{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "eval_steps": 500, "global_step": 506, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0, "forget_loss": 5.228305339813232, "retain_loss": 2.6653778553009033, "step": 0 }, { "epoch": 0, "forget_loss": 5.790935516357422, "retain_loss": 2.794126033782959, "step": 0 }, { "epoch": 0, "forget_loss": 5.287184238433838, "retain_loss": 2.7992446422576904, "step": 0 }, { "epoch": 0, "forget_loss": 5.4225077629089355, "retain_loss": 2.7407283782958984, "step": 0 }, { "epoch": 0.003956478733926805, "grad_norm": 14.0625, "learning_rate": 0.0, "loss": 8.18210220336914, "step": 1 }, { "epoch": 0.03956478733926805, "grad_norm": 18.75, "learning_rate": 4.5e-06, "loss": 7.783626980251736, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 5.050795078277588, "retain_loss": 2.3033814430236816, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 4.908670902252197, "retain_loss": 2.5407941341400146, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 4.856396198272705, "retain_loss": 2.474029541015625, "step": 10 }, { "epoch": 0.03956478733926805, "forget_loss": 4.859658718109131, "retain_loss": 2.726255178451538, "step": 10 }, { "epoch": 0.0791295746785361, "grad_norm": 9.5625, "learning_rate": 9.5e-06, "loss": 7.457444763183593, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 5.263526916503906, "retain_loss": 2.4462852478027344, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 5.065126419067383, "retain_loss": 2.6068644523620605, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 5.208187103271484, "retain_loss": 2.633065700531006, "step": 20 }, { "epoch": 0.0791295746785361, "forget_loss": 4.474795818328857, "retain_loss": 2.2739768028259277, "step": 20 }, { "epoch": 0.11869436201780416, "grad_norm": 5.15625, "learning_rate": 9.814814814814815e-06, "loss": 6.889686584472656, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 4.2715911865234375, "retain_loss": 2.7482452392578125, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 4.270931720733643, "retain_loss": 2.8346025943756104, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 4.168623924255371, "retain_loss": 2.368190050125122, "step": 30 }, { "epoch": 0.11869436201780416, "forget_loss": 4.262532711029053, "retain_loss": 2.6943798065185547, "step": 30 }, { "epoch": 0.1582591493570722, "grad_norm": 3.984375, "learning_rate": 9.609053497942388e-06, "loss": 6.669891357421875, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 3.949828624725342, "retain_loss": 2.471006393432617, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 3.9658637046813965, "retain_loss": 2.365844488143921, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 4.037973403930664, "retain_loss": 2.595186471939087, "step": 40 }, { "epoch": 0.1582591493570722, "forget_loss": 3.8019163608551025, "retain_loss": 2.402667284011841, "step": 40 }, { "epoch": 0.19782393669634027, "grad_norm": 12.0, "learning_rate": 9.403292181069959e-06, "loss": 6.561286163330078, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 3.9541335105895996, "retain_loss": 2.4763028621673584, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 4.136512279510498, "retain_loss": 2.211292266845703, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 4.035154342651367, "retain_loss": 2.7226850986480713, "step": 50 }, { "epoch": 0.19782393669634027, "forget_loss": 4.196267604827881, "retain_loss": 2.3140127658843994, "step": 50 }, { "epoch": 0.23738872403560832, "grad_norm": 4.3125, "learning_rate": 9.197530864197531e-06, "loss": 6.488241577148438, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 4.150100231170654, "retain_loss": 2.4890434741973877, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 3.830029249191284, "retain_loss": 2.056034803390503, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 4.062994956970215, "retain_loss": 2.6220014095306396, "step": 60 }, { "epoch": 0.23738872403560832, "forget_loss": 4.05166482925415, "retain_loss": 2.327568292617798, "step": 60 }, { "epoch": 0.27695351137487634, "grad_norm": 3.28125, "learning_rate": 8.991769547325104e-06, "loss": 6.390378952026367, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 4.1699724197387695, "retain_loss": 2.4814493656158447, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 4.043442249298096, "retain_loss": 2.3001720905303955, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 3.8387367725372314, "retain_loss": 2.364630699157715, "step": 70 }, { "epoch": 0.27695351137487634, "forget_loss": 4.229508876800537, "retain_loss": 2.3277294635772705, "step": 70 }, { "epoch": 0.3165182987141444, "grad_norm": 5.5, "learning_rate": 8.786008230452676e-06, "loss": 6.3377540588378904, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 4.228198528289795, "retain_loss": 2.5336318016052246, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 3.9950098991394043, "retain_loss": 2.2209460735321045, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 3.841887950897217, "retain_loss": 2.3174612522125244, "step": 80 }, { "epoch": 0.3165182987141444, "forget_loss": 3.8216302394866943, "retain_loss": 2.194601535797119, "step": 80 }, { "epoch": 0.3560830860534125, "grad_norm": 4.59375, "learning_rate": 8.580246913580249e-06, "loss": 6.388538360595703, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 3.8051679134368896, "retain_loss": 2.209702730178833, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 4.07779598236084, "retain_loss": 2.443394184112549, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 3.9847471714019775, "retain_loss": 2.5544075965881348, "step": 90 }, { "epoch": 0.3560830860534125, "forget_loss": 4.070178031921387, "retain_loss": 2.382122278213501, "step": 90 }, { "epoch": 0.39564787339268054, "grad_norm": 3.46875, "learning_rate": 8.37448559670782e-06, "loss": 6.273267364501953, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 4.126764297485352, "retain_loss": 2.0702314376831055, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 3.9314870834350586, "retain_loss": 2.451327323913574, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 3.869971752166748, "retain_loss": 2.386556625366211, "step": 100 }, { "epoch": 0.39564787339268054, "forget_loss": 4.190179347991943, "retain_loss": 2.182694435119629, "step": 100 }, { "epoch": 0.4352126607319486, "grad_norm": 3.0, "learning_rate": 8.168724279835392e-06, "loss": 6.3174995422363285, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 3.954601764678955, "retain_loss": 2.387284517288208, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 4.2787604331970215, "retain_loss": 2.4614319801330566, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 3.9109017848968506, "retain_loss": 2.486406087875366, "step": 110 }, { "epoch": 0.4352126607319486, "forget_loss": 4.027335166931152, "retain_loss": 2.030792713165283, "step": 110 }, { "epoch": 0.47477744807121663, "grad_norm": 2.703125, "learning_rate": 7.962962962962963e-06, "loss": 6.260037994384765, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 3.965056896209717, "retain_loss": 2.1373112201690674, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 3.8014025688171387, "retain_loss": 1.9166232347488403, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 4.077919960021973, "retain_loss": 2.3020386695861816, "step": 120 }, { "epoch": 0.47477744807121663, "forget_loss": 4.005480766296387, "retain_loss": 2.2376527786254883, "step": 120 }, { "epoch": 0.5143422354104846, "grad_norm": 2.9375, "learning_rate": 7.757201646090535e-06, "loss": 6.193301010131836, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 3.972163438796997, "retain_loss": 2.4266738891601562, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 4.186028957366943, "retain_loss": 2.346269369125366, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 3.9561634063720703, "retain_loss": 2.5126609802246094, "step": 130 }, { "epoch": 0.5143422354104846, "forget_loss": 3.682123899459839, "retain_loss": 2.355285406112671, "step": 130 }, { "epoch": 0.5539070227497527, "grad_norm": 2.859375, "learning_rate": 7.551440329218107e-06, "loss": 6.207519149780273, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 4.324990272521973, "retain_loss": 2.2892813682556152, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 4.004846572875977, "retain_loss": 2.1679553985595703, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 3.661930561065674, "retain_loss": 1.9277281761169434, "step": 140 }, { "epoch": 0.5539070227497527, "forget_loss": 3.9737062454223633, "retain_loss": 1.9291268587112427, "step": 140 }, { "epoch": 0.5934718100890207, "grad_norm": 2.671875, "learning_rate": 7.34567901234568e-06, "loss": 6.157084655761719, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 3.894688129425049, "retain_loss": 2.094151020050049, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 3.8356785774230957, "retain_loss": 2.1127727031707764, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 3.7613933086395264, "retain_loss": 2.426790714263916, "step": 150 }, { "epoch": 0.5934718100890207, "forget_loss": 4.2197184562683105, "retain_loss": 1.8644764423370361, "step": 150 }, { "epoch": 0.6330365974282888, "grad_norm": 3.171875, "learning_rate": 7.139917695473252e-06, "loss": 6.1496223449707035, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 3.841630220413208, "retain_loss": 2.435896396636963, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 4.037227153778076, "retain_loss": 2.2480666637420654, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 3.795936107635498, "retain_loss": 2.489936351776123, "step": 160 }, { "epoch": 0.6330365974282888, "forget_loss": 4.139549732208252, "retain_loss": 2.3118233680725098, "step": 160 }, { "epoch": 0.6726013847675568, "grad_norm": 2.4375, "learning_rate": 6.934156378600824e-06, "loss": 6.15985336303711, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 3.844918966293335, "retain_loss": 2.0793261528015137, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 3.8478827476501465, "retain_loss": 1.8507996797561646, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 3.738731861114502, "retain_loss": 2.2272250652313232, "step": 170 }, { "epoch": 0.6726013847675568, "forget_loss": 4.023928165435791, "retain_loss": 2.482393264770508, "step": 170 }, { "epoch": 0.712166172106825, "grad_norm": 2.875, "learning_rate": 6.728395061728395e-06, "loss": 6.102357482910156, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 3.858079671859741, "retain_loss": 2.3981428146362305, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 4.243940353393555, "retain_loss": 2.076014995574951, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 3.9097959995269775, "retain_loss": 2.0742626190185547, "step": 180 }, { "epoch": 0.712166172106825, "forget_loss": 3.9718551635742188, "retain_loss": 2.1771230697631836, "step": 180 }, { "epoch": 0.751730959446093, "grad_norm": 2.703125, "learning_rate": 6.522633744855967e-06, "loss": 6.077034759521484, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 4.07846212387085, "retain_loss": 2.193300485610962, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 3.97346830368042, "retain_loss": 2.2364301681518555, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 3.860450267791748, "retain_loss": 2.345485210418701, "step": 190 }, { "epoch": 0.751730959446093, "forget_loss": 4.061342716217041, "retain_loss": 2.19950795173645, "step": 190 }, { "epoch": 0.7912957467853611, "grad_norm": 2.734375, "learning_rate": 6.3168724279835396e-06, "loss": 6.212088012695313, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 3.8057045936584473, "retain_loss": 2.155118465423584, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 3.7121968269348145, "retain_loss": 2.2442877292633057, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 3.89788556098938, "retain_loss": 2.016833782196045, "step": 200 }, { "epoch": 0.7912957467853611, "forget_loss": 4.040591716766357, "retain_loss": 2.054882764816284, "step": 200 }, { "epoch": 0.8308605341246291, "grad_norm": 5.21875, "learning_rate": 6.111111111111112e-06, "loss": 6.104199981689453, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 4.09905481338501, "retain_loss": 2.385629653930664, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 4.130467891693115, "retain_loss": 2.121933698654175, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 3.5963847637176514, "retain_loss": 2.2246758937835693, "step": 210 }, { "epoch": 0.8308605341246291, "forget_loss": 3.781541585922241, "retain_loss": 2.1959068775177, "step": 210 }, { "epoch": 0.8704253214638972, "grad_norm": 3.109375, "learning_rate": 5.905349794238684e-06, "loss": 6.097019195556641, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 3.804974317550659, "retain_loss": 2.4216184616088867, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 4.051773548126221, "retain_loss": 2.2472169399261475, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 3.88171648979187, "retain_loss": 1.885982871055603, "step": 220 }, { "epoch": 0.8704253214638972, "forget_loss": 3.781057834625244, "retain_loss": 2.2794511318206787, "step": 220 }, { "epoch": 0.9099901088031652, "grad_norm": 2.671875, "learning_rate": 5.699588477366255e-06, "loss": 6.0480293273925785, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 3.9442813396453857, "retain_loss": 2.0114781856536865, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 3.9447662830352783, "retain_loss": 2.1324877738952637, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 3.6983299255371094, "retain_loss": 2.309474468231201, "step": 230 }, { "epoch": 0.9099901088031652, "forget_loss": 3.794271945953369, "retain_loss": 2.3472728729248047, "step": 230 }, { "epoch": 0.9495548961424333, "grad_norm": 2.78125, "learning_rate": 5.493827160493828e-06, "loss": 6.008623886108398, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 4.023442268371582, "retain_loss": 2.0604634284973145, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 3.8832573890686035, "retain_loss": 2.165562391281128, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 3.8648319244384766, "retain_loss": 2.203535556793213, "step": 240 }, { "epoch": 0.9495548961424333, "forget_loss": 4.058237075805664, "retain_loss": 2.2459559440612793, "step": 240 }, { "epoch": 0.9891196834817013, "grad_norm": 5.125, "learning_rate": 5.2880658436213996e-06, "loss": 6.081012725830078, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 3.9296000003814697, "retain_loss": 2.2531886100769043, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 3.9538869857788086, "retain_loss": 2.181769609451294, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 3.955592632293701, "retain_loss": 2.205435276031494, "step": 250 }, { "epoch": 0.9891196834817013, "forget_loss": 3.8982174396514893, "retain_loss": 2.0968081951141357, "step": 250 }, { "epoch": 1.0276953511374876, "grad_norm": 2.59375, "learning_rate": 5.082304526748971e-06, "loss": 6.081734848022461, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 4.061560153961182, "retain_loss": 1.9784836769104004, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 3.915540933609009, "retain_loss": 2.2107746601104736, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 3.8902337551116943, "retain_loss": 2.0626654624938965, "step": 260 }, { "epoch": 1.0276953511374876, "forget_loss": 3.9867382049560547, "retain_loss": 2.057222604751587, "step": 260 }, { "epoch": 1.0672601384767557, "grad_norm": 2.5, "learning_rate": 4.876543209876544e-06, "loss": 5.919136428833008, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 3.7137222290039062, "retain_loss": 2.20261549949646, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 3.759235143661499, "retain_loss": 2.0318472385406494, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 4.022280693054199, "retain_loss": 1.8729618787765503, "step": 270 }, { "epoch": 1.0672601384767557, "forget_loss": 3.7953128814697266, "retain_loss": 2.040851593017578, "step": 270 }, { "epoch": 1.1068249258160237, "grad_norm": 2.484375, "learning_rate": 4.670781893004115e-06, "loss": 6.021038055419922, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 4.016865253448486, "retain_loss": 1.941941738128662, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 3.7810330390930176, "retain_loss": 1.8878296613693237, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 3.448192596435547, "retain_loss": 2.1850433349609375, "step": 280 }, { "epoch": 1.1068249258160237, "forget_loss": 3.936615467071533, "retain_loss": 2.0968942642211914, "step": 280 }, { "epoch": 1.1463897131552918, "grad_norm": 2.5625, "learning_rate": 4.465020576131688e-06, "loss": 5.993334579467773, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 3.884347677230835, "retain_loss": 2.1678333282470703, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 3.957656145095825, "retain_loss": 2.159306526184082, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 3.621809482574463, "retain_loss": 2.037255048751831, "step": 290 }, { "epoch": 1.1463897131552918, "forget_loss": 3.965614080429077, "retain_loss": 1.9439359903335571, "step": 290 }, { "epoch": 1.1859545004945597, "grad_norm": 2.484375, "learning_rate": 4.2592592592592596e-06, "loss": 5.992023468017578, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 4.040769577026367, "retain_loss": 2.0465481281280518, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 3.765821933746338, "retain_loss": 2.3230631351470947, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 3.838066577911377, "retain_loss": 2.1758203506469727, "step": 300 }, { "epoch": 1.1859545004945597, "forget_loss": 3.9743764400482178, "retain_loss": 2.1406238079071045, "step": 300 }, { "epoch": 1.225519287833828, "grad_norm": 2.59375, "learning_rate": 4.053497942386831e-06, "loss": 6.039208221435547, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 3.822885274887085, "retain_loss": 2.187807559967041, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 3.873359203338623, "retain_loss": 2.004912853240967, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 4.0021891593933105, "retain_loss": 2.1716084480285645, "step": 310 }, { "epoch": 1.225519287833828, "forget_loss": 4.003294467926025, "retain_loss": 2.1612296104431152, "step": 310 }, { "epoch": 1.265084075173096, "grad_norm": 3.765625, "learning_rate": 3.847736625514404e-06, "loss": 6.015998077392578, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 3.9162092208862305, "retain_loss": 1.9279217720031738, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 3.7619311809539795, "retain_loss": 2.061594009399414, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 3.956817865371704, "retain_loss": 2.146383047103882, "step": 320 }, { "epoch": 1.265084075173096, "forget_loss": 3.6850597858428955, "retain_loss": 2.1570324897766113, "step": 320 }, { "epoch": 1.304648862512364, "grad_norm": 2.59375, "learning_rate": 3.641975308641976e-06, "loss": 6.002947616577148, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 3.892425060272217, "retain_loss": 1.9530080556869507, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 3.9805660247802734, "retain_loss": 2.2374324798583984, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 3.7663347721099854, "retain_loss": 1.7748332023620605, "step": 330 }, { "epoch": 1.304648862512364, "forget_loss": 3.9563276767730713, "retain_loss": 2.096442222595215, "step": 330 }, { "epoch": 1.344213649851632, "grad_norm": 2.578125, "learning_rate": 3.4362139917695475e-06, "loss": 5.931233978271484, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 3.723862648010254, "retain_loss": 2.1462862491607666, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 3.6262669563293457, "retain_loss": 2.1007308959960938, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 3.78971791267395, "retain_loss": 2.274022102355957, "step": 340 }, { "epoch": 1.344213649851632, "forget_loss": 4.297574043273926, "retain_loss": 2.0463507175445557, "step": 340 }, { "epoch": 1.3837784371909, "grad_norm": 2.703125, "learning_rate": 3.230452674897119e-06, "loss": 6.039730453491211, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 4.0523881912231445, "retain_loss": 1.7336885929107666, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 4.002918720245361, "retain_loss": 1.8986440896987915, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 3.752483606338501, "retain_loss": 2.0627171993255615, "step": 350 }, { "epoch": 1.3837784371909, "forget_loss": 3.862380027770996, "retain_loss": 1.6684348583221436, "step": 350 }, { "epoch": 1.4233432245301683, "grad_norm": 2.484375, "learning_rate": 3.0246913580246917e-06, "loss": 5.98388442993164, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 3.8853209018707275, "retain_loss": 2.011307954788208, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 3.9185667037963867, "retain_loss": 2.368375778198242, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 3.948054075241089, "retain_loss": 2.1703217029571533, "step": 360 }, { "epoch": 1.4233432245301683, "forget_loss": 3.9023869037628174, "retain_loss": 2.03293514251709, "step": 360 }, { "epoch": 1.4629080118694362, "grad_norm": 2.671875, "learning_rate": 2.8189300411522637e-06, "loss": 5.96927490234375, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 3.877941370010376, "retain_loss": 1.9830896854400635, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 3.75337815284729, "retain_loss": 2.0410003662109375, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 3.6140739917755127, "retain_loss": 2.0619940757751465, "step": 370 }, { "epoch": 1.4629080118694362, "forget_loss": 4.116907119750977, "retain_loss": 2.0889041423797607, "step": 370 }, { "epoch": 1.5024727992087041, "grad_norm": 2.421875, "learning_rate": 2.6131687242798354e-06, "loss": 5.977725601196289, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 3.852693796157837, "retain_loss": 1.8302228450775146, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 3.919356346130371, "retain_loss": 1.9756414890289307, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 4.005396842956543, "retain_loss": 1.9624111652374268, "step": 380 }, { "epoch": 1.5024727992087041, "forget_loss": 3.8799474239349365, "retain_loss": 1.6203029155731201, "step": 380 }, { "epoch": 1.5420375865479723, "grad_norm": 3.359375, "learning_rate": 2.4074074074074075e-06, "loss": 5.985309219360351, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 4.004049777984619, "retain_loss": 2.2051491737365723, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 4.058261394500732, "retain_loss": 1.9921034574508667, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 3.971198558807373, "retain_loss": 1.9959430694580078, "step": 390 }, { "epoch": 1.5420375865479723, "forget_loss": 3.690929651260376, "retain_loss": 2.367933511734009, "step": 390 }, { "epoch": 1.5816023738872405, "grad_norm": 2.546875, "learning_rate": 2.2016460905349796e-06, "loss": 5.961531066894532, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 3.749250888824463, "retain_loss": 2.06518816947937, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 3.8270249366760254, "retain_loss": 2.048008918762207, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 4.016013145446777, "retain_loss": 1.981733798980713, "step": 400 }, { "epoch": 1.5816023738872405, "forget_loss": 4.077958583831787, "retain_loss": 2.1460680961608887, "step": 400 }, { "epoch": 1.6211671612265084, "grad_norm": 2.484375, "learning_rate": 1.9958847736625517e-06, "loss": 6.026706314086914, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 3.643620729446411, "retain_loss": 1.785910725593567, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 3.890308380126953, "retain_loss": 2.245136022567749, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 3.8318428993225098, "retain_loss": 2.3317747116088867, "step": 410 }, { "epoch": 1.6211671612265084, "forget_loss": 3.908914566040039, "retain_loss": 2.134136915206909, "step": 410 }, { "epoch": 1.6607319485657763, "grad_norm": 2.671875, "learning_rate": 1.7901234567901235e-06, "loss": 5.965105056762695, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 3.7508301734924316, "retain_loss": 2.0185036659240723, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 3.964979410171509, "retain_loss": 2.1062581539154053, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 4.007857322692871, "retain_loss": 2.1970365047454834, "step": 420 }, { "epoch": 1.6607319485657763, "forget_loss": 3.653095006942749, "retain_loss": 1.9304171800613403, "step": 420 }, { "epoch": 1.7002967359050445, "grad_norm": 2.6875, "learning_rate": 1.5843621399176956e-06, "loss": 5.977134704589844, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 3.888289213180542, "retain_loss": 1.867199420928955, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 3.780067205429077, "retain_loss": 2.1667048931121826, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 3.902003049850464, "retain_loss": 1.8836796283721924, "step": 430 }, { "epoch": 1.7002967359050445, "forget_loss": 3.8203608989715576, "retain_loss": 2.3105437755584717, "step": 430 }, { "epoch": 1.7398615232443126, "grad_norm": 2.578125, "learning_rate": 1.3786008230452677e-06, "loss": 5.945446014404297, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 3.8020174503326416, "retain_loss": 2.0557711124420166, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 3.9237802028656006, "retain_loss": 1.8451926708221436, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 4.012683868408203, "retain_loss": 2.0730113983154297, "step": 440 }, { "epoch": 1.7398615232443126, "forget_loss": 3.8095295429229736, "retain_loss": 1.9834192991256714, "step": 440 }, { "epoch": 1.7794263105835806, "grad_norm": 4.21875, "learning_rate": 1.1728395061728396e-06, "loss": 6.011910629272461, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 4.016027450561523, "retain_loss": 1.9786537885665894, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 3.97581148147583, "retain_loss": 1.927825689315796, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 3.886143445968628, "retain_loss": 2.162662982940674, "step": 450 }, { "epoch": 1.7794263105835806, "forget_loss": 3.734543800354004, "retain_loss": 1.834120750427246, "step": 450 }, { "epoch": 1.8189910979228485, "grad_norm": 2.46875, "learning_rate": 9.670781893004117e-07, "loss": 5.98345718383789, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 4.007359027862549, "retain_loss": 1.918708086013794, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 3.963118314743042, "retain_loss": 2.1799988746643066, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 3.9211266040802, "retain_loss": 2.012998580932617, "step": 460 }, { "epoch": 1.8189910979228485, "forget_loss": 3.917017698287964, "retain_loss": 2.2511258125305176, "step": 460 }, { "epoch": 1.8585558852621167, "grad_norm": 2.546875, "learning_rate": 7.613168724279836e-07, "loss": 5.976721572875976, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 3.9052722454071045, "retain_loss": 2.020348310470581, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 3.792335033416748, "retain_loss": 2.179080009460449, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 3.890678644180298, "retain_loss": 2.034893274307251, "step": 470 }, { "epoch": 1.8585558852621167, "forget_loss": 3.8522043228149414, "retain_loss": 2.160851240158081, "step": 470 }, { "epoch": 1.8981206726013848, "grad_norm": 3.828125, "learning_rate": 5.555555555555555e-07, "loss": 5.965145874023437, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 3.8236379623413086, "retain_loss": 2.003793478012085, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 3.934398651123047, "retain_loss": 2.2737514972686768, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 3.869436502456665, "retain_loss": 2.293827772140503, "step": 480 }, { "epoch": 1.8981206726013848, "forget_loss": 4.015378475189209, "retain_loss": 2.0771780014038086, "step": 480 }, { "epoch": 1.9376854599406528, "grad_norm": 2.765625, "learning_rate": 3.497942386831276e-07, "loss": 6.032207870483399, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 3.839055299758911, "retain_loss": 2.1054677963256836, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 3.784615993499756, "retain_loss": 1.9073494672775269, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 3.7499022483825684, "retain_loss": 2.132340908050537, "step": 490 }, { "epoch": 1.9376854599406528, "forget_loss": 3.512561559677124, "retain_loss": 2.2340593338012695, "step": 490 }, { "epoch": 1.9772502472799207, "grad_norm": 2.921875, "learning_rate": 1.440329218106996e-07, "loss": 5.986212158203125, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 3.87085223197937, "retain_loss": 1.7867763042449951, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 3.509723424911499, "retain_loss": 2.2367658615112305, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 4.056915283203125, "retain_loss": 1.8742592334747314, "step": 500 }, { "epoch": 1.9772502472799207, "forget_loss": 3.8325486183166504, "retain_loss": 1.9439557790756226, "step": 500 } ], "logging_steps": 10, "max_steps": 506, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }