Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +81 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -517,3 +517,84 @@
|
|
| 517 |
{"event": "train", "epoch": 0, "step": 10320, "loss": 2.4276498556137085, "perplexity": 11.332218411054889, "learning_rate": 0.0002903612489096236, "grad_norm": 0.45870158076286316, "tokens_per_sec": 134896.02843987982, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 518 |
{"event": "train", "epoch": 0, "step": 10340, "loss": 2.2210040986537933, "perplexity": 9.216580577444036, "learning_rate": 0.00029032426051676886, "grad_norm": 0.44327396154403687, "tokens_per_sec": 134881.4659952638, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 519 |
{"event": "train", "epoch": 0, "step": 10360, "loss": 2.1260549426078796, "perplexity": 8.381735075192855, "learning_rate": 0.00029028720365399706, "grad_norm": 0.4135127365589142, "tokens_per_sec": 134051.99574744506, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 517 |
{"event": "train", "epoch": 0, "step": 10320, "loss": 2.4276498556137085, "perplexity": 11.332218411054889, "learning_rate": 0.0002903612489096236, "grad_norm": 0.45870158076286316, "tokens_per_sec": 134896.02843987982, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 518 |
{"event": "train", "epoch": 0, "step": 10340, "loss": 2.2210040986537933, "perplexity": 9.216580577444036, "learning_rate": 0.00029032426051676886, "grad_norm": 0.44327396154403687, "tokens_per_sec": 134881.4659952638, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 519 |
{"event": "train", "epoch": 0, "step": 10360, "loss": 2.1260549426078796, "perplexity": 8.381735075192855, "learning_rate": 0.00029028720365399706, "grad_norm": 0.4135127365589142, "tokens_per_sec": 134051.99574744506, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 520 |
+
{"event": "train", "epoch": 0, "step": 10380, "loss": 2.561093270778656, "perplexity": 12.949967400139798, "learning_rate": 0.00029025007833938954, "grad_norm": 0.6440810561180115, "tokens_per_sec": 134093.45584219313, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 521 |
+
{"event": "train", "epoch": 0, "step": 10400, "loss": 2.796332001686096, "perplexity": 16.384438324213498, "learning_rate": 0.00029021288459106157, "grad_norm": 0.5771433115005493, "tokens_per_sec": 134168.33204181664, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 522 |
+
{"event": "train", "epoch": 0, "step": 10420, "loss": 2.5249491333961487, "perplexity": 12.490259910344955, "learning_rate": 0.00029017562242716135, "grad_norm": 0.5094009637832642, "tokens_per_sec": 133876.36940574375, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 523 |
+
{"event": "train", "epoch": 0, "step": 10440, "loss": 2.3285775184631348, "perplexity": 10.263331744516451, "learning_rate": 0.00029013829186587035, "grad_norm": 0.4619823694229126, "tokens_per_sec": 133644.9081738525, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 524 |
+
{"event": "train", "epoch": 0, "step": 10460, "loss": 2.3066182732582092, "perplexity": 10.040413244810358, "learning_rate": 0.0002901008929254039, "grad_norm": 0.4666915237903595, "tokens_per_sec": 135320.37763967327, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 525 |
+
{"event": "train", "epoch": 0, "step": 10480, "loss": 2.7665576934814453, "perplexity": 15.90379394222807, "learning_rate": 0.0002900634256240104, "grad_norm": 0.7609860301017761, "tokens_per_sec": 133825.92125009006, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 526 |
+
{"event": "train", "epoch": 0, "step": 10500, "loss": 2.498764753341675, "perplexity": 12.167454866156353, "learning_rate": 0.0002900258899799717, "grad_norm": 0.6262858510017395, "tokens_per_sec": 134365.740260756, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 527 |
+
{"event": "train", "epoch": 0, "step": 10520, "loss": 2.349215090274811, "perplexity": 10.477342728223645, "learning_rate": 0.0002899882860116029, "grad_norm": 0.4293045401573181, "tokens_per_sec": 133245.5817681738, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 528 |
+
{"event": "train", "epoch": 0, "step": 10540, "loss": 2.5769025087356567, "perplexity": 13.156323383833218, "learning_rate": 0.0002899506137372526, "grad_norm": 0.9125115871429443, "tokens_per_sec": 134782.128161677, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 529 |
+
{"event": "train", "epoch": 0, "step": 10560, "loss": 2.6657527089118958, "perplexity": 14.378768500916776, "learning_rate": 0.00028991287317530287, "grad_norm": 0.4853087365627289, "tokens_per_sec": 134843.0891625321, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 530 |
+
{"event": "train", "epoch": 0, "step": 10580, "loss": 2.7172582149505615, "perplexity": 15.138758070430995, "learning_rate": 0.0002898750643441683, "grad_norm": 0.5937750339508057, "tokens_per_sec": 134466.04371158066, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 531 |
+
{"event": "train", "epoch": 0, "step": 10600, "loss": 2.3629326820373535, "perplexity": 10.622056934467151, "learning_rate": 0.00028983718726229783, "grad_norm": 0.47479692101478577, "tokens_per_sec": 134826.15736835857, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 532 |
+
{"event": "train", "epoch": 0, "step": 10620, "loss": 2.7237884402275085, "perplexity": 15.237941062222392, "learning_rate": 0.0002897992419481732, "grad_norm": 0.4846254587173462, "tokens_per_sec": 134450.91630797164, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 533 |
+
{"event": "train", "epoch": 0, "step": 10640, "loss": 3.0550296306610107, "perplexity": 21.221814333982568, "learning_rate": 0.00028976122842030907, "grad_norm": 0.7957623600959778, "tokens_per_sec": 134728.08958153005, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 534 |
+
{"event": "train", "epoch": 0, "step": 10660, "loss": 2.4579885601997375, "perplexity": 11.681291678517386, "learning_rate": 0.0002897231466972542, "grad_norm": 0.5164206027984619, "tokens_per_sec": 135353.02795813716, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 535 |
+
{"event": "train", "epoch": 0, "step": 10680, "loss": 2.7254518866539, "perplexity": 15.263309654625411, "learning_rate": 0.00028968499679758996, "grad_norm": 0.5674614906311035, "tokens_per_sec": 135479.78036809695, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 536 |
+
{"event": "train", "epoch": 0, "step": 10700, "loss": 2.686872661113739, "perplexity": 14.68567694216364, "learning_rate": 0.00028964677873993165, "grad_norm": 0.5078374743461609, "tokens_per_sec": 134484.72746380264, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 537 |
+
{"event": "train", "epoch": 0, "step": 10720, "loss": 2.796050250530243, "perplexity": 16.379822640045354, "learning_rate": 0.000289608492542927, "grad_norm": 0.4605364501476288, "tokens_per_sec": 135407.70252640144, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 538 |
+
{"event": "train", "epoch": 0, "step": 10740, "loss": 3.041315197944641, "perplexity": 20.932755856333408, "learning_rate": 0.0002895701382252574, "grad_norm": 0.6421473026275635, "tokens_per_sec": 134844.14754087102, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 539 |
+
{"event": "train", "epoch": 0, "step": 10760, "loss": 2.356119215488434, "perplexity": 10.549929901457963, "learning_rate": 0.0002895317158056377, "grad_norm": 0.4786122143268585, "tokens_per_sec": 134786.09357819398, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 540 |
+
{"event": "train", "epoch": 0, "step": 10780, "loss": 2.841607928276062, "perplexity": 17.143308598218628, "learning_rate": 0.00028949322530281567, "grad_norm": 0.5010853409767151, "tokens_per_sec": 135178.498618108, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 541 |
+
{"event": "train", "epoch": 0, "step": 10800, "loss": 2.242339551448822, "perplexity": 9.415333196529103, "learning_rate": 0.0002894546667355724, "grad_norm": 0.4561033546924591, "tokens_per_sec": 134160.73589051527, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 542 |
+
{"event": "train", "epoch": 0, "step": 10820, "loss": 3.264196991920471, "perplexity": 26.159096588670934, "learning_rate": 0.00028941604012272217, "grad_norm": 2.445380449295044, "tokens_per_sec": 134564.25724448604, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 543 |
+
{"event": "train", "epoch": 0, "step": 10840, "loss": 3.3153374791145325, "perplexity": 27.531683785188847, "learning_rate": 0.00028937734548311223, "grad_norm": 8.054612159729004, "tokens_per_sec": 134723.7313895745, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 544 |
+
{"event": "train", "epoch": 0, "step": 10860, "loss": 3.245702028274536, "perplexity": 25.679731631349583, "learning_rate": 0.00028933858283562366, "grad_norm": 3.441946268081665, "tokens_per_sec": 134739.9769928561, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 545 |
+
{"event": "train", "epoch": 0, "step": 10880, "loss": 3.3152948021888733, "perplexity": 27.530508842638323, "learning_rate": 0.00028929975219917027, "grad_norm": 10.35793399810791, "tokens_per_sec": 134224.4130049446, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 546 |
+
{"event": "train", "epoch": 0, "step": 10900, "loss": 3.290513515472412, "perplexity": 26.856651421522223, "learning_rate": 0.00028926085359269897, "grad_norm": 8.219490051269531, "tokens_per_sec": 135042.75483029132, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 547 |
+
{"event": "train", "epoch": 0, "step": 10920, "loss": 3.2581329345703125, "perplexity": 26.000946327491015, "learning_rate": 0.0002892218870351902, "grad_norm": 1.8442765474319458, "tokens_per_sec": 134764.02239155013, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 548 |
+
{"event": "train", "epoch": 0, "step": 10940, "loss": 2.985214650630951, "perplexity": 19.79074987742279, "learning_rate": 0.000289182852545657, "grad_norm": 0.6031453609466553, "tokens_per_sec": 134482.3588134855, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 549 |
+
{"event": "train", "epoch": 0, "step": 10960, "loss": 2.034467190504074, "perplexity": 7.648176018871548, "learning_rate": 0.000289143750143146, "grad_norm": 0.5360856056213379, "tokens_per_sec": 134565.97001223871, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 550 |
+
{"event": "train", "epoch": 0, "step": 10980, "loss": 2.4513023495674133, "perplexity": 11.603448629408756, "learning_rate": 0.0002891045798467371, "grad_norm": 0.5067413449287415, "tokens_per_sec": 134328.9691776002, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 551 |
+
{"event": "train", "epoch": 0, "step": 11000, "loss": 2.3346508145332336, "perplexity": 10.325853661747416, "learning_rate": 0.0002890653416755427, "grad_norm": 0.4596998393535614, "tokens_per_sec": 134901.5894755543, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 552 |
+
{"event": "train", "epoch": 0, "step": 11020, "loss": 2.795386493206024, "perplexity": 16.36895402026065, "learning_rate": 0.00028902603564870904, "grad_norm": 0.6009833812713623, "tokens_per_sec": 134639.39650056377, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 553 |
+
{"event": "train", "epoch": 0, "step": 11040, "loss": 2.6567850708961487, "perplexity": 14.250401345237973, "learning_rate": 0.0002889866617854149, "grad_norm": 0.553647518157959, "tokens_per_sec": 134411.8638854361, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 554 |
+
{"event": "train", "epoch": 0, "step": 11060, "loss": 2.9134445786476135, "perplexity": 18.42013903137743, "learning_rate": 0.0002889472201048726, "grad_norm": 0.5711304545402527, "tokens_per_sec": 134607.08952262305, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 555 |
+
{"event": "train", "epoch": 0, "step": 11080, "loss": 2.584758698940277, "perplexity": 13.260089030359637, "learning_rate": 0.0002889077106263272, "grad_norm": 0.58912193775177, "tokens_per_sec": 134296.81098537517, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 556 |
+
{"event": "train", "epoch": 0, "step": 11100, "loss": 3.0587600469589233, "perplexity": 21.301128381274793, "learning_rate": 0.00028886813336905686, "grad_norm": 0.7378895878791809, "tokens_per_sec": 134157.8548167468, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 557 |
+
{"event": "train", "epoch": 0, "step": 11120, "loss": 2.588454782962799, "perplexity": 13.309190118538083, "learning_rate": 0.0002888284883523731, "grad_norm": 0.5712403059005737, "tokens_per_sec": 133783.71481580264, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 558 |
+
{"event": "train", "epoch": 0, "step": 11140, "loss": 2.9673227071762085, "perplexity": 19.439803815307652, "learning_rate": 0.00028878877559562044, "grad_norm": 0.5390475988388062, "tokens_per_sec": 134392.1490606461, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 559 |
+
{"event": "train", "epoch": 0, "step": 11160, "loss": 2.285700798034668, "perplexity": 9.832574458961266, "learning_rate": 0.0002887489951181761, "grad_norm": 1.158182144165039, "tokens_per_sec": 134692.83593299804, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 560 |
+
{"event": "train", "epoch": 0, "step": 11180, "loss": 2.241079807281494, "perplexity": 9.403479753171966, "learning_rate": 0.0002887091469394507, "grad_norm": 0.5269469618797302, "tokens_per_sec": 135014.63076203686, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 561 |
+
{"event": "train", "epoch": 0, "step": 11200, "loss": 2.711267113685608, "perplexity": 15.048331385992975, "learning_rate": 0.0002886692310788878, "grad_norm": 0.45978283882141113, "tokens_per_sec": 134685.97220022735, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 562 |
+
{"event": "train", "epoch": 0, "step": 11220, "loss": 2.28374844789505, "perplexity": 9.813396557924047, "learning_rate": 0.000288629247555964, "grad_norm": 0.5239197611808777, "tokens_per_sec": 135086.4234083437, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 563 |
+
{"event": "train", "epoch": 0, "step": 11240, "loss": 3.335276424884796, "perplexity": 28.0861458604925, "learning_rate": 0.00028858919639018895, "grad_norm": 1.4540077447891235, "tokens_per_sec": 134841.23704041637, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 564 |
+
{"event": "train", "epoch": 0, "step": 11260, "loss": 3.28762286901474, "perplexity": 26.77913043408891, "learning_rate": 0.00028854907760110504, "grad_norm": 1.0307823419570923, "tokens_per_sec": 134049.9038040266, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 565 |
+
{"event": "train", "epoch": 0, "step": 11280, "loss": 3.3135814666748047, "perplexity": 27.483380229210315, "learning_rate": 0.000288508891208288, "grad_norm": 1.4850795269012451, "tokens_per_sec": 134177.10797707725, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 566 |
+
{"event": "train", "epoch": 0, "step": 11300, "loss": 3.302783727645874, "perplexity": 27.188218270530534, "learning_rate": 0.0002884686372313462, "grad_norm": 4.316898822784424, "tokens_per_sec": 134415.67608548526, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 567 |
+
{"event": "train", "epoch": 0, "step": 11320, "loss": 2.2601166367530823, "perplexity": 9.584206972356627, "learning_rate": 0.0002884283156899216, "grad_norm": 0.7005491256713867, "tokens_per_sec": 134913.11022241594, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 568 |
+
{"event": "train", "epoch": 0, "step": 11340, "loss": 2.401206284761429, "perplexity": 11.036481493596234, "learning_rate": 0.00028838792660368834, "grad_norm": 0.45523926615715027, "tokens_per_sec": 135001.89918786075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 569 |
+
{"event": "train", "epoch": 0, "step": 11360, "loss": 2.1290160715579987, "perplexity": 8.406591256593234, "learning_rate": 0.00028834746999235406, "grad_norm": 0.4973062574863434, "tokens_per_sec": 134346.82690397804, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 570 |
+
{"event": "train", "epoch": 0, "step": 11380, "loss": 2.5268081426620483, "perplexity": 12.513501015273212, "learning_rate": 0.0002883069458756592, "grad_norm": 0.5327191352844238, "tokens_per_sec": 134067.1642901039, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 571 |
+
{"event": "train", "epoch": 0, "step": 11400, "loss": 3.2832425236701965, "perplexity": 26.66208513133431, "learning_rate": 0.00028826635427337715, "grad_norm": 0.7593058943748474, "tokens_per_sec": 132792.28580704005, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 572 |
+
{"event": "train", "epoch": 0, "step": 11420, "loss": 2.8444331288337708, "perplexity": 17.191810364579197, "learning_rate": 0.0002882256952053139, "grad_norm": 0.8443586230278015, "tokens_per_sec": 122935.40338665286, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 573 |
+
{"event": "train", "epoch": 0, "step": 11440, "loss": 2.994892716407776, "perplexity": 19.983215903665574, "learning_rate": 0.0002881849686913091, "grad_norm": 0.4673523008823395, "tokens_per_sec": 134754.37677783202, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 574 |
+
{"event": "train", "epoch": 0, "step": 11460, "loss": 2.675844669342041, "perplexity": 14.52461315507558, "learning_rate": 0.0002881441747512346, "grad_norm": 0.5244850516319275, "tokens_per_sec": 134805.26304783445, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 575 |
+
{"event": "train", "epoch": 0, "step": 11480, "loss": 2.4118751883506775, "perplexity": 11.154859006891302, "learning_rate": 0.0002881033134049956, "grad_norm": 0.5670372247695923, "tokens_per_sec": 134822.45410518075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 576 |
+
{"event": "train", "epoch": 0, "step": 11500, "loss": 2.381237268447876, "perplexity": 10.81827970299114, "learning_rate": 0.00028806238467253, "grad_norm": 0.4266142249107361, "tokens_per_sec": 134973.52702731898, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 577 |
+
{"event": "train", "epoch": 0, "step": 11520, "loss": 2.431246519088745, "perplexity": 11.37304997178024, "learning_rate": 0.00028802138857380856, "grad_norm": 0.47468793392181396, "tokens_per_sec": 132353.5336486191, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 578 |
+
{"event": "train", "epoch": 0, "step": 11540, "loss": 2.807965636253357, "perplexity": 16.576161951233235, "learning_rate": 0.0002879803251288348, "grad_norm": 0.5031028389930725, "tokens_per_sec": 132341.55384264293, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 579 |
+
{"event": "train", "epoch": 0, "step": 11560, "loss": 2.4691904187202454, "perplexity": 11.812879493294037, "learning_rate": 0.0002879391943576454, "grad_norm": 0.47706326842308044, "tokens_per_sec": 134915.75894815166, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 580 |
+
{"event": "train", "epoch": 0, "step": 11580, "loss": 2.7079745829105377, "perplexity": 14.998865770009665, "learning_rate": 0.0002878979962803096, "grad_norm": 0.5082792043685913, "tokens_per_sec": 133864.1123516492, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 581 |
+
{"event": "train", "epoch": 0, "step": 11600, "loss": 3.2905282974243164, "perplexity": 26.857048418186032, "learning_rate": 0.0002878567309169298, "grad_norm": 8.563667297363281, "tokens_per_sec": 134568.34160880104, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 582 |
+
{"event": "train", "epoch": 0, "step": 11620, "loss": 3.250087559223175, "perplexity": 25.792598198186855, "learning_rate": 0.00028781539828764095, "grad_norm": 0.9418319463729858, "tokens_per_sec": 134276.08039821993, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 583 |
+
{"event": "train", "epoch": 0, "step": 11640, "loss": 2.7337477803230286, "perplexity": 15.390459128596222, "learning_rate": 0.00028777399841261095, "grad_norm": 0.540590763092041, "tokens_per_sec": 133347.84168898204, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 584 |
+
{"event": "train", "epoch": 0, "step": 11660, "loss": 2.2677418291568756, "perplexity": 9.65756773406578, "learning_rate": 0.00028773253131204033, "grad_norm": 0.4835197329521179, "tokens_per_sec": 134820.07354341907, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 585 |
+
{"event": "train", "epoch": 0, "step": 11680, "loss": 2.5625768303871155, "perplexity": 12.969193706865207, "learning_rate": 0.00028769099700616296, "grad_norm": 0.5887776613235474, "tokens_per_sec": 134754.77314571058, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 586 |
+
{"event": "train", "epoch": 0, "step": 11700, "loss": 4.403090536594391, "perplexity": 81.70298494183972, "learning_rate": 0.00028764939551524486, "grad_norm": 2.576824903488159, "tokens_per_sec": 134982.40861756887, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 587 |
+
{"event": "train", "epoch": 0, "step": 11720, "loss": 2.304483711719513, "perplexity": 10.019004222432166, "learning_rate": 0.000287607726859585, "grad_norm": 0.5357518196105957, "tokens_per_sec": 135256.18909242642, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 588 |
+
{"event": "train", "epoch": 0, "step": 11740, "loss": 2.1637511253356934, "perplexity": 8.70372526450922, "learning_rate": 0.0002875659910595153, "grad_norm": 0.7640151977539062, "tokens_per_sec": 129087.62929477455, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 589 |
+
{"event": "train", "epoch": 0, "step": 11760, "loss": 2.6872137784957886, "perplexity": 14.690687336353466, "learning_rate": 0.00028752418813540037, "grad_norm": 0.512372612953186, "tokens_per_sec": 134648.62991428626, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 590 |
+
{"event": "train", "epoch": 0, "step": 11780, "loss": 3.322639524936676, "perplexity": 27.733457184509284, "learning_rate": 0.0002874823181076374, "grad_norm": 0.7323374152183533, "tokens_per_sec": 134804.86638287856, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 591 |
+
{"event": "train", "epoch": 0, "step": 11800, "loss": 3.3795665502548218, "perplexity": 29.358043118706973, "learning_rate": 0.00028744038099665667, "grad_norm": 0.6540385484695435, "tokens_per_sec": 128526.29738017198, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 592 |
+
{"event": "train", "epoch": 0, "step": 11820, "loss": 2.2893420457839966, "perplexity": 9.86844256127316, "learning_rate": 0.0002873983768229209, "grad_norm": 0.46081703901290894, "tokens_per_sec": 134164.00998818834, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 593 |
+
{"event": "train", "epoch": 0, "step": 11840, "loss": 5.076861500740051, "perplexity": 160.27025767182622, "learning_rate": 0.00028735630560692595, "grad_norm": 0.6969375014305115, "tokens_per_sec": 134947.41927001765, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 594 |
+
{"event": "train", "epoch": 0, "step": 11860, "loss": 5.385310530662537, "perplexity": 218.17784449979425, "learning_rate": 0.0002873141673691998, "grad_norm": 0.8614420294761658, "tokens_per_sec": 134991.02620283773, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 595 |
+
{"event": "train", "epoch": 0, "step": 11880, "loss": 2.9434564113616943, "perplexity": 18.981340380381795, "learning_rate": 0.0002872719621303034, "grad_norm": 0.640351414680481, "tokens_per_sec": 135360.62641403885, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 596 |
+
{"event": "train", "epoch": 0, "step": 11900, "loss": 2.257644087076187, "perplexity": 9.560538816906828, "learning_rate": 0.00028722968991083024, "grad_norm": 0.46941861510276794, "tokens_per_sec": 134193.74610982608, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 597 |
+
{"event": "train", "epoch": 0, "step": 11920, "loss": 2.7557350993156433, "perplexity": 15.732601677850768, "learning_rate": 0.0002871873507314071, "grad_norm": 0.5791509747505188, "tokens_per_sec": 134615.527400306, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 598 |
+
{"event": "train", "epoch": 0, "step": 11940, "loss": 2.8866251707077026, "perplexity": 17.932687605223812, "learning_rate": 0.00028714494461269267, "grad_norm": 0.49778270721435547, "tokens_per_sec": 133516.76498383487, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 599 |
+
{"event": "train", "epoch": 0, "step": 11960, "loss": 3.4292962551116943, "perplexity": 30.854921114315324, "learning_rate": 0.00028710247157537886, "grad_norm": 2.572869062423706, "tokens_per_sec": 132627.49327111064, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 600 |
+
{"event": "train", "epoch": 0, "step": 11980, "loss": 2.4067004323005676, "perplexity": 11.097284428434465, "learning_rate": 0.0002870599316401898, "grad_norm": 0.4559401571750641, "tokens_per_sec": 134252.07765268066, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|