Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +86 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -1553,3 +1553,89 @@
|
|
| 1553 |
{"event": "train", "epoch": 1, "step": 31020, "loss": 2.2266247868537903, "perplexity": 9.26852996202051, "learning_rate": 0.00022024690081966393, "grad_norm": 0.5391517877578735, "tokens_per_sec": 132845.68119886794, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1554 |
{"event": "train", "epoch": 1, "step": 31040, "loss": 2.329028010368347, "perplexity": 10.267956333979713, "learning_rate": 0.00022015430474642213, "grad_norm": 0.5895096063613892, "tokens_per_sec": 134217.596928128, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1555 |
{"event": "train", "epoch": 1, "step": 31060, "loss": 2.0381290912628174, "perplexity": 7.676234222262293, "learning_rate": 0.00022006167444204016, "grad_norm": 0.5365936160087585, "tokens_per_sec": 134469.858985598, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1553 |
{"event": "train", "epoch": 1, "step": 31020, "loss": 2.2266247868537903, "perplexity": 9.26852996202051, "learning_rate": 0.00022024690081966393, "grad_norm": 0.5391517877578735, "tokens_per_sec": 132845.68119886794, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1554 |
{"event": "train", "epoch": 1, "step": 31040, "loss": 2.329028010368347, "perplexity": 10.267956333979713, "learning_rate": 0.00022015430474642213, "grad_norm": 0.5895096063613892, "tokens_per_sec": 134217.596928128, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1555 |
{"event": "train", "epoch": 1, "step": 31060, "loss": 2.0381290912628174, "perplexity": 7.676234222262293, "learning_rate": 0.00022006167444204016, "grad_norm": 0.5365936160087585, "tokens_per_sec": 134469.858985598, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1556 |
+
{"event": "train", "epoch": 1, "step": 31080, "loss": 2.7025020122528076, "perplexity": 14.91700760888043, "learning_rate": 0.0002199690099517161, "grad_norm": 0.5405033230781555, "tokens_per_sec": 134812.8001255542, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1557 |
+
{"event": "train", "epoch": 1, "step": 31100, "loss": 2.1258639097213745, "perplexity": 8.380134041077474, "learning_rate": 0.0002198763113206647, "grad_norm": 0.49189549684524536, "tokens_per_sec": 134287.7569670121, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1558 |
+
{"event": "train", "epoch": 1, "step": 31120, "loss": 2.728902667760849, "perplexity": 15.316070976741544, "learning_rate": 0.00021978357859411753, "grad_norm": 0.7531503438949585, "tokens_per_sec": 134275.55565629338, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1559 |
+
{"event": "train", "epoch": 1, "step": 31140, "loss": 2.558133453130722, "perplexity": 12.911694526393125, "learning_rate": 0.0002196908118173225, "grad_norm": 1.0696262121200562, "tokens_per_sec": 133207.2263066891, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1560 |
+
{"event": "train", "epoch": 1, "step": 31160, "loss": 2.412197768688202, "perplexity": 11.158457925513765, "learning_rate": 0.00021959801103554427, "grad_norm": 0.5453084707260132, "tokens_per_sec": 134130.88349242628, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1561 |
+
{"event": "train", "epoch": 1, "step": 31180, "loss": 2.2701981961727142, "perplexity": 9.681319424392688, "learning_rate": 0.0002195051762940642, "grad_norm": 0.7260033488273621, "tokens_per_sec": 133667.52427459238, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1562 |
+
{"event": "train", "epoch": 1, "step": 31200, "loss": 3.2202088236808777, "perplexity": 25.03334719126226, "learning_rate": 0.00021941230763818017, "grad_norm": 0.8229443430900574, "tokens_per_sec": 128847.89146094972, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1563 |
+
{"event": "train", "epoch": 1, "step": 31220, "loss": 2.6923938393592834, "perplexity": 14.766983429480835, "learning_rate": 0.00021931940511320654, "grad_norm": 0.6730867624282837, "tokens_per_sec": 132864.43110891987, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1564 |
+
{"event": "train", "epoch": 1, "step": 31240, "loss": 2.9324578642845154, "perplexity": 18.773717084892752, "learning_rate": 0.00021922646876447425, "grad_norm": 0.793106734752655, "tokens_per_sec": 135432.25376792072, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1565 |
+
{"event": "train", "epoch": 1, "step": 31260, "loss": 2.528547525405884, "perplexity": 12.535285723484535, "learning_rate": 0.00021913349863733064, "grad_norm": 0.6114016771316528, "tokens_per_sec": 134164.00998818834, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1566 |
+
{"event": "train", "epoch": 1, "step": 31280, "loss": 2.156359761953354, "perplexity": 8.6396300356682, "learning_rate": 0.00021904049477713969, "grad_norm": 0.4857799708843231, "tokens_per_sec": 133723.9679192357, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1567 |
+
{"event": "train", "epoch": 1, "step": 31300, "loss": 1.7400385737419128, "perplexity": 5.697563194765455, "learning_rate": 0.0002189474572292818, "grad_norm": 0.7459017634391785, "tokens_per_sec": 134366.52843285436, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1568 |
+
{"event": "train", "epoch": 1, "step": 31320, "loss": 1.9807321429252625, "perplexity": 7.248047649846496, "learning_rate": 0.00021885438603915378, "grad_norm": 0.4723159670829773, "tokens_per_sec": 134533.56656838907, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1569 |
+
{"event": "train", "epoch": 1, "step": 31340, "loss": 2.329988867044449, "perplexity": 10.277827109814138, "learning_rate": 0.00021876128125216873, "grad_norm": 0.510909378528595, "tokens_per_sec": 134812.93236251065, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1570 |
+
{"event": "train", "epoch": 1, "step": 31360, "loss": 2.3005932569503784, "perplexity": 9.980101463453288, "learning_rate": 0.00021866814291375634, "grad_norm": 0.4973556101322174, "tokens_per_sec": 134602.8709804421, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1571 |
+
{"event": "train", "epoch": 1, "step": 31380, "loss": 2.02332603931427, "perplexity": 7.563439441522764, "learning_rate": 0.00021857497106936256, "grad_norm": 0.48008227348327637, "tokens_per_sec": 134607.61685898804, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1572 |
+
{"event": "train", "epoch": 1, "step": 31400, "loss": 2.2036523818969727, "perplexity": 9.058036565072609, "learning_rate": 0.00021848176576444987, "grad_norm": 0.5170644521713257, "tokens_per_sec": 135375.55919869352, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1573 |
+
{"event": "train", "epoch": 1, "step": 31420, "loss": 2.1573258340358734, "perplexity": 8.647980574010479, "learning_rate": 0.0002183885270444967, "grad_norm": 0.4851105213165283, "tokens_per_sec": 134054.4800150598, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1574 |
+
{"event": "train", "epoch": 1, "step": 31440, "loss": 2.474100559949875, "perplexity": 11.871025034522809, "learning_rate": 0.00021829525495499816, "grad_norm": 1.4702283143997192, "tokens_per_sec": 135234.4959534626, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1575 |
+
{"event": "train", "epoch": 1, "step": 31460, "loss": 2.2097761631011963, "perplexity": 9.113676187697651, "learning_rate": 0.00021820194954146556, "grad_norm": 0.6317822337150574, "tokens_per_sec": 134031.73279267736, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1576 |
+
{"event": "train", "epoch": 1, "step": 31480, "loss": 2.0245066583156586, "perplexity": 7.572374255110802, "learning_rate": 0.00021810861084942636, "grad_norm": 0.46586278080940247, "tokens_per_sec": 134874.18595193795, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1577 |
+
{"event": "train", "epoch": 1, "step": 31500, "loss": 2.643142968416214, "perplexity": 14.057315949618626, "learning_rate": 0.0002180152389244242, "grad_norm": 0.6866251230239868, "tokens_per_sec": 133720.06475120425, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1578 |
+
{"event": "train", "epoch": 1, "step": 31520, "loss": 2.2245808243751526, "perplexity": 9.249604782313396, "learning_rate": 0.00021792183381201912, "grad_norm": 0.5222992897033691, "tokens_per_sec": 134943.17936955998, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1579 |
+
{"event": "train", "epoch": 1, "step": 31540, "loss": 1.900541067123413, "perplexity": 6.689512938789068, "learning_rate": 0.00021782839555778722, "grad_norm": 0.45487454533576965, "tokens_per_sec": 133714.3405169017, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1580 |
+
{"event": "train", "epoch": 1, "step": 31560, "loss": 2.2325634956359863, "perplexity": 9.323736828802215, "learning_rate": 0.00021773492420732083, "grad_norm": 0.493905633687973, "tokens_per_sec": 133658.03497852743, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1581 |
+
{"event": "train", "epoch": 1, "step": 31580, "loss": 2.0107991695404053, "perplexity": 7.469284187154304, "learning_rate": 0.00021764141980622857, "grad_norm": 0.6083336472511292, "tokens_per_sec": 132021.57995992445, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1582 |
+
{"event": "train", "epoch": 1, "step": 31600, "loss": 2.4874267578125, "perplexity": 12.030279433991037, "learning_rate": 0.00021754788240013496, "grad_norm": 0.6397104263305664, "tokens_per_sec": 134227.82130296907, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1583 |
+
{"event": "train", "epoch": 1, "step": 31620, "loss": 2.413846343755722, "perplexity": 11.176868652604847, "learning_rate": 0.00021745431203468082, "grad_norm": 0.5265084505081177, "tokens_per_sec": 134637.15429120013, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1584 |
+
{"event": "train", "epoch": 1, "step": 31640, "loss": 2.8753213584423065, "perplexity": 17.731121251507926, "learning_rate": 0.00021736070875552266, "grad_norm": 0.7959738373756409, "tokens_per_sec": 134678.4492967642, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1585 |
+
{"event": "train", "epoch": 1, "step": 31660, "loss": 2.577507197856903, "perplexity": 13.164281275241718, "learning_rate": 0.00021726707260833362, "grad_norm": 0.5227611064910889, "tokens_per_sec": 134303.76635031818, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1586 |
+
{"event": "train", "epoch": 1, "step": 31680, "loss": 2.264411985874176, "perplexity": 9.625463028502724, "learning_rate": 0.0002171734036388023, "grad_norm": 0.5522229671478271, "tokens_per_sec": 135357.42696020595, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1587 |
+
{"event": "train", "epoch": 1, "step": 31700, "loss": 1.775351107120514, "perplexity": 5.902353130763728, "learning_rate": 0.0002170797018926338, "grad_norm": 0.4616525173187256, "tokens_per_sec": 134648.76182937177, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1588 |
+
{"event": "train", "epoch": 1, "step": 31720, "loss": 2.52315616607666, "perplexity": 12.467885346938294, "learning_rate": 0.00021698596741554887, "grad_norm": 1.178956389427185, "tokens_per_sec": 134845.8674411075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1589 |
+
{"event": "train", "epoch": 1, "step": 31740, "loss": 1.9484655559062958, "perplexity": 7.017910722802182, "learning_rate": 0.00021689220025328468, "grad_norm": 0.49590057134628296, "tokens_per_sec": 133572.17264606894, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1590 |
+
{"event": "train", "epoch": 1, "step": 31760, "loss": 2.2267175912857056, "perplexity": 9.269390162592922, "learning_rate": 0.00021679840045159366, "grad_norm": 0.5290210247039795, "tokens_per_sec": 133560.7499749766, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1591 |
+
{"event": "train", "epoch": 1, "step": 31780, "loss": 2.0616602301597595, "perplexity": 7.859006745191791, "learning_rate": 0.0002167045680562448, "grad_norm": 0.5463327765464783, "tokens_per_sec": 134283.9520193064, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1592 |
+
{"event": "train", "epoch": 1, "step": 31800, "loss": 2.34723436832428, "perplexity": 10.456610564602537, "learning_rate": 0.0002166107031130227, "grad_norm": 0.6516744494438171, "tokens_per_sec": 134190.994637787, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1593 |
+
{"event": "train", "epoch": 1, "step": 31820, "loss": 1.86092147231102, "perplexity": 6.429658795068653, "learning_rate": 0.00021651680566772778, "grad_norm": 0.479316771030426, "tokens_per_sec": 132929.197517037, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1594 |
+
{"event": "train", "epoch": 1, "step": 31840, "loss": 1.9974473118782043, "perplexity": 7.370218197069109, "learning_rate": 0.00021642287576617653, "grad_norm": 0.5761274695396423, "tokens_per_sec": 134104.96981737035, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1595 |
+
{"event": "train", "epoch": 1, "step": 31860, "loss": 2.0479214191436768, "perplexity": 7.751771664754875, "learning_rate": 0.00021632891345420117, "grad_norm": 0.5520979762077332, "tokens_per_sec": 133791.65905451486, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1596 |
+
{"event": "train", "epoch": 1, "step": 31880, "loss": 1.8969351947307587, "perplexity": 6.665434846116843, "learning_rate": 0.0002162349187776496, "grad_norm": 0.4826981723308563, "tokens_per_sec": 133759.36704398697, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1597 |
+
{"event": "train", "epoch": 1, "step": 31900, "loss": 2.086067497730255, "perplexity": 8.053183650089426, "learning_rate": 0.00021614089178238576, "grad_norm": 0.4821702539920807, "tokens_per_sec": 133521.04543577795, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1598 |
+
{"event": "train", "epoch": 1, "step": 31920, "loss": 1.969160556793213, "perplexity": 7.164659638435617, "learning_rate": 0.00021604683251428937, "grad_norm": 0.4513254165649414, "tokens_per_sec": 133529.7368946716, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1599 |
+
{"event": "train", "epoch": 1, "step": 31940, "loss": 2.3141255378723145, "perplexity": 10.11607292716032, "learning_rate": 0.00021595274101925555, "grad_norm": 0.558356761932373, "tokens_per_sec": 134602.34368126112, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1600 |
+
{"event": "train", "epoch": 1, "step": 31960, "loss": 1.8379608392715454, "perplexity": 6.283711689644476, "learning_rate": 0.00021585861734319559, "grad_norm": 0.5965127944946289, "tokens_per_sec": 134250.76627451286, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1601 |
+
{"event": "train", "epoch": 1, "step": 31980, "loss": 2.641543924808502, "perplexity": 14.03485565069232, "learning_rate": 0.00021576446153203608, "grad_norm": 1.0550146102905273, "tokens_per_sec": 134314.66016197237, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1602 |
+
{"event": "train", "epoch": 1, "step": 32000, "loss": 2.0278612971305847, "perplexity": 7.597819491634651, "learning_rate": 0.00021567027363171952, "grad_norm": 0.49684175848960876, "tokens_per_sec": 134653.9067193242, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1603 |
+
{"event": "train", "epoch": 1, "step": 32020, "loss": 2.7926209568977356, "perplexity": 16.32374762236854, "learning_rate": 0.00021557605368820399, "grad_norm": 1.1447405815124512, "tokens_per_sec": 133366.73362742036, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1604 |
+
{"event": "train", "epoch": 1, "step": 32040, "loss": 2.025782734155655, "perplexity": 7.582043346884291, "learning_rate": 0.00021548180174746318, "grad_norm": 0.5397109389305115, "tokens_per_sec": 134722.67490065304, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1605 |
+
{"event": "train", "epoch": 1, "step": 32060, "loss": 2.1116747558116913, "perplexity": 8.262066648477093, "learning_rate": 0.0002153875178554867, "grad_norm": 0.5748896598815918, "tokens_per_sec": 133626.97670746612, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1606 |
+
{"event": "train", "epoch": 1, "step": 32080, "loss": 2.6278246641159058, "perplexity": 13.843622594762202, "learning_rate": 0.00021529320205827926, "grad_norm": 0.5559474229812622, "tokens_per_sec": 134916.1562659824, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1607 |
+
{"event": "train", "epoch": 1, "step": 32100, "loss": 2.0795572996139526, "perplexity": 8.000926117074599, "learning_rate": 0.00021519885440186157, "grad_norm": 0.5606538653373718, "tokens_per_sec": 134411.8638854361, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1608 |
+
{"event": "train", "epoch": 1, "step": 32120, "loss": 2.0070058405399323, "perplexity": 7.441004406002001, "learning_rate": 0.0002151044749322695, "grad_norm": 0.4800957441329956, "tokens_per_sec": 133811.71919965497, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1609 |
+
{"event": "train", "epoch": 1, "step": 32140, "loss": 2.688359409570694, "perplexity": 14.707527088507586, "learning_rate": 0.00021501006369555468, "grad_norm": 0.8397963643074036, "tokens_per_sec": 133752.46795035238, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1610 |
+
{"event": "train", "epoch": 1, "step": 32160, "loss": 2.1220155358314514, "perplexity": 8.347946127328349, "learning_rate": 0.00021491562073778418, "grad_norm": 0.4820932447910309, "tokens_per_sec": 133793.61270036583, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1611 |
+
{"event": "train", "epoch": 1, "step": 32180, "loss": 1.9968559741973877, "perplexity": 7.3658611976892585, "learning_rate": 0.00021482114610504066, "grad_norm": 0.48161372542381287, "tokens_per_sec": 133967.0630655921, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1612 |
+
{"event": "train", "epoch": 1, "step": 32200, "loss": 2.798991560935974, "perplexity": 16.428071705775317, "learning_rate": 0.0002147266398434221, "grad_norm": 0.8239628076553345, "tokens_per_sec": 133517.8026439524, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1613 |
+
{"event": "train", "epoch": 1, "step": 32220, "loss": 2.4227072298526764, "perplexity": 11.276345689129537, "learning_rate": 0.00021463210199904192, "grad_norm": 0.6190411448478699, "tokens_per_sec": 133785.40777117692, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1614 |
+
{"event": "train", "epoch": 1, "step": 32240, "loss": 1.8372160196304321, "perplexity": 6.279033200290602, "learning_rate": 0.00021453753261802905, "grad_norm": 0.6056377291679382, "tokens_per_sec": 134741.4300425482, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1615 |
+
{"event": "train", "epoch": 1, "step": 32260, "loss": 2.1018749475479126, "perplexity": 8.18149541558857, "learning_rate": 0.00021444293174652763, "grad_norm": 0.48140355944633484, "tokens_per_sec": 133078.3743787321, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1616 |
+
{"event": "train", "epoch": 1, "step": 32280, "loss": 2.3534489274024963, "perplexity": 10.521796128690916, "learning_rate": 0.00021434829943069723, "grad_norm": 0.5392767786979675, "tokens_per_sec": 135117.632544879, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1617 |
+
{"event": "train", "epoch": 1, "step": 32300, "loss": 2.3971485793590546, "perplexity": 10.991789437935422, "learning_rate": 0.000214253635716713, "grad_norm": 0.5787780284881592, "tokens_per_sec": 132826.29473474645, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1618 |
+
{"event": "train", "epoch": 1, "step": 32320, "loss": 1.8685628175735474, "perplexity": 6.478978231281577, "learning_rate": 0.0002141589406507652, "grad_norm": 0.4795730710029602, "tokens_per_sec": 134791.77774879147, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1619 |
+
{"event": "train", "epoch": 1, "step": 32340, "loss": 2.000901162624359, "perplexity": 7.395717841322098, "learning_rate": 0.00021406421427905926, "grad_norm": 0.5228570699691772, "tokens_per_sec": 134612.75860508622, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1620 |
+
{"event": "train", "epoch": 1, "step": 32360, "loss": 2.0544599890708923, "perplexity": 7.802623232941072, "learning_rate": 0.00021396945664781613, "grad_norm": 0.9582899212837219, "tokens_per_sec": 134542.65376441347, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1621 |
+
{"event": "train", "epoch": 1, "step": 32380, "loss": 2.295892894268036, "perplexity": 9.933301441670745, "learning_rate": 0.0002138746678032719, "grad_norm": 0.6664041876792908, "tokens_per_sec": 134335.79657120517, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1622 |
+
{"event": "train", "epoch": 1, "step": 32400, "loss": 2.2306176722049713, "perplexity": 9.305612122673736, "learning_rate": 0.0002137798477916779, "grad_norm": 0.5027684569358826, "tokens_per_sec": 134073.0495627277, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1623 |
+
{"event": "train", "epoch": 1, "step": 32420, "loss": 2.3001354932785034, "perplexity": 9.975533981055207, "learning_rate": 0.00021368499665930054, "grad_norm": 0.5002855062484741, "tokens_per_sec": 134669.343748469, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1624 |
+
{"event": "train", "epoch": 1, "step": 32440, "loss": 2.3225149512290955, "perplexity": 10.201297838137306, "learning_rate": 0.00021359011445242176, "grad_norm": 0.4954259991645813, "tokens_per_sec": 134050.94976757432, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1625 |
+
{"event": "train", "epoch": 1, "step": 32460, "loss": 2.2144112586975098, "perplexity": 9.156016999111207, "learning_rate": 0.00021349520121733833, "grad_norm": 0.6051504015922546, "tokens_per_sec": 133512.4848063306, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1626 |
+
{"event": "train", "epoch": 1, "step": 32480, "loss": 2.58212348818779, "perplexity": 13.225191902006609, "learning_rate": 0.0002134002570003623, "grad_norm": 0.5892339944839478, "tokens_per_sec": 133328.6964188012, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1627 |
+
{"event": "train", "epoch": 1, "step": 32500, "loss": 2.797270655632019, "perplexity": 16.399824862085776, "learning_rate": 0.0002133052818478209, "grad_norm": 1.2186973094940186, "tokens_per_sec": 134415.93900380444, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1628 |
+
{"event": "train", "epoch": 1, "step": 32520, "loss": 2.2123448252677917, "perplexity": 9.137116234812527, "learning_rate": 0.0002132102758060563, "grad_norm": 0.52647465467453, "tokens_per_sec": 134683.9924073085, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1629 |
+
{"event": "train", "epoch": 1, "step": 32540, "loss": 2.0299054384231567, "perplexity": 7.6133663928087625, "learning_rate": 0.00021311523892142582, "grad_norm": 0.4654105305671692, "tokens_per_sec": 134889.14398328398, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1630 |
+
{"event": "train", "epoch": 1, "step": 32560, "loss": 2.033451020717621, "perplexity": 7.6404081208997345, "learning_rate": 0.00021302017124030185, "grad_norm": 0.503710150718689, "tokens_per_sec": 134850.89509336808, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1631 |
+
{"event": "train", "epoch": 1, "step": 32580, "loss": 2.246149241924286, "perplexity": 9.451271114436892, "learning_rate": 0.00021292507280907172, "grad_norm": 0.5175694823265076, "tokens_per_sec": 134303.6351103489, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1632 |
+
{"event": "train", "epoch": 1, "step": 32600, "loss": 2.555019438266754, "perplexity": 12.871549855710063, "learning_rate": 0.00021282994367413785, "grad_norm": 0.6011718511581421, "tokens_per_sec": 133293.91266971003, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1633 |
+
{"event": "train", "epoch": 1, "step": 32620, "loss": 2.2244591116905212, "perplexity": 9.248479056592501, "learning_rate": 0.0002127347838819176, "grad_norm": 0.5381768941879272, "tokens_per_sec": 133723.5775921789, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1634 |
+
{"event": "train", "epoch": 1, "step": 32640, "loss": 2.32526695728302, "perplexity": 10.229410536961714, "learning_rate": 0.00021263959347884328, "grad_norm": 0.5974282622337341, "tokens_per_sec": 134659.5797460417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1635 |
+
{"event": "train", "epoch": 1, "step": 32660, "loss": 2.2705079913139343, "perplexity": 9.68431911473172, "learning_rate": 0.00021254437251136228, "grad_norm": 0.5026917457580566, "tokens_per_sec": 134189.55343555796, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1636 |
+
{"event": "train", "epoch": 1, "step": 32680, "loss": 1.9345754981040955, "perplexity": 6.921105409323982, "learning_rate": 0.0002124491210259367, "grad_norm": 0.5161557197570801, "tokens_per_sec": 134183.0029064744, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1637 |
+
{"event": "train", "epoch": 1, "step": 32700, "loss": 2.417285740375519, "perplexity": 11.215376520823858, "learning_rate": 0.0002123538390690435, "grad_norm": 0.6105350255966187, "tokens_per_sec": 133875.19576084628, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1638 |
+
{"event": "train", "epoch": 1, "step": 32720, "loss": 1.798217087984085, "perplexity": 6.038871084779604, "learning_rate": 0.00021225852668717486, "grad_norm": 0.5105339288711548, "tokens_per_sec": 134256.7988262209, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1639 |
+
{"event": "train", "epoch": 1, "step": 32740, "loss": 2.405452936887741, "perplexity": 11.083449248471176, "learning_rate": 0.0002121631839268375, "grad_norm": 0.5526533722877502, "tokens_per_sec": 132646.43767866967, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1640 |
+
{"event": "train", "epoch": 1, "step": 32760, "loss": 2.5538654923439026, "perplexity": 12.856705349759078, "learning_rate": 0.00021206781083455293, "grad_norm": 0.5606786608695984, "tokens_per_sec": 134402.26861491462, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1641 |
+
{"event": "train", "epoch": 1, "step": 32780, "loss": 2.039138823747635, "perplexity": 7.6839890798244985, "learning_rate": 0.00021197240745685772, "grad_norm": 0.47092658281326294, "tokens_per_sec": 134808.30422338485, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|