{ "config": "C2", "description": "RoPE instead of sinusoidal absolute", "parameters": 2763264, "windowed": true, "window": { "src_window": 1024, "src_stride": 768, "train_examples": 25654 }, "best_epoch": 100, "best_val_loss": 0.09927769922586938, "test": { "loss": 0.11979607583332225, "ppl": 1.1272669511654074, "token_acc": 0.9746887539870357, "seq_acc": 0.56318347509113 }, "test_full": { "loss": 5.920605090370729, "ppl": 372.63712480831, "token_acc": 0.30378654455794296, "seq_acc": 0.052752293577981654 }, "gen_test_win": { "n": 3292.0, "seq_acc": 0.6597812879708383, "bit_acc": 0.9560890886286575, "lev": 1.537667071688943, "lev_norm": 0.010323837570599004, "token_acc": 0.8211193850588517, "bleu": 0.9453322550492058, "rouge1": 0.9753460409870339, "rouge2": 0.9555024454872828, "rougeL": 0.9753460409870339 }, "gen_test": { "n": 500.0, "seq_acc": 0.204, "bit_acc": 0.8181579795401267, "lev": 22.95, "lev_norm": 0.034200148953254766, "token_acc": 0.44049529344873334, "bleu": 0.9245620567121133, "rouge1": 0.9619928822649874, "rouge2": 0.9411761597516598, "rougeL": 0.9619324395189165 }, "gen_test_full": { "n": 500.0, "seq_acc": 0.054, "bit_acc": 0.3517906454336038, "lev": 505.6, "lev_norm": 0.6312979292199293, "token_acc": 0.04248673746235677, "bleu": 0.1026796740588983, "rouge1": 0.36243435714957306, "rouge2": 0.25601572873722805, "rougeL": 0.3112700877732138 }, "checkpoint": "/home2/winterdew/ANLP_A1/ANLP-M26-Assignment-1/outputs/checkpoints/C2_best.pt", "history": [ { "train/epoch_loss": 6.995370723362873, "epoch": 1, "val/loss": 5.975532108766493, "val/ppl": 393.67752463013096, "val/token_acc": 0.12048348974279702, "val/seq_acc": 0.0, "val_full/loss": 6.270644691050796, "val_full/ppl": 528.818192361973, "val_full/token_acc": 0.07332962884899116, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 5.833912618736971, "epoch": 2, "val/loss": 5.281581922754959, "val/ppl": 196.68076316142339, "val/token_acc": 0.18133857246861007, "val/seq_acc": 0.0, "val_full/loss": 5.8494414531797885, "val_full/ppl": 347.04048797371195, "val_full/token_acc": 0.10122822817068731, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 5.267378161375659, "epoch": 3, "val/loss": 4.599296502182211, "val/ppl": 99.41435325505417, "val/token_acc": 0.26174068274974815, "val/seq_acc": 0.0, "val_full/loss": 5.996897170021461, "val_full/ppl": 402.17896254727714, "val_full/token_acc": 0.0968594914637184, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 4.551063683859428, "epoch": 4, "val/loss": 3.5084439771734224, "val/ppl": 33.39626198735998, "val/token_acc": 0.38315165900409887, "val/seq_acc": 0.0, "val_full/loss": 6.7181524872339, "val_full/ppl": 827.2876741803625, "val_full/token_acc": 0.1017839008220123, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 3.580359074242989, "epoch": 5, "val/loss": 2.288402971404525, "val/ppl": 9.859179709632619, "val/token_acc": 0.5378125280489873, "val/seq_acc": 0.00032123353678124, "val_full/loss": 7.171605172426181, "val_full/ppl": 1301.9327529909626, "val_full/token_acc": 0.1069382436911993, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.0, "gen_val_win/bit_acc": 0.6191782853752595, "gen_val_win/lev": 62.53125, "gen_val_win/lev_norm": 0.45235598570141544, "gen_val_win/token_acc": 0.07652274479568234, "gen_val_win/bleu": 0.15222595619365087, "gen_val_win/rouge1": 0.42920903000545196, "gen_val_win/rouge2": 0.21950845824242435, "gen_val_win/rougeL": 0.4219266487165362, "gen_val/n": 128.0, "gen_val/seq_acc": 0.0, "gen_val/bit_acc": 0.5144970288021381, "gen_val/lev": 425.484375, "gen_val/lev_norm": 0.5156729614081624, "gen_val/token_acc": 0.015991277485008178, "gen_val/bleu": 0.12413074730544832, "gen_val/rouge1": 0.40704193764823327, "gen_val/rouge2": 0.19375547017279462, "gen_val/rougeL": 0.37537546462183163, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0, "gen_val_full/bit_acc": 0.1293192520602159, "gen_val_full/lev": 533.875, "gen_val_full/lev_norm": 0.8060229206813219, "gen_val_full/token_acc": 0.011334698157487392, "gen_val_full/bleu": 0.0005790934932958215, "gen_val_full/rouge1": 0.1511024572918868, "gen_val_full/rouge2": 0.047324442374536424, "gen_val_full/rougeL": 0.13051932962132223 }, { "train/epoch_loss": 2.7552905706990685, "epoch": 6, "val/loss": 1.5545950089515226, "val/ppl": 4.733169245769669, "val/token_acc": 0.6527709906154322, "val/seq_acc": 0.00128493414712496, "val_full/loss": 7.066416803588879, "val_full/ppl": 1171.941206280403, "val_full/token_acc": 0.1304489451800188, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 2.2788533207187034, "epoch": 7, "val/loss": 1.151090940280356, "val/ppl": 3.1616401895944373, "val/token_acc": 0.7287251548304096, "val/seq_acc": 0.004497269514937359, "val_full/loss": 7.7966491831372515, "val_full/ppl": 2432.4376535790266, "val_full/token_acc": 0.12631014198394297, "val_full/seq_acc": 0.0, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.9741628886458285, "epoch": 8, "val/loss": 0.9012647076520286, "val/ppl": 2.4627157579117815, "val/token_acc": 0.7783406967119108, "val/seq_acc": 0.008352071956312239, "val_full/loss": 7.011426917070647, "val_full/ppl": 1109.2361642020871, "val_full/token_acc": 0.16438329916265879, "val_full/seq_acc": 0.0023094688221709007, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.7607509602335027, "epoch": 9, "val/loss": 0.7322117839463648, "val/ppl": 2.0796753170286784, "val/token_acc": 0.8156196706924236, "val/seq_acc": 0.01574044330228076, "val_full/loss": 6.722441211270574, "val_full/ppl": 830.8433018167867, "val_full/token_acc": 0.1844833202398973, "val_full/seq_acc": 0.0023094688221709007, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.6072707785632545, "epoch": 10, "val/loss": 0.638098754085413, "val/ppl": 1.892878628248731, "val/token_acc": 0.8388168064545083, "val/seq_acc": 0.0273048506264054, "val_full/loss": 6.620422362111747, "val_full/ppl": 750.261912407472, "val_full/token_acc": 0.20368276839947116, "val_full/seq_acc": 0.009237875288683603, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.0625, "gen_val_win/bit_acc": 0.7596008799497171, "gen_val_win/lev": 16.0078125, "gen_val_win/lev_norm": 0.12610161520236987, "gen_val_win/token_acc": 0.3641123274631128, "gen_val_win/bleu": 0.6404382598407361, "gen_val_win/rouge1": 0.8049537970832047, "gen_val_win/rouge2": 0.68604201758791, "gen_val_win/rougeL": 0.8042237353216579, "gen_val/n": 128.0, "gen_val/seq_acc": 0.0078125, "gen_val/bit_acc": 0.5627391010305949, "gen_val/lev": 197.3828125, "gen_val/lev_norm": 0.2425218587382373, "gen_val/token_acc": 0.06832864333557925, "gen_val/bleu": 0.527264293628675, "gen_val/rouge1": 0.7405585252014839, "gen_val/rouge2": 0.6223908061616639, "gen_val/rougeL": 0.7340760854405636, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0078125, "gen_val_full/bit_acc": 0.24735600187574205, "gen_val_full/lev": 610.1328125, "gen_val_full/lev_norm": 0.714567196137067, "gen_val_full/token_acc": 0.018877166541070082, "gen_val_full/bleu": 0.04987182029735536, "gen_val_full/rouge1": 0.26309011076183286, "gen_val_full/rouge2": 0.13794812321551939, "gen_val_full/rougeL": 0.22134718759848468 }, { "train/epoch_loss": 1.4831541544778686, "epoch": 11, "val/loss": 0.53435421331434, "val/ppl": 1.706345950873767, "val/token_acc": 0.8636794287480927, "val/seq_acc": 0.043366527465467396, "val_full/loss": 6.250100071973979, "val_full/ppl": 518.0646658281336, "val_full/token_acc": 0.24012722987602753, "val_full/seq_acc": 0.006928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.3880311767953888, "epoch": 12, "val/loss": 0.46726363072187654, "val/ppl": 1.5956220029535655, "val/token_acc": 0.8795663751234155, "val/seq_acc": 0.06488917442981047, "val_full/loss": 6.244773578364215, "val_full/ppl": 515.3125338098071, "val_full/token_acc": 0.24543486175247659, "val_full/seq_acc": 0.009237875288683603, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.3117436664062843, "epoch": 13, "val/loss": 0.4200845250387076, "val/ppl": 1.5220902049150007, "val/token_acc": 0.8917234295060387, "val/seq_acc": 0.07805974943784132, "val_full/loss": 6.244714148934642, "val_full/ppl": 515.281909989858, "val_full/token_acc": 0.2527352507233325, "val_full/seq_acc": 0.006928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.2502200576729905, "epoch": 14, "val/loss": 0.3869152949997694, "val/ppl": 1.4724317636578035, "val/token_acc": 0.899023645919558, "val/seq_acc": 0.09251525859299711, "val_full/loss": 6.005919665183276, "val_full/ppl": 405.8240394078294, "val_full/token_acc": 0.2704976144398245, "val_full/seq_acc": 0.011547344110854504, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.199270762112967, "epoch": 15, "val/loss": 0.35389138099325046, "val/ppl": 1.4246004393910614, "val/token_acc": 0.9086774840183104, "val/seq_acc": 0.10986186957918406, "val_full/loss": 6.192432492958286, "val_full/ppl": 489.03423290158656, "val_full/token_acc": 0.2578895935925195, "val_full/seq_acc": 0.016166281755196306, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.2265625, "gen_val_win/bit_acc": 0.8369911956105652, "gen_val_win/lev": 8.1875, "gen_val_win/lev_norm": 0.06252630647566886, "gen_val_win/token_acc": 0.5716736094094584, "gen_val_win/bleu": 0.789540025710193, "gen_val_win/rouge1": 0.8967285636361726, "gen_val_win/rouge2": 0.8273286565962761, "gen_val_win/rougeL": 0.8967285636361726, "gen_val/n": 128.0, "gen_val/seq_acc": 0.0078125, "gen_val/bit_acc": 0.6282546518508778, "gen_val/lev": 113.0859375, "gen_val/lev_norm": 0.14689947429206585, "gen_val/token_acc": 0.12761000483537738, "gen_val/bleu": 0.6973874922009002, "gen_val/rouge1": 0.8463161177245291, "gen_val/rouge2": 0.7678106345672289, "gen_val/rougeL": 0.843933897528292, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0078125, "gen_val_full/bit_acc": 0.2395406291805118, "gen_val_full/lev": 706.4765625, "gen_val_full/lev_norm": 0.7012449348833671, "gen_val_full/token_acc": 0.02245118616040196, "gen_val_full/bleu": 0.07419303060286554, "gen_val_full/rouge1": 0.2679363814726357, "gen_val_full/rouge2": 0.14769219870334008, "gen_val_full/rougeL": 0.22277268388857696 }, { "train/epoch_loss": 1.1549483582265954, "epoch": 16, "val/loss": 0.31556516386491046, "val/ppl": 1.371033950883735, "val/token_acc": 0.9177030248027844, "val/seq_acc": 0.13363315130099582, "val_full/loss": 5.822308026475886, "val_full/ppl": 337.75069226512585, "val_full/token_acc": 0.29075092452432505, "val_full/seq_acc": 0.013856812933025405, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.118490819621859, "epoch": 17, "val/loss": 0.2969314056733401, "val/ppl": 1.3457229873199037, "val/token_acc": 0.9234674033369569, "val/seq_acc": 0.1625441696113074, "val_full/loss": 6.052475362983579, "val_full/ppl": 425.1641641727175, "val_full/token_acc": 0.27586272969399683, "val_full/seq_acc": 0.023094688221709007, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.0856515222654082, "epoch": 18, "val/loss": 0.28207586462902334, "val/ppl": 1.3258793035128686, "val/token_acc": 0.9284538899582132, "val/seq_acc": 0.17860584645036942, "val_full/loss": 5.706772050863209, "val_full/ppl": 300.89821483396275, "val_full/token_acc": 0.29916265879783094, "val_full/seq_acc": 0.03233256351039261, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.0577344556996353, "epoch": 19, "val/loss": 0.26301759980696454, "val/ppl": 1.3008496134523408, "val/token_acc": 0.9329118089976165, "val/seq_acc": 0.19241888853196273, "val_full/loss": 5.9044769222202, "val_full/ppl": 366.67537587410794, "val_full/token_acc": 0.2831247964130372, "val_full/seq_acc": 0.025404157043879907, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 1.0343907209405876, "epoch": 20, "val/loss": 0.25522526245135446, "val/ppl": 1.2907523461645032, "val/token_acc": 0.9350161063517867, "val/seq_acc": 0.20558946353999358, "val_full/loss": 5.448175741056544, "val_full/ppl": 232.3339418067963, "val_full/token_acc": 0.31219222441510663, "val_full/seq_acc": 0.02771362586605081, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.234375, "gen_val_win/bit_acc": 0.8554692484369019, "gen_val_win/lev": 5.328125, "gen_val_win/lev_norm": 0.03916613653801227, "gen_val_win/token_acc": 0.6049562682215743, "gen_val_win/bleu": 0.8381115916312833, "gen_val_win/rouge1": 0.9267279458070664, "gen_val_win/rouge2": 0.8700091054588754, "gen_val_win/rougeL": 0.9267279458070664, "gen_val/n": 128.0, "gen_val/seq_acc": 0.0234375, "gen_val/bit_acc": 0.6670638690052131, "gen_val/lev": 81.796875, "gen_val/lev_norm": 0.1102067239985438, "gen_val/token_acc": 0.1839252506009888, "gen_val/bleu": 0.7737386319407429, "gen_val/rouge1": 0.8910038846170638, "gen_val/rouge2": 0.8324292306391359, "gen_val/rougeL": 0.8898227137651546, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0234375, "gen_val_full/bit_acc": 0.32297812137955617, "gen_val_full/lev": 644.3828125, "gen_val_full/lev_norm": 0.6681950287002262, "gen_val_full/token_acc": 0.022691022901682265, "gen_val_full/bleu": 0.09839284557174968, "gen_val_full/rouge1": 0.3288217650557424, "gen_val_full/rouge2": 0.19653596049434877, "gen_val_full/rougeL": 0.2578958906244661 }, { "train/epoch_loss": 1.0122020176223983, "epoch": 21, "val/loss": 0.23777997387577712, "val/ppl": 1.26843007436625, "val/token_acc": 0.940042484866013, "val/seq_acc": 0.22518470928364923, "val_full/loss": 5.419640183994711, "val_full/ppl": 225.79786219875206, "val_full/token_acc": 0.3242062503592711, "val_full/seq_acc": 0.03695150115473441, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.9939897338351109, "epoch": 22, "val/loss": 0.22978879153996576, "val/ppl": 1.2583342110300577, "val/token_acc": 0.9416481335580577, "val/seq_acc": 0.23514294892386764, "val_full/loss": 5.366829052218859, "val_full/ppl": 214.18262784325685, "val_full/token_acc": 0.32307574393071337, "val_full/seq_acc": 0.03233256351039261, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.9776245095486058, "epoch": 23, "val/loss": 0.22080388952579136, "val/ppl": 1.2470788413586418, "val/token_acc": 0.9448694039153893, "val/seq_acc": 0.254416961130742, "val_full/loss": 5.202963702240893, "val_full/ppl": 181.8102757168398, "val_full/token_acc": 0.33005039376113743, "val_full/seq_acc": 0.03926096997690531, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.961638304955347, "epoch": 24, "val/loss": 0.21207630866494379, "val/ppl": 1.2362422174724308, "val/token_acc": 0.9466445931525566, "val/seq_acc": 0.26019916479280436, "val_full/loss": 5.282296377110119, "val_full/ppl": 196.82133279855557, "val_full/token_acc": 0.330242005020215, "val_full/seq_acc": 0.046189376443418015, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.9482644060603401, "epoch": 25, "val/loss": 0.20377230890322381, "val/ppl": 1.2260189680698952, "val/token_acc": 0.9485494310418765, "val/seq_acc": 0.28172181175714744, "val_full/loss": 5.015769382077641, "val_full/ppl": 150.77209349922734, "val_full/token_acc": 0.3471037958190423, "val_full/seq_acc": 0.03695150115473441, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.390625, "gen_val_win/bit_acc": 0.9124491640307275, "gen_val_win/lev": 3.171875, "gen_val_win/lev_norm": 0.024217467751750245, "gen_val_win/token_acc": 0.7240009928021842, "gen_val_win/bleu": 0.886857168388725, "gen_val_win/rouge1": 0.9444426307964542, "gen_val_win/rouge2": 0.905085973956554, "gen_val_win/rougeL": 0.9444426307964542, "gen_val/n": 128.0, "gen_val/seq_acc": 0.0546875, "gen_val/bit_acc": 0.715185808048957, "gen_val/lev": 56.625, "gen_val/lev_norm": 0.07757158657628029, "gen_val/token_acc": 0.2344697860333882, "gen_val/bleu": 0.8272041845035049, "gen_val/rouge1": 0.9173450261938174, "gen_val/rouge2": 0.8721517697627691, "gen_val/rougeL": 0.9165902105502579, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0390625, "gen_val_full/bit_acc": 0.2619298668434054, "gen_val_full/lev": 594.4765625, "gen_val_full/lev_norm": 0.6808404487174158, "gen_val_full/token_acc": 0.025328769591064674, "gen_val_full/bleu": 0.09182151966783342, "gen_val_full/rouge1": 0.33321025973798307, "gen_val_full/rouge2": 0.21227233105110935, "gen_val_full/rougeL": 0.2732308977848592 }, { "train/epoch_loss": 0.9360308318661336, "epoch": 26, "val/loss": 0.19933261227064888, "val/ppl": 1.2205878808965673, "val/token_acc": 0.9503844581184989, "val/seq_acc": 0.2894314166398972, "val_full/loss": 5.190170132834505, "val_full/ppl": 179.49908902296949, "val_full/token_acc": 0.33873038379735193, "val_full/seq_acc": 0.04157043879907621, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.9254991490941987, "epoch": 27, "val/loss": 0.19405014994204572, "val/ppl": 1.214157171341223, "val/token_acc": 0.9513518365230226, "val/seq_acc": 0.2942499196916158, "val_full/loss": 5.019456327003775, "val_full/ppl": 151.32900792990333, "val_full/token_acc": 0.36066987296173525, "val_full/seq_acc": 0.04849884526558892, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.9159192450325983, "epoch": 28, "val/loss": 0.1910588487145636, "val/ppl": 1.2105306881601412, "val/token_acc": 0.9533264852250402, "val/seq_acc": 0.3135239318984902, "val_full/loss": 5.137287221684646, "val_full/ppl": 170.25328188214968, "val_full/token_acc": 0.3437505987851846, "val_full/seq_acc": 0.03695150115473441, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.9065227007628082, "epoch": 29, "val/loss": 0.18516592114836566, "val/ppl": 1.2034180960761303, "val/token_acc": 0.9541143501111986, "val/seq_acc": 0.32123353678123995, "val_full/loss": 5.164095957122191, "val_full/ppl": 174.87928871096668, "val_full/token_acc": 0.3512042767633026, "val_full/seq_acc": 0.04849884526558892, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8977805069854432, "epoch": 30, "val/loss": 0.1796077594319005, "val/ppl": 1.1967478579886603, "val/token_acc": 0.9550418366227523, "val/seq_acc": 0.32605203983295855, "val_full/loss": 5.207041279651842, "val_full/ppl": 182.55313469340206, "val_full/token_acc": 0.34631818965682426, "val_full/seq_acc": 0.04849884526558892, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.4765625, "gen_val_win/bit_acc": 0.9290884067357513, "gen_val_win/lev": 2.0859375, "gen_val_win/lev_norm": 0.017199462753933442, "gen_val_win/token_acc": 0.775118424333084, "gen_val_win/bleu": 0.9065473022474428, "gen_val_win/rouge1": 0.9534948095035979, "gen_val_win/rouge2": 0.9190306638019985, "gen_val_win/rougeL": 0.9534948095035979, "gen_val/n": 128.0, "gen_val/seq_acc": 0.0703125, "gen_val/bit_acc": 0.7390455845506547, "gen_val/lev": 46.875, "gen_val/lev_norm": 0.06591076993169548, "gen_val/token_acc": 0.27367025055652916, "gen_val/bleu": 0.8555323033741441, "gen_val/rouge1": 0.9286462069303709, "gen_val/rouge2": 0.8880547235922424, "gen_val/rougeL": 0.9281386055939489, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.03125, "gen_val_full/bit_acc": 0.31932685358138485, "gen_val_full/lev": 614.6484375, "gen_val_full/lev_norm": 0.656036086833709, "gen_val_full/token_acc": 0.02236951118475559, "gen_val_full/bleu": 0.10616910255511497, "gen_val_full/rouge1": 0.34030767142198753, "gen_val_full/rouge2": 0.22211420901161488, "gen_val_full/rougeL": 0.28067522370149545 }, { "train/epoch_loss": 0.8909414646928745, "epoch": 31, "val/loss": 0.17644704366060554, "val/ppl": 1.1929712496964127, "val/token_acc": 0.9560491069202461, "val/seq_acc": 0.3369739800835207, "val_full/loss": 4.972597494203759, "val_full/ppl": 144.40148266999464, "val_full/token_acc": 0.3619153461457395, "val_full/seq_acc": 0.04157043879907621, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8828385904840104, "epoch": 32, "val/loss": 0.17426898212655229, "val/ppl": 1.1903757125521166, "val/token_acc": 0.9571461339769225, "val/seq_acc": 0.3472534532605204, "val_full/loss": 5.272345390671406, "val_full/ppl": 194.87247897562096, "val_full/token_acc": 0.33982256797409416, "val_full/seq_acc": 0.04849884526558892, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8764852108503518, "epoch": 33, "val/loss": 0.1680034250126493, "val/ppl": 1.182940662227604, "val/token_acc": 0.9589512421338173, "val/seq_acc": 0.36684869900417605, "val_full/loss": 5.243562909569066, "val_full/ppl": 189.3435158621706, "val_full/token_acc": 0.342294353216195, "val_full/seq_acc": 0.053117782909930716, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8692677410463443, "epoch": 34, "val/loss": 0.1662353731105501, "val/ppl": 1.180851009591197, "val/token_acc": 0.9597391070199759, "val/seq_acc": 0.3710247349823322, "val_full/loss": 5.125430601395888, "val_full/ppl": 168.24657326545085, "val_full/token_acc": 0.35971181666634733, "val_full/seq_acc": 0.050808314087759814, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8641646281442142, "epoch": 35, "val/loss": 0.16242443123179154, "val/ppl": 1.1763594190574553, "val/token_acc": 0.9608361340766522, "val/seq_acc": 0.3787343398650819, "val_full/loss": 5.187735621970147, "val_full/ppl": 179.06262804067197, "val_full/token_acc": 0.3523731054436759, "val_full/seq_acc": 0.050808314087759814, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.5078125, "gen_val_win/bit_acc": 0.934458933437135, "gen_val_win/lev": 1.515625, "gen_val_win/lev_norm": 0.012608255240775584, "gen_val_win/token_acc": 0.7862842892768079, "gen_val_win/bleu": 0.9190175899136278, "gen_val_win/rouge1": 0.9629300041650583, "gen_val_win/rouge2": 0.9325652026415067, "gen_val_win/rougeL": 0.9629300041650583, "gen_val/n": 128.0, "gen_val/seq_acc": 0.09375, "gen_val/bit_acc": 0.758615883558934, "gen_val/lev": 36.1484375, "gen_val/lev_norm": 0.04977548676594661, "gen_val/token_acc": 0.2996242050491424, "gen_val/bleu": 0.8759583484058917, "gen_val/rouge1": 0.9427869829855836, "gen_val/rouge2": 0.9081305703292875, "gen_val/rougeL": 0.9425731720874744, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0546875, "gen_val_full/bit_acc": 0.2445423834667575, "gen_val_full/lev": 529.265625, "gen_val_full/lev_norm": 0.6818199514553388, "gen_val_full/token_acc": 0.033334754252099405, "gen_val_full/bleu": 0.06715355758530663, "gen_val_full/rouge1": 0.33110716118594996, "gen_val_full/rouge2": 0.23193709834381152, "gen_val_full/rougeL": 0.28028347532296566 }, { "train/epoch_loss": 0.8560918325795199, "epoch": 36, "val/loss": 0.1609244988561467, "val/ppl": 1.174596282103075, "val/token_acc": 0.9611752151668977, "val/seq_acc": 0.384195309990363, "val_full/loss": 5.36468143430608, "val_full/ppl": 213.72313897496375, "val_full/token_acc": 0.34240931997164153, "val_full/seq_acc": 0.050808314087759814, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8503464565907334, "epoch": 37, "val/loss": 0.15771310727542617, "val/ppl": 1.170830243840375, "val/token_acc": 0.9625714314208494, "val/seq_acc": 0.39029874718920654, "val_full/loss": 5.132414195580007, "val_full/ppl": 169.42565136546253, "val_full/token_acc": 0.35858131023778955, "val_full/seq_acc": 0.053117782909930716, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8460360304673117, "epoch": 38, "val/loss": 0.15577538608773545, "val/ppl": 1.168563697945893, "val/token_acc": 0.9631997287351278, "val/seq_acc": 0.4050754898811436, "val_full/loss": 5.050719949366725, "val_full/ppl": 156.1348332059374, "val_full/token_acc": 0.3668589166299412, "val_full/seq_acc": 0.053117782909930716, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8416729933603149, "epoch": 39, "val/loss": 0.15192259107822684, "val/ppl": 1.1640701235312552, "val/token_acc": 0.964137188219924, "val/seq_acc": 0.4124638612271121, "val_full/loss": 5.31732957675468, "val_full/ppl": 203.838818622384, "val_full/token_acc": 0.3524114276954914, "val_full/seq_acc": 0.057736720554272515, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8355102124356867, "epoch": 40, "val/loss": 0.14684852000297943, "val/ppl": 1.1581785088642205, "val/token_acc": 0.965712917992241, "val/seq_acc": 0.42788307099261164, "val_full/loss": 5.166461495418096, "val_full/ppl": 175.29346204387824, "val_full/token_acc": 0.3586196324896051, "val_full/seq_acc": 0.053117782909930716, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.5703125, "gen_val_win/bit_acc": 0.9493728472086826, "gen_val_win/lev": 1.4453125, "gen_val_win/lev_norm": 0.0121548993457164, "gen_val_win/token_acc": 0.8024937655860349, "gen_val_win/bleu": 0.9298298073616039, "gen_val_win/rouge1": 0.9655077197660651, "gen_val_win/rouge2": 0.9393315916832448, "gen_val_win/rougeL": 0.9655077197660651, "gen_val/n": 128.0, "gen_val/seq_acc": 0.078125, "gen_val/bit_acc": 0.7715689040421624, "gen_val/lev": 33.25, "gen_val/lev_norm": 0.04768892438456196, "gen_val/token_acc": 0.3463888755232642, "gen_val/bleu": 0.8883294517665103, "gen_val/rouge1": 0.9483271124692857, "gen_val/rouge2": 0.9168530717913945, "gen_val/rougeL": 0.9479645382085561, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.03125, "gen_val_full/bit_acc": 0.32233132791388186, "gen_val_full/lev": 518.3203125, "gen_val_full/lev_norm": 0.6488054957973685, "gen_val_full/token_acc": 0.030329453690715397, "gen_val_full/bleu": 0.0938472377889516, "gen_val_full/rouge1": 0.3601287541890134, "gen_val_full/rouge2": 0.24646846885896284, "gen_val_full/rougeL": 0.3034602996508524 }, { "train/epoch_loss": 0.8304888334060251, "epoch": 41, "val/loss": 0.14268469852096327, "val/ppl": 1.1533660862961466, "val/token_acc": 0.9666304315305522, "val/seq_acc": 0.4416961130742049, "val_full/loss": 5.358802112274617, "val_full/ppl": 212.47027840962429, "val_full/token_acc": 0.35145337140010346, "val_full/seq_acc": 0.06004618937644342, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.826247526523181, "epoch": 42, "val/loss": 0.13970463727989532, "val/ppl": 1.1499341010286386, "val/token_acc": 0.9673385126307706, "val/seq_acc": 0.4500481850305172, "val_full/loss": 5.450080551576961, "val_full/ppl": 232.77691570012897, "val_full/token_acc": 0.3342466803349365, "val_full/seq_acc": 0.057736720554272515, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8227115215506042, "epoch": 43, "val/loss": 0.1390314311509267, "val/ppl": 1.1491602188642516, "val/token_acc": 0.9682261072493542, "val/seq_acc": 0.4465146161259235, "val_full/loss": 5.193236661461228, "val_full/ppl": 180.0503729498964, "val_full/token_acc": 0.36227940753798693, "val_full/seq_acc": 0.06466512702078522, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8173399352076047, "epoch": 44, "val/loss": 0.1356666351058748, "val/ppl": 1.1453000271256764, "val/token_acc": 0.9691137018679379, "val/seq_acc": 0.46514616125923547, "val_full/loss": 5.500843089539941, "val_full/ppl": 244.89831646077423, "val_full/token_acc": 0.3399375347295407, "val_full/seq_acc": 0.05542725173210162, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8136944240168146, "epoch": 45, "val/loss": 0.1337494416315866, "val/ppl": 1.1431063688931162, "val/token_acc": 0.969562485663851, "val/seq_acc": 0.46900096370061034, "val_full/loss": 5.126671359146563, "val_full/ppl": 168.4554560649331, "val_full/token_acc": 0.3674145892812662, "val_full/seq_acc": 0.06235565819861432, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.5859375, "gen_val_win/bit_acc": 0.9471821841319309, "gen_val_win/lev": 1.25, "gen_val_win/lev_norm": 0.010280795885099246, "gen_val_win/token_acc": 0.8131539611360239, "gen_val_win/bleu": 0.9391197762238938, "gen_val_win/rouge1": 0.9711827774040873, "gen_val_win/rouge2": 0.9486937066297777, "gen_val_win/rougeL": 0.9711827774040873, "gen_val/n": 128.0, "gen_val/seq_acc": 0.1015625, "gen_val/bit_acc": 0.7821110627283063, "gen_val/lev": 28.25, "gen_val/lev_norm": 0.03832984563677537, "gen_val/token_acc": 0.351591737545565, "gen_val/bleu": 0.9043745164371645, "gen_val/rouge1": 0.9555825978731163, "gen_val/rouge2": 0.9295113795421714, "gen_val/rougeL": 0.9555155377872794, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0546875, "gen_val_full/bit_acc": 0.3742634223616899, "gen_val_full/lev": 552.1796875, "gen_val_full/lev_norm": 0.6319002319393148, "gen_val_full/token_acc": 0.033000545728869055, "gen_val_full/bleu": 0.12726190921975344, "gen_val_full/rouge1": 0.3730387399774349, "gen_val_full/rouge2": 0.25131445966374627, "gen_val_full/rougeL": 0.30975626539946866 }, { "train/epoch_loss": 0.8100172247078056, "epoch": 46, "val/loss": 0.13167286623952215, "val/ppl": 1.1407350852636906, "val/token_acc": 0.9703702964964945, "val/seq_acc": 0.4779955027304851, "val_full/loss": 5.344028135718255, "val_full/ppl": 209.35432170964398, "val_full/token_acc": 0.3525263944509379, "val_full/seq_acc": 0.06928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8063960959786489, "epoch": 47, "val/loss": 0.1294364190767629, "val/ppl": 1.1381867421982717, "val/token_acc": 0.9712479181418356, "val/seq_acc": 0.4815290716350787, "val_full/loss": 5.178985873459925, "val_full/ppl": 177.50270947732005, "val_full/token_acc": 0.36779781179942134, "val_full/seq_acc": 0.06928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.8027049875318856, "epoch": 48, "val/loss": 0.12705491853648876, "val/ppl": 1.1354793749346737, "val/token_acc": 0.9717066749109912, "val/seq_acc": 0.48827497590748475, "val_full/loss": 5.290455498883865, "val_full/ppl": 198.43379119747243, "val_full/token_acc": 0.36042077832493435, "val_full/seq_acc": 0.06466512702078522, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7997443438170854, "epoch": 49, "val/loss": 0.12609530869098648, "val/ppl": 1.134390280383623, "val/token_acc": 0.9716368640982936, "val/seq_acc": 0.49020237712817216, "val_full/loss": 5.219981027491426, "val_full/ppl": 184.93067543857435, "val_full/token_acc": 0.36523022092778173, "val_full/seq_acc": 0.06928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7962049421229564, "epoch": 50, "val/loss": 0.12423698263302155, "val/ppl": 1.1322841708908584, "val/token_acc": 0.9724347019576947, "val/seq_acc": 0.5024092515258592, "val_full/loss": 5.256799355467627, "val_full/ppl": 191.8664113158115, "val_full/token_acc": 0.36300753032248173, "val_full/seq_acc": 0.06928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.546875, "gen_val_win/bit_acc": 0.9418925947597685, "gen_val_win/lev": 1.2734375, "gen_val_win/lev_norm": 0.010597328446190442, "gen_val_win/token_acc": 0.7919680718383637, "gen_val_win/bleu": 0.9346683056433839, "gen_val_win/rouge1": 0.9676755114021661, "gen_val_win/rouge2": 0.9431690249745048, "gen_val_win/rougeL": 0.9676755114021661, "gen_val/n": 128.0, "gen_val/seq_acc": 0.1015625, "gen_val/bit_acc": 0.7954056396765551, "gen_val/lev": 25.1796875, "gen_val/lev_norm": 0.03469096867452959, "gen_val/token_acc": 0.4104618971295352, "gen_val/bleu": 0.9148643251858689, "gen_val/rouge1": 0.9612759444173142, "gen_val/rouge2": 0.9383258717336574, "gen_val/rougeL": 0.9611566745210505, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0546875, "gen_val_full/bit_acc": 0.3792326461507818, "gen_val_full/lev": 568.140625, "gen_val_full/lev_norm": 0.6416667424147317, "gen_val_full/token_acc": 0.03232423696152172, "gen_val_full/bleu": 0.12455500585887742, "gen_val_full/rouge1": 0.36369442228980675, "gen_val_full/rouge2": 0.2485710655072301, "gen_val_full/rougeL": 0.29943750735353 }, { "train/epoch_loss": 0.7935096948521393, "epoch": 51, "val/loss": 0.12211543709954227, "val/ppl": 1.1298845248447096, "val/token_acc": 0.973102891164943, "val/seq_acc": 0.5101188564086091, "val_full/loss": 5.14476754560348, "val_full/ppl": 171.53160676021028, "val_full/token_acc": 0.37293299354270054, "val_full/seq_acc": 0.07390300230946882, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7909734631119821, "epoch": 52, "val/loss": 0.11956716591250602, "val/ppl": 1.1270089381088781, "val/token_acc": 0.9738109722651614, "val/seq_acc": 0.5165435271442339, "val_full/loss": 5.223838476498879, "val_full/ppl": 185.6454137359905, "val_full/token_acc": 0.3640039088696852, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7875587506187229, "epoch": 53, "val/loss": 0.12078814995346984, "val/ppl": 1.128385838451948, "val/token_acc": 0.9736613776665237, "val/seq_acc": 0.512688724702859, "val_full/loss": 5.316853243140317, "val_full/ppl": 203.7417464623638, "val_full/token_acc": 0.3535419341240491, "val_full/seq_acc": 0.07390300230946882, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7846282487200977, "epoch": 54, "val/loss": 0.11707850749736769, "val/ppl": 1.124207684958009, "val/token_acc": 0.9744691884991673, "val/seq_acc": 0.52071956312239, "val_full/loss": 5.237640699548756, "val_full/ppl": 188.22549763227875, "val_full/token_acc": 0.36473203165418, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7828222343749239, "epoch": 55, "val/loss": 0.11614146314732193, "val/ppl": 1.1231547459008828, "val/token_acc": 0.9748381885091403, "val/seq_acc": 0.5300353356890459, "val_full/loss": 5.197993672936347, "val_full/ppl": 180.908915068467, "val_full/token_acc": 0.3668972388817567, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.5234375, "gen_val_win/bit_acc": 0.9366707856399584, "gen_val_win/lev": 1.4296875, "gen_val_win/lev_norm": 0.01176483948693388, "gen_val_win/token_acc": 0.799800299550674, "gen_val_win/bleu": 0.9290983820929968, "gen_val_win/rouge1": 0.965755473669198, "gen_val_win/rouge2": 0.9411171707758922, "gen_val_win/rougeL": 0.965755473669198, "gen_val/n": 128.0, "gen_val/seq_acc": 0.1640625, "gen_val/bit_acc": 0.8011712959587859, "gen_val/lev": 24.7265625, "gen_val/lev_norm": 0.03567827127089045, "gen_val/token_acc": 0.39545210559703314, "gen_val/bleu": 0.9164283567075726, "gen_val/rouge1": 0.9624503069182171, "gen_val/rouge2": 0.9409586148036581, "gen_val/rougeL": 0.9624026697230952, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.3757417117538352, "gen_val_full/lev": 500.8515625, "gen_val_full/lev_norm": 0.6181358328513049, "gen_val_full/token_acc": 0.03886472224203095, "gen_val_full/bleu": 0.11915294477701922, "gen_val_full/rouge1": 0.39043593908446095, "gen_val_full/rouge2": 0.2787649988337111, "gen_val_full/rougeL": 0.3261082252920391 }, { "train/epoch_loss": 0.7798933871369113, "epoch": 56, "val/loss": 0.11468604560401079, "val/ppl": 1.1215212757587365, "val/token_acc": 0.9749079993218378, "val/seq_acc": 0.5326052039832958, "val_full/loss": 5.27014859752534, "val_full/ppl": 194.4448543228764, "val_full/token_acc": 0.3668589166299412, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7770069360435752, "epoch": 57, "val/loss": 0.11484508718129241, "val/ppl": 1.1216996584561447, "val/token_acc": 0.9753468101445084, "val/seq_acc": 0.5374237070350144, "val_full/loss": 5.390086617270881, "val_full/ppl": 219.22237317529513, "val_full/token_acc": 0.3521431719327828, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7758665425224495, "epoch": 58, "val/loss": 0.11302239058068163, "val/ppl": 1.1196570024383736, "val/token_acc": 0.975516350689631, "val/seq_acc": 0.5415997430131706, "val_full/loss": 5.211792827212152, "val_full/ppl": 183.42260863271198, "val_full/token_acc": 0.36733794477763515, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7734360559622843, "epoch": 59, "val/loss": 0.11200202885361335, "val/ppl": 1.1185151299462062, "val/token_acc": 0.9760449182714843, "val/seq_acc": 0.5470607131384516, "val_full/loss": 5.517750911949836, "val_full/ppl": 249.0742169141425, "val_full/token_acc": 0.3381938722719347, "val_full/seq_acc": 0.06928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7710748243510277, "epoch": 60, "val/loss": 0.11054647731977883, "val/ppl": 1.1168882578185346, "val/token_acc": 0.9764238912546997, "val/seq_acc": 0.5566977192418888, "val_full/loss": 5.328467205733009, "val_full/ppl": 206.12178959787667, "val_full/token_acc": 0.3525838778286612, "val_full/seq_acc": 0.06928406466512702, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.6015625, "gen_val_win/bit_acc": 0.9388405147576669, "gen_val_win/lev": 3.1015625, "gen_val_win/lev_norm": 0.015023159623174206, "gen_val_win/token_acc": 0.8217016029593095, "gen_val_win/bleu": 0.9252322601744094, "gen_val_win/rouge1": 0.9687495089006884, "gen_val_win/rouge2": 0.9471590340864586, "gen_val_win/rougeL": 0.9687495089006884, "gen_val/n": 128.0, "gen_val/seq_acc": 0.171875, "gen_val/bit_acc": 0.787561823285657, "gen_val/lev": 24.1015625, "gen_val/lev_norm": 0.0344065360000764, "gen_val/token_acc": 0.4139342265529842, "gen_val/bleu": 0.9224372884442231, "gen_val/rouge1": 0.9641586755754376, "gen_val/rouge2": 0.9443289078504787, "gen_val/rougeL": 0.9641255014777731, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0546875, "gen_val_full/bit_acc": 0.3918829288835079, "gen_val_full/lev": 525.8671875, "gen_val_full/lev_norm": 0.6262776687851552, "gen_val_full/token_acc": 0.03767860638089646, "gen_val_full/bleu": 0.12005813143076255, "gen_val_full/rouge1": 0.3709799318307528, "gen_val_full/rouge2": 0.25285862530257186, "gen_val_full/rougeL": 0.3057391487110338 }, { "train/epoch_loss": 0.7692888557465, "epoch": 61, "val/loss": 0.10926879314589201, "val/ppl": 1.1154621386264896, "val/token_acc": 0.9764937020673974, "val/seq_acc": 0.5544490844844202, "val_full/loss": 5.5136764781611065, "val_full/ppl": 248.06144514568888, "val_full/token_acc": 0.34595412826457683, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.767349373967273, "epoch": 62, "val/loss": 0.10924034142619984, "val/ppl": 1.1154304022618733, "val/token_acc": 0.9767330534252177, "val/seq_acc": 0.5560552521683264, "val_full/loss": 5.578898181369637, "val_full/ppl": 264.7797057925477, "val_full/token_acc": 0.3364502098143287, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7658214072634157, "epoch": 63, "val/loss": 0.10810739653960619, "val/ppl": 1.114167396684231, "val/token_acc": 0.9771020534351906, "val/seq_acc": 0.5650497911982011, "val_full/loss": 5.402696284777443, "val_full/ppl": 222.00419648460908, "val_full/token_acc": 0.34903906953572594, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7634172452952797, "epoch": 64, "val/loss": 0.10817990131509235, "val/ppl": 1.1142481820698091, "val/token_acc": 0.9772516480338284, "val/seq_acc": 0.5611949887568262, "val_full/loss": 5.360503036439671, "val_full/ppl": 212.83198176831377, "val_full/token_acc": 0.35151085477782673, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7615973996997177, "epoch": 65, "val/loss": 0.1061087566860069, "val/ppl": 1.111942801144812, "val/token_acc": 0.9775907291240737, "val/seq_acc": 0.5660134918085448, "val_full/loss": 5.4703817644762305, "val_full/ppl": 237.55086393369695, "val_full/token_acc": 0.3476594684703673, "val_full/seq_acc": 0.08545034642032333, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.625, "gen_val_win/bit_acc": 0.9240180500826236, "gen_val_win/lev": 3.984375, "gen_val_win/lev_norm": 0.016169124204950366, "gen_val_win/token_acc": 0.8010792249202845, "gen_val_win/bleu": 0.9230816637610787, "gen_val_win/rouge1": 0.9682264551272864, "gen_val_win/rouge2": 0.9492211007789696, "gen_val_win/rougeL": 0.9682264551272864, "gen_val/n": 128.0, "gen_val/seq_acc": 0.203125, "gen_val/bit_acc": 0.8057363574600247, "gen_val/lev": 24.1953125, "gen_val/lev_norm": 0.03162261251077031, "gen_val/token_acc": 0.4415028226591396, "gen_val/bleu": 0.9247399313901736, "gen_val/rouge1": 0.9666540727733139, "gen_val/rouge2": 0.9480247759222942, "gen_val/rougeL": 0.9666056981603108, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.34558169266172334, "gen_val_full/lev": 556.265625, "gen_val_full/lev_norm": 0.6461050540429291, "gen_val_full/token_acc": 0.035525197725341524, "gen_val_full/bleu": 0.1176912809591565, "gen_val_full/rouge1": 0.35404522179498393, "gen_val_full/rouge2": 0.24455123647055177, "gen_val_full/rougeL": 0.29453265957843894 }, { "train/epoch_loss": 0.7608585213485204, "epoch": 66, "val/loss": 0.1095111308717146, "val/ppl": 1.1157324899412437, "val/token_acc": 0.9770920804619482, "val/seq_acc": 0.5634436235142949, "val_full/loss": 5.420797560908429, "val_full/ppl": 226.0593467205135, "val_full/token_acc": 0.347448696085382, "val_full/seq_acc": 0.07390300230946882, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7589125273530918, "epoch": 67, "val/loss": 0.10670304391846144, "val/ppl": 1.1126038109501415, "val/token_acc": 0.9779497561608043, "val/seq_acc": 0.5737230966912946, "val_full/loss": 5.493937000613156, "val_full/ppl": 243.21285358639852, "val_full/token_acc": 0.3474870183371975, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7570422626195703, "epoch": 68, "val/loss": 0.10556327852618037, "val/ppl": 1.1113364260290377, "val/token_acc": 0.9778899183213492, "val/seq_acc": 0.5692258271763572, "val_full/loss": 5.432119765419916, "val_full/ppl": 228.6333812509411, "val_full/token_acc": 0.34719960144858114, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7554686881062991, "epoch": 69, "val/loss": 0.10390940998502457, "val/ppl": 1.1094999407466137, "val/token_acc": 0.9779397831875617, "val/seq_acc": 0.5753292643752008, "val_full/loss": 5.518696318908199, "val_full/ppl": 249.3098047576079, "val_full/token_acc": 0.34212190308302515, "val_full/seq_acc": 0.07621247113163972, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7545096016584192, "epoch": 70, "val/loss": 0.10571780794785876, "val/ppl": 1.111508173473917, "val/token_acc": 0.977850026428379, "val/seq_acc": 0.5685833601027948, "val_full/loss": 5.426737060850946, "val_full/ppl": 227.40602152587084, "val_full/token_acc": 0.3519707217996129, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.6171875, "gen_val_win/bit_acc": 0.9549847074900761, "gen_val_win/lev": 1.109375, "gen_val_win/lev_norm": 0.009229312465825404, "gen_val_win/token_acc": 0.8390316945345645, "gen_val_win/bleu": 0.9452998286784585, "gen_val_win/rouge1": 0.9735241637088105, "gen_val_win/rouge2": 0.954144793721871, "gen_val_win/rougeL": 0.9735241637088105, "gen_val/n": 128.0, "gen_val/seq_acc": 0.1953125, "gen_val/bit_acc": 0.8208642782107346, "gen_val/lev": 19.8359375, "gen_val/lev_norm": 0.029189174862170115, "gen_val/token_acc": 0.4288859533245266, "gen_val/bleu": 0.9331933491874678, "gen_val/rouge1": 0.9686683826556418, "gen_val/rouge2": 0.9502815413471055, "gen_val/rougeL": 0.9685862607424227, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0546875, "gen_val_full/bit_acc": 0.3671029687313845, "gen_val_full/lev": 521.71875, "gen_val_full/lev_norm": 0.6435382618336629, "gen_val_full/token_acc": 0.03712480252764613, "gen_val_full/bleu": 0.10995822767720215, "gen_val_full/rouge1": 0.36238189831126444, "gen_val_full/rouge2": 0.2479118388065199, "gen_val_full/rougeL": 0.300873198260952 }, { "train/epoch_loss": 0.7535619166426528, "epoch": 71, "val/loss": 0.10508038048431718, "val/ppl": 1.110799893400761, "val/token_acc": 0.978199080491867, "val/seq_acc": 0.5820751686476068, "val_full/loss": 5.591022234091475, "val_full/ppl": 268.00944812614483, "val_full/token_acc": 0.3339401023204123, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7523766571744125, "epoch": 72, "val/loss": 0.10527595871316661, "val/ppl": 1.11101716292242, "val/token_acc": 0.9781492156256545, "val/seq_acc": 0.5788628332797944, "val_full/loss": 5.439379586694514, "val_full/ppl": 230.2992584047655, "val_full/token_acc": 0.34662476767134837, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7505472156769617, "epoch": 73, "val/loss": 0.10373660443856816, "val/ppl": 1.1093082295679142, "val/token_acc": 0.9783087831975347, "val/seq_acc": 0.5804690009637006, "val_full/loss": 5.596636443982448, "val_full/ppl": 269.51834107828176, "val_full/token_acc": 0.3318515395964667, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7498734516098612, "epoch": 74, "val/loss": 0.10334283763851525, "val/ppl": 1.1088715068051902, "val/token_acc": 0.9786678102342652, "val/seq_acc": 0.5833601027947317, "val_full/loss": 5.616988029685374, "val_full/ppl": 275.05966277632007, "val_full/token_acc": 0.3316982505892046, "val_full/seq_acc": 0.07621247113163972, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7482658022656999, "epoch": 75, "val/loss": 0.10210829080438107, "val/ppl": 1.107503397668166, "val/token_acc": 0.9787176751004777, "val/seq_acc": 0.583681336331513, "val_full/loss": 5.644142339022591, "val_full/ppl": 282.63105062072907, "val_full/token_acc": 0.3300120715093219, "val_full/seq_acc": 0.07159353348729793, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.6484375, "gen_val_win/bit_acc": 0.9476811781423589, "gen_val_win/lev": 1.90625, "gen_val_win/lev_norm": 0.01187550800905095, "gen_val_win/token_acc": 0.8533498759305211, "gen_val_win/bleu": 0.9433507117683827, "gen_val_win/rouge1": 0.9736264156205103, "gen_val_win/rouge2": 0.9563676193284467, "gen_val_win/rougeL": 0.9736264156205103, "gen_val/n": 128.0, "gen_val/seq_acc": 0.2421875, "gen_val/bit_acc": 0.8271393643031785, "gen_val/lev": 18.8125, "gen_val/lev_norm": 0.026790460131742057, "gen_val/token_acc": 0.4637106641217638, "gen_val/bleu": 0.935424432617851, "gen_val/rouge1": 0.9709846665014976, "gen_val/rouge2": 0.95441035265913, "gen_val/rougeL": 0.9709517717646555, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.34337851148687143, "gen_val_full/lev": 600.1015625, "gen_val_full/lev_norm": 0.6594099125553171, "gen_val_full/token_acc": 0.032464076636508785, "gen_val_full/bleu": 0.11365944691942924, "gen_val_full/rouge1": 0.34628273502360013, "gen_val_full/rouge2": 0.2298203567705291, "gen_val_full/rougeL": 0.2840028360407299 }, { "train/epoch_loss": 0.7471561143523142, "epoch": 76, "val/loss": 0.10176311059638243, "val/ppl": 1.1071211753865784, "val/token_acc": 0.978588026448325, "val/seq_acc": 0.5843238034050755, "val_full/loss": 5.645573211069382, "val_full/ppl": 283.035748957338, "val_full/token_acc": 0.32731035275632797, "val_full/seq_acc": 0.07621247113163972, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7469619139471554, "epoch": 77, "val/loss": 0.10257696416575826, "val/ppl": 1.1080225766614666, "val/token_acc": 0.9790866751104507, "val/seq_acc": 0.5878573723096692, "val_full/loss": 5.633106204133055, "val_full/ppl": 279.5290447838613, "val_full/token_acc": 0.3300120715093219, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7454144519166161, "epoch": 78, "val/loss": 0.10266572934941523, "val/ppl": 1.1081209348543068, "val/token_acc": 0.9785979994215676, "val/seq_acc": 0.5843238034050755, "val_full/loss": 5.641723022332292, "val_full/ppl": 281.9481030690838, "val_full/token_acc": 0.32993542700569084, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7446135588714904, "epoch": 79, "val/loss": 0.10225602011188839, "val/ppl": 1.1076670204638093, "val/token_acc": 0.9788473237526304, "val/seq_acc": 0.5881786058464503, "val_full/loss": 5.644910430718159, "val_full/ppl": 282.8482205761633, "val_full/token_acc": 0.331429994826496, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7434256564351983, "epoch": 80, "val/loss": 0.10209512184139186, "val/ppl": 1.107488813092944, "val/token_acc": 0.9790368102442382, "val/seq_acc": 0.5913909412142627, "val_full/loss": 5.628657005666903, "val_full/ppl": 278.28812717795273, "val_full/token_acc": 0.33104677230834084, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.6328125, "gen_val_win/bit_acc": 0.9589264820592823, "gen_val_win/lev": 1.46875, "gen_val_win/lev_norm": 0.011870788211873987, "gen_val_win/token_acc": 0.837447020693094, "gen_val_win/bleu": 0.9458252409859191, "gen_val_win/rouge1": 0.9736477876341096, "gen_val_win/rouge2": 0.9547537997486909, "gen_val_win/rougeL": 0.9736477876341096, "gen_val/n": 128.0, "gen_val/seq_acc": 0.2265625, "gen_val/bit_acc": 0.8308444020449598, "gen_val/lev": 19.421875, "gen_val/lev_norm": 0.027883458066460365, "gen_val/token_acc": 0.48211485691885536, "gen_val/bleu": 0.9346281821713329, "gen_val/rouge1": 0.970167401334752, "gen_val/rouge2": 0.952445849287562, "gen_val/rougeL": 0.9700579496696512, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.33647078636998906, "gen_val_full/lev": 574.203125, "gen_val_full/lev_norm": 0.6577566522728574, "gen_val_full/token_acc": 0.03564499687613035, "gen_val_full/bleu": 0.10642088306875847, "gen_val_full/rouge1": 0.3347771511495598, "gen_val_full/rouge2": 0.2283691153378196, "gen_val_full/rougeL": 0.2830156498412585 }, { "train/epoch_loss": 0.7427948204953772, "epoch": 81, "val/loss": 0.1017492253433381, "val/ppl": 1.1071058028356335, "val/token_acc": 0.9790368102442382, "val/seq_acc": 0.5920334082878252, "val_full/loss": 5.572770887423595, "val_full/ppl": 263.1622829770605, "val_full/token_acc": 0.33472570848263045, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7426253491803595, "epoch": 82, "val/loss": 0.1011202050432829, "val/ppl": 1.106409629787785, "val/token_acc": 0.9790966480836932, "val/seq_acc": 0.5920334082878252, "val_full/loss": 5.594441147799345, "val_full/ppl": 268.9273174661375, "val_full/token_acc": 0.33342275192090287, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7415810908165359, "epoch": 83, "val/loss": 0.10094672155499612, "val/ppl": 1.1062177026343127, "val/token_acc": 0.9789869453780256, "val/seq_acc": 0.5907484741407003, "val_full/loss": 5.5852166374858685, "val_full/ppl": 266.4580012787265, "val_full/token_acc": 0.33643104868842094, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7408571424626947, "epoch": 84, "val/loss": 0.1009505429860274, "val/ppl": 1.1062219299770462, "val/token_acc": 0.9792362697090884, "val/seq_acc": 0.5929971088981689, "val_full/loss": 5.707580410862442, "val_full/ppl": 301.1415472515213, "val_full/token_acc": 0.32464695625514955, "val_full/seq_acc": 0.07621247113163972, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7399543473250848, "epoch": 85, "val/loss": 0.10107410515099295, "val/ppl": 1.1063586255986761, "val/token_acc": 0.9791963778161183, "val/seq_acc": 0.5929971088981689, "val_full/loss": 5.687350004311253, "val_full/ppl": 295.11054185742114, "val_full/token_acc": 0.3250493398992125, "val_full/seq_acc": 0.07852193995381063, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.65625, "gen_val_win/bit_acc": 0.936570142391929, "gen_val_win/lev": 3.5703125, "gen_val_win/lev_norm": 0.015941331271142705, "gen_val_win/token_acc": 0.8228346456692913, "gen_val_win/bleu": 0.9309919214200336, "gen_val_win/rouge1": 0.9705177049034177, "gen_val_win/rouge2": 0.9530451958188538, "gen_val_win/rougeL": 0.9705177049034177, "gen_val/n": 128.0, "gen_val/seq_acc": 0.265625, "gen_val/bit_acc": 0.8383745107632093, "gen_val/lev": 19.0625, "gen_val/lev_norm": 0.025524067608020805, "gen_val/token_acc": 0.4900353426271353, "gen_val/bleu": 0.936846118862419, "gen_val/rouge1": 0.971977426375064, "gen_val/rouge2": 0.9559564493688095, "gen_val/rougeL": 0.9719442522773994, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0703125, "gen_val_full/bit_acc": 0.33686501520985346, "gen_val_full/lev": 549.171875, "gen_val_full/lev_norm": 0.6643301571195462, "gen_val_full/token_acc": 0.03650568672866181, "gen_val_full/bleu": 0.0961379279523444, "gen_val_full/rouge1": 0.33109183900682865, "gen_val_full/rouge2": 0.2272072947724632, "gen_val_full/rougeL": 0.2799981833360931 }, { "train/epoch_loss": 0.7401719008597947, "epoch": 86, "val/loss": 0.10011620043191252, "val/ppl": 1.1052993468752617, "val/token_acc": 0.9791963778161183, "val/seq_acc": 0.5920334082878252, "val_full/loss": 5.726109444159689, "val_full/ppl": 306.77342449878154, "val_full/token_acc": 0.3233248385675142, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7391969173626413, "epoch": 87, "val/loss": 0.1004848159704973, "val/ppl": 1.1057068524911442, "val/token_acc": 0.9795254859331213, "val/seq_acc": 0.5974943784131064, "val_full/loss": 5.680714416591619, "val_full/ppl": 293.1587926329848, "val_full/token_acc": 0.3283833758071624, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7387118480746585, "epoch": 88, "val/loss": 0.10021054421704194, "val/ppl": 1.1054036299184982, "val/token_acc": 0.9794157832274536, "val/seq_acc": 0.5907484741407003, "val_full/loss": 5.7281915699429, "val_full/ppl": 307.4128307866947, "val_full/token_acc": 0.3224625879016651, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7382475876451431, "epoch": 89, "val/loss": 0.10013674062370566, "val/ppl": 1.1053220501689995, "val/token_acc": 0.979365918361241, "val/seq_acc": 0.5926758753613878, "val_full/loss": 5.713398656805074, "val_full/ppl": 302.89876985650113, "val_full/token_acc": 0.3234781275747763, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7378488964868009, "epoch": 90, "val/loss": 0.10072225372146756, "val/ppl": 1.105969420209943, "val/token_acc": 0.9793758913344836, "val/seq_acc": 0.5936395759717314, "val_full/loss": 5.65790436861695, "val_full/ppl": 286.54751494005836, "val_full/token_acc": 0.32836421468125465, "val_full/seq_acc": 0.08545034642032333, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.65625, "gen_val_win/bit_acc": 0.9397923322683706, "gen_val_win/lev": 3.453125, "gen_val_win/lev_norm": 0.015042173523238098, "gen_val_win/token_acc": 0.8247600295348265, "gen_val_win/bleu": 0.9334896953527388, "gen_val_win/rouge1": 0.971912387630821, "gen_val_win/rouge2": 0.9553194334608761, "gen_val_win/rougeL": 0.971912387630821, "gen_val/n": 128.0, "gen_val/seq_acc": 0.25, "gen_val/bit_acc": 0.8328767206872987, "gen_val/lev": 20.34375, "gen_val/lev_norm": 0.028285833316708944, "gen_val/token_acc": 0.4728242258206546, "gen_val/bleu": 0.9344087206825136, "gen_val/rouge1": 0.9704404877921282, "gen_val/rouge2": 0.9538362591267303, "gen_val/rougeL": 0.9704068856415906, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.355097238973122, "gen_val_full/lev": 522.0703125, "gen_val_full/lev_norm": 0.639539040440194, "gen_val_full/token_acc": 0.038817536055105115, "gen_val_full/bleu": 0.10566389750921357, "gen_val_full/rouge1": 0.34989792471680825, "gen_val_full/rouge2": 0.237879814249418, "gen_val_full/rougeL": 0.2906843767576277 }, { "train/epoch_loss": 0.737777391425392, "epoch": 91, "val/loss": 0.09986305337632671, "val/ppl": 1.1050195790127586, "val/token_acc": 0.9795055399866363, "val/seq_acc": 0.5962094442659813, "val_full/loss": 5.693944874638333, "val_full/ppl": 297.0631892642554, "val_full/token_acc": 0.3261032018241392, "val_full/seq_acc": 0.08545034642032333, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7368834512192115, "epoch": 92, "val/loss": 0.09936335234456517, "val/ppl": 1.104467537528346, "val/token_acc": 0.9794756210669087, "val/seq_acc": 0.5962094442659813, "val_full/loss": 5.677080015065435, "val_full/ppl": 292.09526967501756, "val_full/token_acc": 0.3263331353350323, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.737032663643806, "epoch": 93, "val/loss": 0.09950750781057163, "val/ppl": 1.1046267640373175, "val/token_acc": 0.9795753507993338, "val/seq_acc": 0.5965306778027626, "val_full/loss": 5.6828042517340815, "val_full/ppl": 293.7720867987923, "val_full/token_acc": 0.32495353426967366, "val_full/seq_acc": 0.08545034642032333, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7366967462839331, "epoch": 94, "val/loss": 0.09931489304731249, "val/ppl": 1.1044140171044305, "val/token_acc": 0.9795952967458188, "val/seq_acc": 0.5978156119498875, "val_full/loss": 5.678785280423078, "val_full/ppl": 292.5937945573253, "val_full/token_acc": 0.3254134012914599, "val_full/seq_acc": 0.08545034642032333, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7365873708689302, "epoch": 95, "val/loss": 0.09952979725968017, "val/ppl": 1.1046513858337605, "val/token_acc": 0.9796651075585164, "val/seq_acc": 0.5949245101188564, "val_full/loss": 5.713141852809021, "val_full/ppl": 302.82099422897585, "val_full/token_acc": 0.3227500047902815, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.6328125, "gen_val_win/bit_acc": 0.9246040544821033, "gen_val_win/lev": 4.5625, "gen_val_win/lev_norm": 0.016696424812685837, "gen_val_win/token_acc": 0.8137542829172785, "gen_val_win/bleu": 0.9220869414589093, "gen_val_win/rouge1": 0.9701212523269171, "gen_val_win/rouge2": 0.9529250186396473, "gen_val_win/rougeL": 0.9701212523269171, "gen_val/n": 128.0, "gen_val/seq_acc": 0.2421875, "gen_val/bit_acc": 0.8262547146854848, "gen_val/lev": 22.0546875, "gen_val/lev_norm": 0.03003813391315662, "gen_val/token_acc": 0.47081484371190324, "gen_val/bleu": 0.931260298878397, "gen_val/rouge1": 0.9696616974165042, "gen_val/rouge2": 0.9529502680523851, "gen_val/rougeL": 0.9696281673735857, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.3551788052891724, "gen_val_full/lev": 540.96875, "gen_val_full/lev_norm": 0.6521694306109027, "gen_val_full/token_acc": 0.03675133059354578, "gen_val_full/bleu": 0.1015755239554731, "gen_val_full/rouge1": 0.34066941377132964, "gen_val_full/rouge2": 0.2313868540644178, "gen_val_full/rougeL": 0.2839247701539138 }, { "train/epoch_loss": 0.7363973328300247, "epoch": 96, "val/loss": 0.09948764023446655, "val/ppl": 1.1046048179990233, "val/token_acc": 0.9797049994514865, "val/seq_acc": 0.5971731448763251, "val_full/loss": 5.77722736112974, "val_full/ppl": 322.86276640559106, "val_full/token_acc": 0.31815133457241945, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7360071862427671, "epoch": 97, "val/loss": 0.10000829233185782, "val/ppl": 1.1051800825576574, "val/token_acc": 0.9797448913444565, "val/seq_acc": 0.59845807902345, "val_full/loss": 5.743435518021039, "val_full/ppl": 312.1349161880764, "val_full/token_acc": 0.3214662093544617, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7357251083761677, "epoch": 98, "val/loss": 0.09932624503489099, "val/ppl": 1.1044265544697962, "val/token_acc": 0.9797249453979715, "val/seq_acc": 0.5974943784131064, "val_full/loss": 5.735954670165169, "val_full/ppl": 309.80859464732436, "val_full/token_acc": 0.32181110962080134, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7358884735594962, "epoch": 99, "val/loss": 0.09961430160556396, "val/ppl": 1.1047447376208104, "val/token_acc": 0.9796252156655464, "val/seq_acc": 0.5974943784131064, "val_full/loss": 5.736716100375558, "val_full/ppl": 310.04458210331444, "val_full/token_acc": 0.32100634233267544, "val_full/seq_acc": 0.08314087759815242, "system/peak_gpu_mem_gb": 1.0650625228881836 }, { "train/epoch_loss": 0.7359878973770617, "epoch": 100, "val/loss": 0.09927769922586938, "val/ppl": 1.104372940490582, "val/token_acc": 0.9795753507993338, "val/seq_acc": 0.5978156119498875, "val_full/loss": 5.748038529431489, "val_full/ppl": 313.5749885610196, "val_full/token_acc": 0.3209488589549522, "val_full/seq_acc": 0.08083140877598152, "system/peak_gpu_mem_gb": 1.0650625228881836, "gen_val_win/n": 128.0, "gen_val_win/seq_acc": 0.6328125, "gen_val_win/bit_acc": 0.9441492797622892, "gen_val_win/lev": 2.1796875, "gen_val_win/lev_norm": 0.014352797667106544, "gen_val_win/token_acc": 0.8238946845504223, "gen_val_win/bleu": 0.9441925726890976, "gen_val_win/rouge1": 0.9727711842923557, "gen_val_win/rouge2": 0.9555740634354313, "gen_val_win/rougeL": 0.9727711842923557, "gen_val/n": 128.0, "gen_val/seq_acc": 0.25, "gen_val/bit_acc": 0.8362129070137241, "gen_val/lev": 18.828125, "gen_val/lev_norm": 0.027976049053401193, "gen_val/token_acc": 0.48612336961851527, "gen_val/bleu": 0.9366272176056055, "gen_val/rouge1": 0.9708670633474195, "gen_val/rouge2": 0.9542004447299458, "gen_val/rougeL": 0.9708336765952826, "gen_val_full/n": 128.0, "gen_val_full/seq_acc": 0.0625, "gen_val_full/bit_acc": 0.3532414452475822, "gen_val_full/lev": 573.65625, "gen_val_full/lev_norm": 0.6462366040228255, "gen_val_full/token_acc": 0.03648509763617677, "gen_val_full/bleu": 0.10978491718305944, "gen_val_full/rouge1": 0.33893754392655, "gen_val_full/rouge2": 0.23012860870149393, "gen_val_full/rougeL": 0.2863141438838297 } ] }