Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +87 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -1972,3 +1972,90 @@
|
|
| 1972 |
{"event": "train", "epoch": 1, "step": 39400, "loss": 2.2444751858711243, "perplexity": 9.435462392852674, "learning_rate": 0.00017902040304676077, "grad_norm": 0.5323309302330017, "tokens_per_sec": 133292.49067217013, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1973 |
{"event": "train", "epoch": 1, "step": 39420, "loss": 1.914375126361847, "perplexity": 6.7826991426233905, "learning_rate": 0.00017891759650849894, "grad_norm": 0.5142626166343689, "tokens_per_sec": 134376.38136432564, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1974 |
{"event": "train", "epoch": 1, "step": 39440, "loss": 1.9836176335811615, "perplexity": 7.2689920264831756, "learning_rate": 0.00017881477586016503, "grad_norm": 0.5103498101234436, "tokens_per_sec": 134252.07765268066, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1972 |
{"event": "train", "epoch": 1, "step": 39400, "loss": 2.2444751858711243, "perplexity": 9.435462392852674, "learning_rate": 0.00017902040304676077, "grad_norm": 0.5323309302330017, "tokens_per_sec": 133292.49067217013, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1973 |
{"event": "train", "epoch": 1, "step": 39420, "loss": 1.914375126361847, "perplexity": 6.7826991426233905, "learning_rate": 0.00017891759650849894, "grad_norm": 0.5142626166343689, "tokens_per_sec": 134376.38136432564, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1974 |
{"event": "train", "epoch": 1, "step": 39440, "loss": 1.9836176335811615, "perplexity": 7.2689920264831756, "learning_rate": 0.00017881477586016503, "grad_norm": 0.5103498101234436, "tokens_per_sec": 134252.07765268066, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1975 |
+
{"event": "train", "epoch": 1, "step": 39460, "loss": 2.1538383066654205, "perplexity": 8.617873036007003, "learning_rate": 0.00017871194115192938, "grad_norm": 0.5693441033363342, "tokens_per_sec": 133385.63091961475, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1976 |
+
{"event": "train", "epoch": 1, "step": 39480, "loss": 2.621026247739792, "perplexity": 13.749827075961539, "learning_rate": 0.00017860909243396918, "grad_norm": 0.6121850609779358, "tokens_per_sec": 134559.51442286512, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1977 |
+
{"event": "train", "epoch": 1, "step": 39500, "loss": 2.697242021560669, "perplexity": 14.838750284533958, "learning_rate": 0.00017850622975646848, "grad_norm": 0.7053834795951843, "tokens_per_sec": 134020.23138948562, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1978 |
+
{"event": "train", "epoch": 1, "step": 39520, "loss": 2.4551562666893005, "perplexity": 11.648253640698467, "learning_rate": 0.0001784033531696183, "grad_norm": 0.6213880181312561, "tokens_per_sec": 132991.19591579225, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1979 |
+
{"event": "train", "epoch": 1, "step": 39540, "loss": 2.559121787548065, "perplexity": 12.924461906661179, "learning_rate": 0.0001783004627236162, "grad_norm": 0.8141724467277527, "tokens_per_sec": 134405.9488306867, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1980 |
+
{"event": "train", "epoch": 1, "step": 39560, "loss": 2.0048362612724304, "perplexity": 7.42487805713509, "learning_rate": 0.00017819755846866666, "grad_norm": 0.5248403549194336, "tokens_per_sec": 133897.62898071797, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1981 |
+
{"event": "train", "epoch": 1, "step": 39580, "loss": 1.7294455468654633, "perplexity": 5.637527297047794, "learning_rate": 0.00017809464045498094, "grad_norm": 0.4810648560523987, "tokens_per_sec": 133605.02371157272, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1982 |
+
{"event": "train", "epoch": 1, "step": 39600, "loss": 2.2356386184692383, "perplexity": 9.35245259444006, "learning_rate": 0.00017799170873277682, "grad_norm": 0.5187187194824219, "tokens_per_sec": 132975.49790918044, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1983 |
+
{"event": "train", "epoch": 1, "step": 39620, "loss": 2.1090598106384277, "perplexity": 8.240490020318276, "learning_rate": 0.0001778887633522789, "grad_norm": 0.5435354709625244, "tokens_per_sec": 134243.42303018938, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1984 |
+
{"event": "train", "epoch": 1, "step": 39640, "loss": 1.9756657779216766, "perplexity": 7.211419259634404, "learning_rate": 0.00017778580436371834, "grad_norm": 0.56279456615448, "tokens_per_sec": 133823.96666059727, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1985 |
+
{"event": "train", "epoch": 1, "step": 39660, "loss": 1.7117465734481812, "perplexity": 5.538626652949153, "learning_rate": 0.00017768283181733308, "grad_norm": 0.5523565411567688, "tokens_per_sec": 134041.4059552522, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1986 |
+
{"event": "train", "epoch": 1, "step": 39680, "loss": 2.202930122613907, "perplexity": 9.051496676109192, "learning_rate": 0.00017757984576336765, "grad_norm": 0.5711151361465454, "tokens_per_sec": 133501.4613703431, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1987 |
+
{"event": "train", "epoch": 1, "step": 39700, "loss": 1.9654100239276886, "perplexity": 7.137838674883754, "learning_rate": 0.00017747684625207314, "grad_norm": 0.5301913619041443, "tokens_per_sec": 134242.76742243915, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1988 |
+
{"event": "train", "epoch": 1, "step": 39720, "loss": 4.395219087600708, "perplexity": 81.0623885816195, "learning_rate": 0.00017737383333370723, "grad_norm": 3.2519636154174805, "tokens_per_sec": 133329.60181447605, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1989 |
+
{"event": "train", "epoch": 1, "step": 39740, "loss": 2.3975889086723328, "perplexity": 10.996630510785248, "learning_rate": 0.00017727080705853409, "grad_norm": 0.5990543961524963, "tokens_per_sec": 134523.55890344456, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1990 |
+
{"event": "train", "epoch": 1, "step": 39760, "loss": 2.108452796936035, "perplexity": 8.23548944782307, "learning_rate": 0.00017716776747682435, "grad_norm": 0.5635353922843933, "tokens_per_sec": 130239.32366007126, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1991 |
+
{"event": "train", "epoch": 1, "step": 39780, "loss": 3.5083101987838745, "perplexity": 33.3917945880418, "learning_rate": 0.0001770647146388554, "grad_norm": 1.089340090751648, "tokens_per_sec": 134597.99362064895, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1992 |
+
{"event": "train", "epoch": 1, "step": 39800, "loss": 1.9961585104465485, "perplexity": 7.360725567676231, "learning_rate": 0.00017696164859491072, "grad_norm": 0.5118879079818726, "tokens_per_sec": 134082.07465103664, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1993 |
+
{"event": "train", "epoch": 1, "step": 39820, "loss": 2.348485827445984, "perplexity": 10.469704777002532, "learning_rate": 0.00017685856939528044, "grad_norm": 0.560077428817749, "tokens_per_sec": 135172.11706875695, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1994 |
+
{"event": "train", "epoch": 1, "step": 39840, "loss": 1.8103930056095123, "perplexity": 6.11284934429965, "learning_rate": 0.00017675547709026113, "grad_norm": 0.48523783683776855, "tokens_per_sec": 134009.51594934042, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1995 |
+
{"event": "train", "epoch": 1, "step": 39860, "loss": 2.7330020666122437, "perplexity": 15.378986530377295, "learning_rate": 0.00017665237173015584, "grad_norm": 0.6957278251647949, "tokens_per_sec": 135058.1484314504, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1996 |
+
{"event": "train", "epoch": 1, "step": 39880, "loss": 2.4907389879226685, "perplexity": 12.070192551991536, "learning_rate": 0.00017654925336527366, "grad_norm": 0.6570205688476562, "tokens_per_sec": 134135.98888177518, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1997 |
+
{"event": "train", "epoch": 1, "step": 39900, "loss": 2.832532823085785, "perplexity": 16.98843507903461, "learning_rate": 0.00017644612204593043, "grad_norm": 0.7613231539726257, "tokens_per_sec": 134698.51224046768, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1998 |
+
{"event": "train", "epoch": 1, "step": 39920, "loss": 1.9618513882160187, "perplexity": 7.112482850101563, "learning_rate": 0.00017634297782244803, "grad_norm": 0.5872564911842346, "tokens_per_sec": 134753.31980822212, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1999 |
+
{"event": "train", "epoch": 1, "step": 39940, "loss": 2.3295917212963104, "perplexity": 10.273746124903916, "learning_rate": 0.00017623982074515472, "grad_norm": 0.5558028221130371, "tokens_per_sec": 135115.2415478599, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2000 |
+
{"event": "train", "epoch": 1, "step": 39960, "loss": 2.670915186405182, "perplexity": 14.453190505359325, "learning_rate": 0.00017613665086438506, "grad_norm": 0.8167163729667664, "tokens_per_sec": 134743.14732349876, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2001 |
+
{"event": "train", "epoch": 1, "step": 39980, "loss": 2.320322632789612, "perplexity": 10.178957841920568, "learning_rate": 0.00017603346823047986, "grad_norm": 0.5435994863510132, "tokens_per_sec": 133911.06642698118, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2002 |
+
{"event": "train", "epoch": 1, "step": 40000, "loss": 2.1369487941265106, "perplexity": 8.473543618938255, "learning_rate": 0.00017593027289378602, "grad_norm": 0.5455033779144287, "tokens_per_sec": 134364.4266611398, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2003 |
+
{"event": "train", "epoch": 1, "step": 40020, "loss": 2.5812657177448273, "perplexity": 13.21385258725034, "learning_rate": 0.00017582706490465683, "grad_norm": 0.6892513632774353, "tokens_per_sec": 132214.49759936123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2004 |
+
{"event": "train", "epoch": 1, "step": 40040, "loss": 2.345073491334915, "perplexity": 10.434039510864816, "learning_rate": 0.00017572384431345167, "grad_norm": 0.672498881816864, "tokens_per_sec": 130883.02285708844, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2005 |
+
{"event": "train", "epoch": 1, "step": 40060, "loss": 1.9493986368179321, "perplexity": 7.024462057324469, "learning_rate": 0.00017562061117053608, "grad_norm": 0.5878283977508545, "tokens_per_sec": 133527.27203765683, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2006 |
+
{"event": "train", "epoch": 1, "step": 40080, "loss": 2.234648436307907, "perplexity": 9.34319654605961, "learning_rate": 0.00017551736552628167, "grad_norm": 0.5532965064048767, "tokens_per_sec": 133234.8606781613, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2007 |
+
{"event": "train", "epoch": 1, "step": 40100, "loss": 1.9051883816719055, "perplexity": 6.720673560113551, "learning_rate": 0.00017541410743106617, "grad_norm": 0.544782817363739, "tokens_per_sec": 133828.00620848272, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2008 |
+
{"event": "train", "epoch": 1, "step": 40120, "loss": 2.0316661596298218, "perplexity": 7.626783216647653, "learning_rate": 0.00017531083693527345, "grad_norm": 0.5120164752006531, "tokens_per_sec": 133722.92705214786, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2009 |
+
{"event": "train", "epoch": 1, "step": 40140, "loss": 1.8455587923526764, "perplexity": 6.3316368723270475, "learning_rate": 0.00017520755408929336, "grad_norm": 0.5475971102714539, "tokens_per_sec": 134665.12130339272, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2010 |
+
{"event": "train", "epoch": 1, "step": 40160, "loss": 2.4590063095092773, "perplexity": 11.69318635693138, "learning_rate": 0.0001751042589435219, "grad_norm": 0.5325367450714111, "tokens_per_sec": 134087.96123283033, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2011 |
+
{"event": "train", "epoch": 1, "step": 40180, "loss": 1.7954696714878082, "perplexity": 6.022302561481515, "learning_rate": 0.0001750009515483608, "grad_norm": 0.5745704174041748, "tokens_per_sec": 134539.75626426144, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2012 |
+
{"event": "train", "epoch": 1, "step": 40200, "loss": 1.956707239151001, "perplexity": 7.075989123159458, "learning_rate": 0.000174897631954218, "grad_norm": 0.6354648470878601, "tokens_per_sec": 133973.98431564448, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2013 |
+
{"event": "train", "epoch": 1, "step": 40220, "loss": 2.1200332939624786, "perplexity": 8.331414868884146, "learning_rate": 0.00017479430021150733, "grad_norm": 0.5996856689453125, "tokens_per_sec": 134646.12357677062, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2014 |
+
{"event": "train", "epoch": 1, "step": 40240, "loss": 1.8381826281547546, "perplexity": 6.285105501602817, "learning_rate": 0.00017469095637064858, "grad_norm": 0.5607914924621582, "tokens_per_sec": 133827.22434147235, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2015 |
+
{"event": "train", "epoch": 1, "step": 40260, "loss": 2.1328941583633423, "perplexity": 8.439256044726623, "learning_rate": 0.0001745876004820675, "grad_norm": 0.541839599609375, "tokens_per_sec": 134838.32666560056, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2016 |
+
{"event": "train", "epoch": 1, "step": 40280, "loss": 1.7630846798419952, "perplexity": 5.830394584718528, "learning_rate": 0.0001744842325961956, "grad_norm": 0.5890209078788757, "tokens_per_sec": 134695.8720379472, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2017 |
+
{"event": "train", "epoch": 1, "step": 40300, "loss": 2.6755820512771606, "perplexity": 14.520799230100376, "learning_rate": 0.00017438085276347016, "grad_norm": 0.8728213310241699, "tokens_per_sec": 135323.308861263, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2018 |
+
{"event": "train", "epoch": 1, "step": 40320, "loss": 1.9929937422275543, "perplexity": 7.337467400098048, "learning_rate": 0.00017427746103433454, "grad_norm": 0.5283652544021606, "tokens_per_sec": 134887.68774597268, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2019 |
+
{"event": "train", "epoch": 1, "step": 40340, "loss": 2.232502728700638, "perplexity": 9.323170271103294, "learning_rate": 0.00017417405745923773, "grad_norm": 0.5501741766929626, "tokens_per_sec": 134333.6957607935, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2020 |
+
{"event": "train", "epoch": 1, "step": 40360, "loss": 2.0587914884090424, "perplexity": 7.836493592073504, "learning_rate": 0.0001740706420886346, "grad_norm": 0.5259140729904175, "tokens_per_sec": 133557.37546875168, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2021 |
+
{"event": "train", "epoch": 1, "step": 40380, "loss": 2.6331884264945984, "perplexity": 13.918075993488726, "learning_rate": 0.0001739672149729855, "grad_norm": 0.7967926263809204, "tokens_per_sec": 134641.77068599837, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2022 |
+
{"event": "train", "epoch": 1, "step": 40400, "loss": 1.9424088299274445, "perplexity": 6.975533623396032, "learning_rate": 0.00017386377616275704, "grad_norm": 0.5685714483261108, "tokens_per_sec": 133744.5283649743, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2023 |
+
{"event": "train", "epoch": 1, "step": 40420, "loss": 2.0360927879810333, "perplexity": 7.660618985398548, "learning_rate": 0.00017376032570842092, "grad_norm": 0.6293671727180481, "tokens_per_sec": 134407.65757080563, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2024 |
+
{"event": "train", "epoch": 1, "step": 40440, "loss": 1.9713495671749115, "perplexity": 7.180360330985481, "learning_rate": 0.000173656863660455, "grad_norm": 0.5294460654258728, "tokens_per_sec": 134598.78451984673, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2025 |
+
{"event": "train", "epoch": 1, "step": 40460, "loss": 2.523072600364685, "perplexity": 12.466843502754285, "learning_rate": 0.00017355339006934267, "grad_norm": 0.6271876692771912, "tokens_per_sec": 135296.26649944825, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2026 |
+
{"event": "train", "epoch": 1, "step": 40480, "loss": 1.8661071956157684, "perplexity": 6.463087828468934, "learning_rate": 0.0001734499049855728, "grad_norm": 0.5844515562057495, "tokens_per_sec": 133454.92363667436, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2027 |
+
{"event": "train", "epoch": 1, "step": 40500, "loss": 2.6467111706733704, "perplexity": 14.10756489200003, "learning_rate": 0.00017334640845964, "grad_norm": 1.2352241277694702, "tokens_per_sec": 134796.4047532179, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2028 |
+
{"event": "train", "epoch": 1, "step": 40520, "loss": 3.1762687265872955, "perplexity": 23.95719572614617, "learning_rate": 0.00017324290054204438, "grad_norm": 1.0158488750457764, "tokens_per_sec": 135265.24104835297, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2029 |
+
{"event": "train", "epoch": 1, "step": 40540, "loss": 2.0808333456516266, "perplexity": 8.011142183843544, "learning_rate": 0.0001731393812832918, "grad_norm": 0.693794846534729, "tokens_per_sec": 135148.85610529577, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2030 |
+
{"event": "train", "epoch": 1, "step": 40560, "loss": 2.1722927391529083, "perplexity": 8.778387538705354, "learning_rate": 0.00017303585073389334, "grad_norm": 0.649462878704071, "tokens_per_sec": 134353.91878877222, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2031 |
+
{"event": "train", "epoch": 1, "step": 40580, "loss": 2.9344632029533386, "perplexity": 18.81140251912093, "learning_rate": 0.00017293230894436598, "grad_norm": 0.6591626405715942, "tokens_per_sec": 133252.42866810545, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2032 |
+
{"event": "train", "epoch": 1, "step": 40600, "loss": 2.2000539898872375, "perplexity": 9.025500772049051, "learning_rate": 0.00017282875596523185, "grad_norm": 0.6318605542182922, "tokens_per_sec": 134736.1463757034, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2033 |
+
{"event": "train", "epoch": 1, "step": 40620, "loss": 2.7567859292030334, "perplexity": 15.749142655254865, "learning_rate": 0.00017272519184701862, "grad_norm": 0.8493112921714783, "tokens_per_sec": 134590.74414441586, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2034 |
+
{"event": "train", "epoch": 1, "step": 40640, "loss": 2.0556384921073914, "perplexity": 7.811824068655269, "learning_rate": 0.00017262161664025955, "grad_norm": 0.5299574732780457, "tokens_per_sec": 134684.25637633482, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2035 |
+
{"event": "train", "epoch": 1, "step": 40660, "loss": 2.3244916200637817, "perplexity": 10.22148236814089, "learning_rate": 0.00017251803039549315, "grad_norm": 0.622124433517456, "tokens_per_sec": 135243.94425671353, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2036 |
+
{"event": "train", "epoch": 1, "step": 40680, "loss": 2.7168303728103638, "perplexity": 15.13228245714711, "learning_rate": 0.0001724144331632634, "grad_norm": 0.8342093229293823, "tokens_per_sec": 135290.67287738007, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2037 |
+
{"event": "train", "epoch": 1, "step": 40700, "loss": 2.304787129163742, "perplexity": 10.022044624319216, "learning_rate": 0.0001723108249941195, "grad_norm": 0.6029829382896423, "tokens_per_sec": 134751.60226796987, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2038 |
+
{"event": "train", "epoch": 1, "step": 40720, "loss": 2.317167818546295, "perplexity": 10.146895722346967, "learning_rate": 0.00017220720593861622, "grad_norm": 0.6799089312553406, "tokens_per_sec": 134005.98806963215, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2039 |
+
{"event": "train", "epoch": 1, "step": 40740, "loss": 1.951445758342743, "perplexity": 7.038856713580587, "learning_rate": 0.00017210357604731337, "grad_norm": 0.568684458732605, "tokens_per_sec": 134558.46050790724, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2040 |
+
{"event": "train", "epoch": 1, "step": 40760, "loss": 2.566755473613739, "perplexity": 13.023500726110749, "learning_rate": 0.00017199993537077619, "grad_norm": 0.6405164003372192, "tokens_per_sec": 134346.82690397804, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2041 |
+
{"event": "train", "epoch": 1, "step": 40780, "loss": 2.0553934276103973, "perplexity": 7.809909902475944, "learning_rate": 0.00017189628395957525, "grad_norm": 0.6333890557289124, "tokens_per_sec": 134805.52749243527, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2042 |
+
{"event": "train", "epoch": 1, "step": 40800, "loss": 1.9097800850868225, "perplexity": 6.75160385688087, "learning_rate": 0.00017179262186428637, "grad_norm": 0.5627815127372742, "tokens_per_sec": 134426.19362057565, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2043 |
+
{"event": "train", "epoch": 1, "step": 40820, "loss": 2.075845181941986, "perplexity": 7.971280795346899, "learning_rate": 0.00017168894913549047, "grad_norm": 0.5741996765136719, "tokens_per_sec": 134269.3902458563, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2044 |
+
{"event": "train", "epoch": 1, "step": 40840, "loss": 1.8226849734783173, "perplexity": 6.188451992776024, "learning_rate": 0.0001715852658237737, "grad_norm": 0.511803925037384, "tokens_per_sec": 134564.78424530453, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2045 |
+
{"event": "train", "epoch": 1, "step": 40860, "loss": 2.266589730978012, "perplexity": 9.646447674797127, "learning_rate": 0.0001714815719797273, "grad_norm": 0.615704357624054, "tokens_per_sec": 133899.1943696441, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2046 |
+
{"event": "train", "epoch": 1, "step": 40880, "loss": 2.12179733812809, "perplexity": 8.346124823364978, "learning_rate": 0.00017137786765394778, "grad_norm": 0.5366615653038025, "tokens_per_sec": 133415.4114922177, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2047 |
+
{"event": "train", "epoch": 1, "step": 40900, "loss": 1.9619366526603699, "perplexity": 7.113089317854436, "learning_rate": 0.00017127415289703665, "grad_norm": 0.5448105335235596, "tokens_per_sec": 133747.00124950855, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2048 |
+
{"event": "train", "epoch": 1, "step": 40920, "loss": 2.5281534492969513, "perplexity": 12.530346840074161, "learning_rate": 0.00017117042775960056, "grad_norm": 0.7288480997085571, "tokens_per_sec": 133455.57157172708, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2049 |
+
{"event": "train", "epoch": 1, "step": 40940, "loss": 2.1475529074668884, "perplexity": 8.563876137532397, "learning_rate": 0.00017106669229225127, "grad_norm": 0.6251428723335266, "tokens_per_sec": 135022.0586226545, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2050 |
+
{"event": "train", "epoch": 1, "step": 40960, "loss": 2.557748556137085, "perplexity": 12.906725810270446, "learning_rate": 0.00017096294654560547, "grad_norm": 0.6042811870574951, "tokens_per_sec": 134071.08774777586, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2051 |
+
{"event": "train", "epoch": 1, "step": 40980, "loss": 2.7087680101394653, "perplexity": 15.010771000857227, "learning_rate": 0.00017085919057028501, "grad_norm": 0.7859557271003723, "tokens_per_sec": 134349.71609999667, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2052 |
+
{"event": "train", "epoch": 1, "step": 41000, "loss": 1.9979293644428253, "perplexity": 7.373771886116466, "learning_rate": 0.00017075542441691654, "grad_norm": 0.5495272278785706, "tokens_per_sec": 133629.57517255595, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2053 |
+
{"event": "train", "epoch": 1, "step": 41020, "loss": 2.499220609664917, "perplexity": 12.173002741815882, "learning_rate": 0.000170651648136132, "grad_norm": 0.795656681060791, "tokens_per_sec": 134005.46543658385, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2054 |
+
{"event": "train", "epoch": 1, "step": 41040, "loss": 2.4765357971191406, "perplexity": 11.899969024360033, "learning_rate": 0.0001705478617785678, "grad_norm": 0.5982843041419983, "tokens_per_sec": 134813.85802846964, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2055 |
+
{"event": "train", "epoch": 1, "step": 41060, "loss": 2.2284310162067413, "perplexity": 9.28528618112847, "learning_rate": 0.00017044406539486572, "grad_norm": 0.5942403078079224, "tokens_per_sec": 134161.91454726842, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2056 |
+
{"event": "train", "epoch": 1, "step": 41080, "loss": 2.089510530233383, "perplexity": 8.080958811103098, "learning_rate": 0.00017034025903567222, "grad_norm": 0.5203970074653625, "tokens_per_sec": 134343.0186081746, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2057 |
+
{"event": "train", "epoch": 1, "step": 41100, "loss": 2.129141539335251, "perplexity": 8.4076460790842, "learning_rate": 0.0001702364427516385, "grad_norm": 0.69418865442276, "tokens_per_sec": 134152.48587308126, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2058 |
+
{"event": "train", "epoch": 1, "step": 41120, "loss": 2.1212494671344757, "perplexity": 8.34155347603382, "learning_rate": 0.00017013261659342094, "grad_norm": 0.5510379672050476, "tokens_per_sec": 134069.7799030367, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2059 |
+
{"event": "train", "epoch": 1, "step": 41140, "loss": 2.4040462374687195, "perplexity": 11.067869127695182, "learning_rate": 0.00017002878061168038, "grad_norm": 0.9795651435852051, "tokens_per_sec": 134199.51146475234, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2060 |
+
{"event": "train", "epoch": 1, "step": 41160, "loss": 2.5507077276706696, "perplexity": 12.816170932549133, "learning_rate": 0.0001699249348570827, "grad_norm": 0.7334607243537903, "tokens_per_sec": 134571.10857711313, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2061 |
+
{"event": "train", "epoch": 1, "step": 41180, "loss": 2.0979593992233276, "perplexity": 8.149523010518797, "learning_rate": 0.0001698210793802984, "grad_norm": 0.6085432171821594, "tokens_per_sec": 134069.9106863624, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|