.02028178796172142, + "step": 1360 + }, + { + "epoch": 1.5578465063001146, + "forget_loss": 4.021210919802343e-09, + "retain_loss": 0.024784252047538757, + "step": 1360 + }, + { + "epoch": 1.5693012600229095, + "grad_norm": 0.162109375, + "learning_rate": 1.0858294930875576e-05, + "loss": 0.021194690465927125, + "step": 1370 + }, + { + "epoch": 1.5693012600229095, + "forget_loss": 1.9185983912706206e-09, + "retain_loss": 0.020529447123408318, + "step": 1370 + }, + { + "epoch": 1.5693012600229095, + "forget_loss": 4.003663178764327e-09, + "retain_loss": 0.024104757234454155, + "step": 1370 + }, + { + "epoch": 1.5693012600229095, + "forget_loss": 2.1751702661276795e-08, + "retain_loss": 0.02209138683974743, + "step": 1370 + }, + { + "epoch": 1.5693012600229095, + "forget_loss": 1.2200887411495387e-08, + "retain_loss": 0.022040190175175667, + "step": 1370 + }, + { + "epoch": 1.5807560137457046, + "grad_norm": 0.1923828125, + "learning_rate": 1.0570276497695852e-05, + "loss": 0.020398758351802826, + "step": 1380 + }, + { + "epoch": 1.5807560137457046, + "forget_loss": 2.143367439089161e-08, + "retain_loss": 0.01812758296728134, + "step": 1380 + }, + { + "epoch": 1.5807560137457046, + "forget_loss": 4.7683954562671715e-08, + "retain_loss": 0.01754043623805046, + "step": 1380 + }, + { + "epoch": 1.5807560137457046, + "forget_loss": 3.031481199400332e-08, + "retain_loss": 0.03185281530022621, + "step": 1380 + }, + { + "epoch": 1.5807560137457046, + "forget_loss": 2.201096727105778e-08, + "retain_loss": 0.037519581615924835, + "step": 1380 + }, + { + "epoch": 1.5922107674684995, + "grad_norm": 0.205078125, + "learning_rate": 1.028225806451613e-05, + "loss": 0.02095855325460434, + "step": 1390 + }, + { + "epoch": 1.5922107674684995, + "forget_loss": 8.277861240912898e-08, + "retain_loss": 0.027487264946103096, + "step": 1390 + }, + { + "epoch": 1.5922107674684995, + "forget_loss": 1.217109542039907e-07, + "retain_loss": 0.020043643191456795, + "step": 1390 + }, + { + "epoch": 1.5922107674684995, + "forget_loss": 2.6641883366096408e-08, + "retain_loss": 0.019022131338715553, + "step": 1390 + }, + { + "epoch": 1.5922107674684995, + "forget_loss": 2.9094383791061773e-08, + "retain_loss": 0.018960099667310715, + "step": 1390 + }, + { + "epoch": 1.6036655211912945, + "grad_norm": 0.171875, + "learning_rate": 9.994239631336407e-06, + "loss": 0.021135611832141875, + "step": 1400 + }, + { + "epoch": 1.6036655211912945, + "forget_loss": 6.217426573584817e-08, + "retain_loss": 0.016695497557520866, + "step": 1400 + }, + { + "epoch": 1.6036655211912945, + "forget_loss": 4.8250729633991796e-08, + "retain_loss": 0.021785398945212364, + "step": 1400 + }, + { + "epoch": 1.6036655211912945, + "forget_loss": 7.732725748610392e-08, + "retain_loss": 0.025909021496772766, + "step": 1400 + }, + { + "epoch": 1.6036655211912945, + "forget_loss": 2.7628921372979676e-09, + "retain_loss": 0.014592481777071953, + "step": 1400 + }, + { + "epoch": 1.6151202749140894, + "grad_norm": 0.1962890625, + "learning_rate": 9.706221198156683e-06, + "loss": 0.02171867936849594, + "step": 1410 + }, + { + "epoch": 1.6151202749140894, + "forget_loss": 4.1544343076793666e-08, + "retain_loss": 0.019604764878749847, + "step": 1410 + }, + { + "epoch": 1.6151202749140894, + "forget_loss": 3.411736315683811e-08, + "retain_loss": 0.019353380426764488, + "step": 1410 + }, + { + "epoch": 1.6151202749140894, + "forget_loss": 1.0411536521814924e-08, + "retain_loss": 0.017512496560811996, + "step": 1410 + }, + { + "epoch": 1.6151202749140894, + "forget_loss": 2.658219777629256e-08, + "retain_loss": 0.02085181139409542, + "step": 1410 + }, + { + "epoch": 1.6265750286368843, + "grad_norm": 0.2119140625, + "learning_rate": 9.418202764976959e-06, + "loss": 0.020016534626483916, + "step": 1420 + }, + { + "epoch": 1.6265750286368843, + "forget_loss": 5.425387428203976e-08, + "retain_loss": 0.031197555363178253, + "step": 1420 + }, + { + "epoch": 1.6265750286368843, + "forget_loss": 7.546157121396391e-08, + "retain_loss": 0.017391281202435493, + "step": 1420 + }, + { + "epoch": 1.6265750286368843, + "forget_loss": 1.0410505346669652e-08, + "retain_loss": 0.015930164605379105, + "step": 1420 + }, + { + "epoch": 1.6265750286368843, + "forget_loss": 5.6592699415602965e-09, + "retain_loss": 0.017635004594922066, + "step": 1420 + }, + { + "epoch": 1.6380297823596792, + "grad_norm": 0.1923828125, + "learning_rate": 9.130184331797236e-06, + "loss": 0.02109528034925461, + "step": 1430 + }, + { + "epoch": 1.6380297823596792, + "forget_loss": 6.951045072867146e-09, + "retain_loss": 0.014017445035278797, + "step": 1430 + }, + { + "epoch": 1.6380297823596792, + "forget_loss": 1.156479445540981e-08, + "retain_loss": 0.017461460083723068, + "step": 1430 + }, + { + "epoch": 1.6380297823596792, + "forget_loss": 6.22810780726013e-09, + "retain_loss": 0.019582783803343773, + "step": 1430 + }, + { + "epoch": 1.6380297823596792, + "forget_loss": 8.705757004179304e-09, + "retain_loss": 0.016309870406985283, + "step": 1430 + }, + { + "epoch": 1.6494845360824741, + "grad_norm": 0.173828125, + "learning_rate": 8.842165898617512e-06, + "loss": 0.020758749544620515, + "step": 1440 + }, + { + "epoch": 1.6494845360824741, + "forget_loss": 1.6525364188169078e-08, + "retain_loss": 0.015881458297371864, + "step": 1440 + }, + { + "epoch": 1.6494845360824741, + "forget_loss": 2.484807204439221e-08, + "retain_loss": 0.009261966682970524, + "step": 1440 + }, + { + "epoch": 1.6494845360824741, + "forget_loss": 3.562422890013295e-08, + "retain_loss": 0.022935394197702408, + "step": 1440 + }, + { + "epoch": 1.6494845360824741, + "forget_loss": 1.280076844523137e-07, + "retain_loss": 0.014195682480931282, + "step": 1440 + }, + { + "epoch": 1.660939289805269, + "grad_norm": 0.1943359375, + "learning_rate": 8.554147465437788e-06, + "loss": 0.02167462408542633, + "step": 1450 + }, + { + "epoch": 1.660939289805269, + "forget_loss": 6.6190928293963225e-09, + "retain_loss": 0.024128355085849762, + "step": 1450 + }, + { + "epoch": 1.660939289805269, + "forget_loss": 8.966759779127642e-09, + "retain_loss": 0.018851345404982567, + "step": 1450 + }, + { + "epoch": 1.660939289805269, + "forget_loss": 7.917346778185674e-08, + "retain_loss": 0.017255203798413277, + "step": 1450 + }, + { + "epoch": 1.660939289805269, + "forget_loss": 1.073428208542282e-07, + "retain_loss": 0.02466348372399807, + "step": 1450 + }, + { + "epoch": 1.6723940435280642, + "grad_norm": 0.2451171875, + "learning_rate": 8.266129032258065e-06, + "loss": 0.022821980714797973, + "step": 1460 + }, + { + "epoch": 1.6723940435280642, + "forget_loss": 4.6970569655968575e-09, + "retain_loss": 0.034727588295936584, + "step": 1460 + }, + { + "epoch": 1.6723940435280642, + "forget_loss": 9.013169233185181e-08, + "retain_loss": 0.016363129019737244, + "step": 1460 + }, + { + "epoch": 1.6723940435280642, + "forget_loss": 1.7886146963519423e-07, + "retain_loss": 0.027345962822437286, + "step": 1460 + }, + { + "epoch": 1.6723940435280642, + "forget_loss": 3.0995966682212384e-08, + "retain_loss": 0.03211313113570213, + "step": 1460 + }, + { + "epoch": 1.6838487972508591, + "grad_norm": 0.197265625, + "learning_rate": 7.978110599078341e-06, + "loss": 0.023417454957962037, + "step": 1470 + }, + { + "epoch": 1.6838487972508591, + "forget_loss": 1.5241577955293906e-07, + "retain_loss": 0.030100328847765923, + "step": 1470 + }, + { + "epoch": 1.6838487972508591, + "forget_loss": 1.787868164626616e-08, + "retain_loss": 0.020799852907657623, + "step": 1470 + }, + { + "epoch": 1.6838487972508591, + "forget_loss": 3.712963092539212e-08, + "retain_loss": 0.016399649903178215, + "step": 1470 + }, + { + "epoch": 1.6838487972508591, + "forget_loss": 2.06813961511898e-08, + "retain_loss": 0.016628945246338844, + "step": 1470 + }, + { + "epoch": 1.695303550973654, + "grad_norm": 0.2490234375, + "learning_rate": 7.690092165898617e-06, + "loss": 0.02219371348619461, + "step": 1480 + }, + { + "epoch": 1.695303550973654, + "forget_loss": 3.038276830125142e-08, + "retain_loss": 0.022364655509591103, + "step": 1480 + }, + { + "epoch": 1.695303550973654, + "forget_loss": 1.5132698649722442e-07, + "retain_loss": 0.02092198096215725, + "step": 1480 + }, + { + "epoch": 1.695303550973654, + "forget_loss": 5.258186064338588e-08, + "retain_loss": 0.009369916282594204, + "step": 1480 + }, + { + "epoch": 1.695303550973654, + "forget_loss": 3.0446482668367025e-08, + "retain_loss": 0.024398265406489372, + "step": 1480 + }, + { + "epoch": 1.7067583046964492, + "grad_norm": 0.1728515625, + "learning_rate": 7.402073732718895e-06, + "loss": 0.021649637818336488, + "step": 1490 + }, + { + "epoch": 1.7067583046964492, + "forget_loss": 2.699783063775385e-09, + "retain_loss": 0.015917476266622543, + "step": 1490 + }, + { + "epoch": 1.7067583046964492, + "forget_loss": 5.224163146522187e-08, + "retain_loss": 0.026943163946270943, + "step": 1490 + }, + { + "epoch": 1.7067583046964492, + "forget_loss": 5.959521764964393e-09, + "retain_loss": 0.020420579239726067, + "step": 1490 + }, + { + "epoch": 1.7067583046964492, + "forget_loss": 5.0872539958390917e-08, + "retain_loss": 0.02119573950767517, + "step": 1490 + }, + { + "epoch": 1.718213058419244, + "grad_norm": 0.2099609375, + "learning_rate": 7.114055299539171e-06, + "loss": 0.021461085975170137, + "step": 1500 + }, + { + "epoch": 1.718213058419244, + "forget_loss": 1.1111339972558198e-06, + "retain_loss": 0.015246644616127014, + "step": 1500 + }, + { + "epoch": 1.718213058419244, + "forget_loss": 1.2093730461515406e-08, + "retain_loss": 0.02578875608742237, + "step": 1500 + }, + { + "epoch": 1.718213058419244, + "forget_loss": 1.0410992956622067e-08, + "retain_loss": 0.029737161472439766, + "step": 1500 + }, + { + "epoch": 1.718213058419244, + "forget_loss": 1.3550183197708066e-08, + "retain_loss": 0.0380815826356411, + "step": 1500 + }, + { + "epoch": 1.729667812142039, + "grad_norm": 0.171875, + "learning_rate": 6.826036866359448e-06, + "loss": 0.022089402377605438, + "step": 1510 + }, + { + "epoch": 1.729667812142039, + "forget_loss": 1.4068325171479046e-08, + "retain_loss": 0.025930780917406082, + "step": 1510 + }, + { + "epoch": 1.729667812142039, + "forget_loss": 3.942339077411816e-08, + "retain_loss": 0.021967818960547447, + "step": 1510 + }, + { + "epoch": 1.729667812142039, + "forget_loss": 1.2184702313788875e-07, + "retain_loss": 0.04010249674320221, + "step": 1510 + }, + { + "epoch": 1.729667812142039, + "forget_loss": 2.8449733235902386e-08, + "retain_loss": 0.02482871524989605, + "step": 1510 + }, + { + "epoch": 1.741122565864834, + "grad_norm": 0.18359375, + "learning_rate": 6.538018433179724e-06, + "loss": 0.022809343039989473, + "step": 1520 + }, + { + "epoch": 1.741122565864834, + "forget_loss": 8.75739303296541e-09, + "retain_loss": 0.02512333169579506, + "step": 1520 + }, + { + "epoch": 1.741122565864834, + "forget_loss": 2.932968001800873e-08, + "retain_loss": 0.028969930484890938, + "step": 1520 + }, + { + "epoch": 1.741122565864834, + "forget_loss": 1.3895233408334207e-08, + "retain_loss": 0.022200997918844223, + "step": 1520 + }, + { + "epoch": 1.741122565864834, + "forget_loss": 1.0810396844362913e-07, + "retain_loss": 0.02343927137553692, + "step": 1520 + }, + { + "epoch": 1.7525773195876289, + "grad_norm": 0.1650390625, + "learning_rate": 6.25e-06, + "loss": 0.021387623250484468, + "step": 1530 + }, + { + "epoch": 1.7525773195876289, + "forget_loss": 3.949251095036743e-07, + "retain_loss": 0.03315495699644089, + "step": 1530 + }, + { + "epoch": 1.7525773195876289, + "forget_loss": 1.1532841170946995e-07, + "retain_loss": 0.023524099960923195, + "step": 1530 + }, + { + "epoch": 1.7525773195876289, + "forget_loss": 1.737878463359266e-08, + "retain_loss": 0.021044719964265823, + "step": 1530 + }, + { + "epoch": 1.7525773195876289, + "forget_loss": 1.5254467689018725e-09, + "retain_loss": 0.016680333763360977, + "step": 1530 + }, + { + "epoch": 1.7640320733104238, + "grad_norm": 0.17578125, + "learning_rate": 5.961981566820277e-06, + "loss": 0.023655688762664794, + "step": 1540 + }, + { + "epoch": 1.7640320733104238, + "forget_loss": 5.484626974805451e-09, + "retain_loss": 0.015781564638018608, + "step": 1540 + }, + { + "epoch": 1.7640320733104238, + "forget_loss": 8.721925404131525e-09, + "retain_loss": 0.02230190671980381, + "step": 1540 + }, + { + "epoch": 1.7640320733104238, + "forget_loss": 6.269440433470663e-08, + "retain_loss": 0.043873514980077744, + "step": 1540 + }, + { + "epoch": 1.7640320733104238, + "forget_loss": 2.495698225857268e-08, + "retain_loss": 0.04776827245950699, + "step": 1540 + }, + { + "epoch": 1.7754868270332187, + "grad_norm": 0.244140625, + "learning_rate": 5.673963133640553e-06, + "loss": 0.021500414609909056, + "step": 1550 + }, + { + "epoch": 1.7754868270332187, + "forget_loss": 1.2894139089780765e-08, + "retain_loss": 0.021685905754566193, + "step": 1550 + }, + { + "epoch": 1.7754868270332187, + "forget_loss": 1.01795150087014e-08, + "retain_loss": 0.02003580518066883, + "step": 1550 + }, + { + "epoch": 1.7754868270332187, + "forget_loss": 4.5272781079575e-09, + "retain_loss": 0.02798016183078289, + "step": 1550 + }, + { + "epoch": 1.7754868270332187, + "forget_loss": 9.158154767874294e-08, + "retain_loss": 0.020200006663799286, + "step": 1550 + }, + { + "epoch": 1.7869415807560136, + "grad_norm": 0.2099609375, + "learning_rate": 5.38594470046083e-06, + "loss": 0.021815258264541625, + "step": 1560 + }, + { + "epoch": 1.7869415807560136, + "forget_loss": 1.9465607792312767e-08, + "retain_loss": 0.023672498762607574, + "step": 1560 + }, + { + "epoch": 1.7869415807560136, + "forget_loss": 2.209292802035634e-07, + "retain_loss": 0.019269166514277458, + "step": 1560 + }, + { + "epoch": 1.7869415807560136, + "forget_loss": 5.099231259464432e-09, + "retain_loss": 0.026050956919789314, + "step": 1560 + }, + { + "epoch": 1.7869415807560136, + "forget_loss": 6.805104924012539e-09, + "retain_loss": 0.02401677332818508, + "step": 1560 + }, + { + "epoch": 1.7983963344788088, + "grad_norm": 0.162109375, + "learning_rate": 5.097926267281106e-06, + "loss": 0.021849420666694642, + "step": 1570 + }, + { + "epoch": 1.7983963344788088, + "forget_loss": 1.097847146525055e-08, + "retain_loss": 0.017921848222613335, + "step": 1570 + }, + { + "epoch": 1.7983963344788088, + "forget_loss": 1.049954523324459e-08, + "retain_loss": 0.016865499317646027, + "step": 1570 + }, + { + "epoch": 1.7983963344788088, + "forget_loss": 8.912421023410388e-09, + "retain_loss": 0.01202414557337761, + "step": 1570 + }, + { + "epoch": 1.7983963344788088, + "forget_loss": 4.532140707169674e-08, + "retain_loss": 0.024097174406051636, + "step": 1570 + }, + { + "epoch": 1.8098510882016037, + "grad_norm": 0.2109375, + "learning_rate": 4.809907834101383e-06, + "loss": 0.021765342354774474, + "step": 1580 + }, + { + "epoch": 1.8098510882016037, + "forget_loss": 1.4280697513413543e-08, + "retain_loss": 0.027266938239336014, + "step": 1580 + }, + { + "epoch": 1.8098510882016037, + "forget_loss": 3.764376899084709e-08, + "retain_loss": 0.02171538770198822, + "step": 1580 + }, + { + "epoch": 1.8098510882016037, + "forget_loss": 5.152051674173208e-09, + "retain_loss": 0.01970984973013401, + "step": 1580 + }, + { + "epoch": 1.8098510882016037, + "forget_loss": 1.993336837813331e-07, + "retain_loss": 0.03570767864584923, + "step": 1580 + }, + { + "epoch": 1.8213058419243986, + "grad_norm": 0.16796875, + "learning_rate": 4.5218894009216595e-06, + "loss": 0.022261792421340944, + "step": 1590 + }, + { + "epoch": 1.8213058419243986, + "forget_loss": 6.277179664948562e-08, + "retain_loss": 0.026909146457910538, + "step": 1590 + }, + { + "epoch": 1.8213058419243986, + "forget_loss": 6.691417975268621e-10, + "retain_loss": 0.015054693445563316, + "step": 1590 + }, + { + "epoch": 1.8213058419243986, + "forget_loss": 2.381351116298447e-09, + "retain_loss": 0.023661844432353973, + "step": 1590 + }, + { + "epoch": 1.8213058419243986, + "forget_loss": 1.1333929350598737e-08, + "retain_loss": 0.018988439813256264, + "step": 1590 + }, + { + "epoch": 1.8327605956471937, + "grad_norm": 0.1923828125, + "learning_rate": 4.233870967741935e-06, + "loss": 0.02133145332336426, + "step": 1600 + } + ], + "logging_steps": 10, + "max_steps": 1746, + "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": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +}