Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +84 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -37,3 +37,87 @@
|
|
| 37 |
{"event": "train", "epoch": 0, "step": 720, "loss": 4.7866517305374146, "perplexity": 119.89924085602638, "learning_rate": 0.00029995257466183546, "grad_norm": 0.7744899392127991, "tokens_per_sec": 135045.67403672123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 38 |
{"event": "train", "epoch": 0, "step": 740, "loss": 4.666993021965027, "perplexity": 106.3773865575008, "learning_rate": 0.00029994990347590475, "grad_norm": 0.7438130974769592, "tokens_per_sec": 134972.20151707396, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 39 |
{"event": "train", "epoch": 0, "step": 760, "loss": 4.43474018573761, "perplexity": 84.33021184996495, "learning_rate": 0.00029994715912331387, "grad_norm": 0.7224310636520386, "tokens_per_sec": 135022.19127045022, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
{"event": "train", "epoch": 0, "step": 720, "loss": 4.7866517305374146, "perplexity": 119.89924085602638, "learning_rate": 0.00029995257466183546, "grad_norm": 0.7744899392127991, "tokens_per_sec": 135045.67403672123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 38 |
{"event": "train", "epoch": 0, "step": 740, "loss": 4.666993021965027, "perplexity": 106.3773865575008, "learning_rate": 0.00029994990347590475, "grad_norm": 0.7438130974769592, "tokens_per_sec": 134972.20151707396, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 39 |
{"event": "train", "epoch": 0, "step": 760, "loss": 4.43474018573761, "perplexity": 84.33021184996495, "learning_rate": 0.00029994715912331387, "grad_norm": 0.7224310636520386, "tokens_per_sec": 135022.19127045022, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 40 |
+
{"event": "train", "epoch": 0, "step": 780, "loss": 4.11405873298645, "perplexity": 61.19458669967641, "learning_rate": 0.00029994434160540213, "grad_norm": 0.8463916182518005, "tokens_per_sec": 135000.7057258848, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 41 |
+
{"event": "train", "epoch": 0, "step": 800, "loss": 5.154967427253723, "perplexity": 173.29016213047174, "learning_rate": 0.0002999414509235442, "grad_norm": 1.035745620727539, "tokens_per_sec": 134745.26096014574, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 42 |
+
{"event": "train", "epoch": 0, "step": 820, "loss": 4.68245005607605, "perplexity": 108.0344390165837, "learning_rate": 0.00029993848707915075, "grad_norm": 0.7734178900718689, "tokens_per_sec": 134917.48067565274, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 43 |
+
{"event": "train", "epoch": 0, "step": 840, "loss": 4.926585555076599, "perplexity": 137.9078288434663, "learning_rate": 0.00029993545007366766, "grad_norm": 0.9916408658027649, "tokens_per_sec": 134949.27431025018, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 44 |
+
{"event": "train", "epoch": 0, "step": 860, "loss": 4.183092296123505, "perplexity": 65.56829663469706, "learning_rate": 0.0002999323399085769, "grad_norm": 1.1518107652664185, "tokens_per_sec": 135024.18101866223, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 45 |
+
{"event": "train", "epoch": 0, "step": 880, "loss": 4.418570637702942, "perplexity": 82.97759550663422, "learning_rate": 0.0002999291565853962, "grad_norm": 0.8036053776741028, "tokens_per_sec": 135066.3774850968, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 46 |
+
{"event": "train", "epoch": 0, "step": 900, "loss": 3.836549162864685, "perplexity": 46.36519930559673, "learning_rate": 0.00029992590010567856, "grad_norm": 0.7838431000709534, "tokens_per_sec": 134988.77226658913, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 47 |
+
{"event": "train", "epoch": 0, "step": 920, "loss": 4.65129017829895, "perplexity": 104.72000593734202, "learning_rate": 0.00029992257047101303, "grad_norm": 0.7456551790237427, "tokens_per_sec": 134937.34994143574, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 48 |
+
{"event": "train", "epoch": 0, "step": 940, "loss": 4.928666949272156, "perplexity": 138.19516832757571, "learning_rate": 0.0002999191676830243, "grad_norm": 1.0507354736328125, "tokens_per_sec": 134983.7343283025, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 49 |
+
{"event": "train", "epoch": 0, "step": 960, "loss": 4.44944953918457, "perplexity": 85.57982269518875, "learning_rate": 0.00029991569174337286, "grad_norm": 2.0399932861328125, "tokens_per_sec": 134872.4653292497, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 50 |
+
{"event": "train", "epoch": 0, "step": 980, "loss": 4.066194295883179, "perplexity": 58.33453563234353, "learning_rate": 0.00029991214265375453, "grad_norm": 0.8431903719902039, "tokens_per_sec": 134975.647897864, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 51 |
+
{"event": "train", "epoch": 0, "step": 1000, "loss": 4.395802140235901, "perplexity": 81.10966600217003, "learning_rate": 0.0002999085204159013, "grad_norm": 0.7001171708106995, "tokens_per_sec": 135030.81393626073, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 52 |
+
{"event": "train", "epoch": 0, "step": 1020, "loss": 4.127518594264984, "perplexity": 62.023825548337925, "learning_rate": 0.00029990482503158063, "grad_norm": 0.7451835870742798, "tokens_per_sec": 134990.89361626896, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 53 |
+
{"event": "train", "epoch": 0, "step": 1040, "loss": 3.617012619972229, "perplexity": 37.22619276030402, "learning_rate": 0.0002999010565025954, "grad_norm": 0.9016689658164978, "tokens_per_sec": 134814.7837071405, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 54 |
+
{"event": "train", "epoch": 0, "step": 1060, "loss": 4.48874032497406, "perplexity": 89.00925249744537, "learning_rate": 0.0002998972148307845, "grad_norm": 0.8111092448234558, "tokens_per_sec": 104135.09287823891, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 55 |
+
{"event": "train", "epoch": 0, "step": 1080, "loss": 4.205470681190491, "perplexity": 67.05215042606443, "learning_rate": 0.0002998933000180222, "grad_norm": 0.8757339715957642, "tokens_per_sec": 134828.53814497514, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 56 |
+
{"event": "train", "epoch": 0, "step": 1100, "loss": 3.9755062460899353, "perplexity": 53.277081393965226, "learning_rate": 0.00029988931206621934, "grad_norm": 0.9147546291351318, "tokens_per_sec": 134846.13204503802, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 57 |
+
{"event": "train", "epoch": 0, "step": 1120, "loss": 4.135497808456421, "perplexity": 62.52070666149828, "learning_rate": 0.0002998852509773214, "grad_norm": 0.7158412337303162, "tokens_per_sec": 134842.1630951143, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 58 |
+
{"event": "train", "epoch": 0, "step": 1140, "loss": 3.8115729093551636, "perplexity": 45.221512295378794, "learning_rate": 0.0002998811167533099, "grad_norm": 0.7186790108680725, "tokens_per_sec": 133811.4586397354, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 59 |
+
{"event": "train", "epoch": 0, "step": 1160, "loss": 3.8230870962142944, "perplexity": 45.745210431271495, "learning_rate": 0.00029987690939620236, "grad_norm": 0.7453848123550415, "tokens_per_sec": 134819.54454111075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 60 |
+
{"event": "train", "epoch": 0, "step": 1180, "loss": 4.347952842712402, "perplexity": 77.32001456494545, "learning_rate": 0.0002998726289080514, "grad_norm": 0.689865231513977, "tokens_per_sec": 134928.07688916943, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 61 |
+
{"event": "train", "epoch": 0, "step": 1200, "loss": 3.961314022541046, "perplexity": 52.526301365540434, "learning_rate": 0.0002998682752909458, "grad_norm": 0.6931058764457703, "tokens_per_sec": 134923.04348022427, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 62 |
+
{"event": "train", "epoch": 0, "step": 1220, "loss": 3.66582989692688, "perplexity": 39.08856218374329, "learning_rate": 0.0002998638485470098, "grad_norm": 0.6872467994689941, "tokens_per_sec": 134966.23704313277, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 63 |
+
{"event": "train", "epoch": 0, "step": 1240, "loss": 3.7248464226722717, "perplexity": 41.46486419277333, "learning_rate": 0.00029985934867840345, "grad_norm": 0.7044392228126526, "tokens_per_sec": 135435.18983834173, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 64 |
+
{"event": "train", "epoch": 0, "step": 1260, "loss": 3.9544836282730103, "perplexity": 52.168748521334216, "learning_rate": 0.0002998547756873224, "grad_norm": 0.8836699724197388, "tokens_per_sec": 135480.04746569353, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 65 |
+
{"event": "train", "epoch": 0, "step": 1280, "loss": 3.9815263748168945, "perplexity": 53.59878365507288, "learning_rate": 0.0002998501295759979, "grad_norm": 0.8123427629470825, "tokens_per_sec": 135471.90146276072, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 66 |
+
{"event": "train", "epoch": 0, "step": 1300, "loss": 4.011739075183868, "perplexity": 55.24285855850677, "learning_rate": 0.0002998454103466972, "grad_norm": 0.7360571622848511, "tokens_per_sec": 135510.50349673792, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 67 |
+
{"event": "train", "epoch": 0, "step": 1320, "loss": 3.4758657813072205, "perplexity": 32.32580350117473, "learning_rate": 0.0002998406180017229, "grad_norm": 1.0972497463226318, "tokens_per_sec": 104583.35912071388, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 68 |
+
{"event": "train", "epoch": 0, "step": 1340, "loss": 3.8330634236335754, "perplexity": 46.20386366176792, "learning_rate": 0.00029983575254341325, "grad_norm": 0.7529428601264954, "tokens_per_sec": 103373.15667500538, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 69 |
+
{"event": "train", "epoch": 0, "step": 1360, "loss": 4.201993048191071, "perplexity": 66.81937264750262, "learning_rate": 0.0002998308139741426, "grad_norm": 0.8064389824867249, "tokens_per_sec": 101098.0472216215, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 70 |
+
{"event": "train", "epoch": 0, "step": 1380, "loss": 3.3918808698654175, "perplexity": 29.721802573133825, "learning_rate": 0.0002998258022963205, "grad_norm": 0.7430403232574463, "tokens_per_sec": 103993.10652413408, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 71 |
+
{"event": "train", "epoch": 0, "step": 1400, "loss": 3.6433388590812683, "perplexity": 38.219232560391596, "learning_rate": 0.00029982071751239234, "grad_norm": 0.6664264798164368, "tokens_per_sec": 103147.93741730427, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 72 |
+
{"event": "train", "epoch": 0, "step": 1420, "loss": 3.137860894203186, "perplexity": 23.054498064557794, "learning_rate": 0.00029981555962483897, "grad_norm": 1.1427690982818604, "tokens_per_sec": 134690.32794953385, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 73 |
+
{"event": "train", "epoch": 0, "step": 1440, "loss": 4.018308162689209, "perplexity": 55.60694828978166, "learning_rate": 0.0002998103286361776, "grad_norm": 0.6634413599967957, "tokens_per_sec": 134905.9591823914, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 74 |
+
{"event": "train", "epoch": 0, "step": 1460, "loss": 3.775096118450165, "perplexity": 43.60169904205368, "learning_rate": 0.00029980502454896046, "grad_norm": 0.6687666773796082, "tokens_per_sec": 134975.11767397917, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 75 |
+
{"event": "train", "epoch": 0, "step": 1480, "loss": 3.0987940430641174, "perplexity": 22.17119764324156, "learning_rate": 0.0002997996473657755, "grad_norm": 0.5723268389701843, "tokens_per_sec": 134863.46586438778, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 76 |
+
{"event": "train", "epoch": 0, "step": 1500, "loss": 4.166404128074646, "perplexity": 64.48316152392603, "learning_rate": 0.0002997941970892467, "grad_norm": 1.1436331272125244, "tokens_per_sec": 134981.61320362796, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 77 |
+
{"event": "train", "epoch": 0, "step": 1520, "loss": 3.611291229724884, "perplexity": 37.01381531056966, "learning_rate": 0.00029978867372203324, "grad_norm": 0.7986361384391785, "tokens_per_sec": 134887.5553623217, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 78 |
+
{"event": "train", "epoch": 0, "step": 1540, "loss": 3.8298484086990356, "perplexity": 46.05555608334787, "learning_rate": 0.0002997830772668303, "grad_norm": 0.8444863557815552, "tokens_per_sec": 134897.08764942194, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 79 |
+
{"event": "train", "epoch": 0, "step": 1560, "loss": 3.403752028942108, "perplexity": 30.07673739515371, "learning_rate": 0.0002997774077263686, "grad_norm": 0.6463981866836548, "tokens_per_sec": 134859.4958940348, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 80 |
+
{"event": "train", "epoch": 0, "step": 1580, "loss": 3.825500249862671, "perplexity": 45.85573395420638, "learning_rate": 0.00029977166510341456, "grad_norm": 0.6561903357505798, "tokens_per_sec": 134818.4865489481, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 81 |
+
{"event": "train", "epoch": 0, "step": 1600, "loss": 3.753374934196472, "perplexity": 42.66483028783862, "learning_rate": 0.0002997658494007699, "grad_norm": 1.1277724504470825, "tokens_per_sec": 134849.9689187293, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 82 |
+
{"event": "train", "epoch": 0, "step": 1620, "loss": 3.2838041186332703, "perplexity": 26.677062629299176, "learning_rate": 0.0002997599606212728, "grad_norm": 0.8155786991119385, "tokens_per_sec": 134987.04871906686, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 83 |
+
{"event": "train", "epoch": 0, "step": 1640, "loss": 3.2464561462402344, "perplexity": 25.699104482116397, "learning_rate": 0.00029975399876779665, "grad_norm": 0.7335466146469116, "tokens_per_sec": 134883.58397361977, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 84 |
+
{"event": "train", "epoch": 0, "step": 1660, "loss": 3.514792799949646, "perplexity": 33.60896342351157, "learning_rate": 0.00029974796384325017, "grad_norm": 0.8131045699119568, "tokens_per_sec": 134803.0153106162, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 85 |
+
{"event": "train", "epoch": 0, "step": 1680, "loss": 3.6981714367866516, "perplexity": 40.37341148671972, "learning_rate": 0.0002997418558505784, "grad_norm": 1.2860559225082397, "tokens_per_sec": 134967.29735552822, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 86 |
+
{"event": "train", "epoch": 0, "step": 1700, "loss": 3.889654576778412, "perplexity": 48.89399448535724, "learning_rate": 0.00029973567479276147, "grad_norm": 0.6860939264297485, "tokens_per_sec": 134984.6623413103, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 87 |
+
{"event": "train", "epoch": 0, "step": 1720, "loss": 3.2484377026557922, "perplexity": 25.750079195432498, "learning_rate": 0.0002997294206728152, "grad_norm": 0.5407994985580444, "tokens_per_sec": 134914.30213612667, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 88 |
+
{"event": "train", "epoch": 0, "step": 1740, "loss": 3.762909233570099, "perplexity": 43.07355490778997, "learning_rate": 0.0002997230934937915, "grad_norm": 1.1168618202209473, "tokens_per_sec": 135094.65590802798, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 89 |
+
{"event": "train", "epoch": 0, "step": 1760, "loss": 3.664206802845001, "perplexity": 39.02516923007473, "learning_rate": 0.00029971669325877764, "grad_norm": 0.6551121473312378, "tokens_per_sec": 134823.11538725646, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 90 |
+
{"event": "train", "epoch": 0, "step": 1780, "loss": 3.106242835521698, "perplexity": 22.33696290207731, "learning_rate": 0.00029971021997089633, "grad_norm": 0.533824622631073, "tokens_per_sec": 135040.4991663072, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 91 |
+
{"event": "train", "epoch": 0, "step": 1800, "loss": 3.743643283843994, "perplexity": 42.25164482624064, "learning_rate": 0.00029970367363330627, "grad_norm": 0.7927202582359314, "tokens_per_sec": 134952.71952027793, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 92 |
+
{"event": "train", "epoch": 0, "step": 1820, "loss": 3.306063711643219, "perplexity": 27.277541600840653, "learning_rate": 0.00029969705424920173, "grad_norm": 0.7511147260665894, "tokens_per_sec": 134986.51840560694, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 93 |
+
{"event": "train", "epoch": 0, "step": 1840, "loss": 3.1633567810058594, "perplexity": 23.649850202980648, "learning_rate": 0.00029969036182181255, "grad_norm": 1.3380285501480103, "tokens_per_sec": 134934.56789192586, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 94 |
+
{"event": "train", "epoch": 0, "step": 1860, "loss": 3.282947838306427, "perplexity": 26.654229362625408, "learning_rate": 0.00029968359635440436, "grad_norm": 0.7367568612098694, "tokens_per_sec": 134982.67375763235, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 95 |
+
{"event": "train", "epoch": 0, "step": 1880, "loss": 3.260329842567444, "perplexity": 26.058130805930244, "learning_rate": 0.00029967675785027827, "grad_norm": 0.6488712430000305, "tokens_per_sec": 134974.72000879934, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 96 |
+
{"event": "train", "epoch": 0, "step": 1900, "loss": 4.099326848983765, "perplexity": 60.29968313944519, "learning_rate": 0.00029966984631277064, "grad_norm": 0.8896181583404541, "tokens_per_sec": 134897.08764942194, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 97 |
+
{"event": "train", "epoch": 0, "step": 1920, "loss": 3.2325087785720825, "perplexity": 25.343157653951693, "learning_rate": 0.00029966286174525433, "grad_norm": 0.5709714293479919, "tokens_per_sec": 134444.20872270182, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 98 |
+
{"event": "train", "epoch": 0, "step": 1940, "loss": 2.805585563182831, "perplexity": 16.536756387223168, "learning_rate": 0.0002996558041511374, "grad_norm": 0.5968441963195801, "tokens_per_sec": 134758.7369527468, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 99 |
+
{"event": "train", "epoch": 0, "step": 1960, "loss": 3.1132136583328247, "perplexity": 22.493213878826293, "learning_rate": 0.0002996486735338635, "grad_norm": 0.6275550127029419, "tokens_per_sec": 134298.38572189352, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 100 |
+
{"event": "train", "epoch": 0, "step": 1980, "loss": 3.7214125990867615, "perplexity": 41.32272534366938, "learning_rate": 0.0002996414698969118, "grad_norm": 1.8389214277267456, "tokens_per_sec": 134824.9670115795, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 101 |
+
{"event": "train", "epoch": 0, "step": 2000, "loss": 3.5565409064292908, "perplexity": 35.0417744831828, "learning_rate": 0.0002996341932437975, "grad_norm": 0.6495166420936584, "tokens_per_sec": 103958.81060049075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 102 |
+
{"event": "train", "epoch": 0, "step": 2020, "loss": 3.535371959209442, "perplexity": 34.307773433504224, "learning_rate": 0.0002996268435780708, "grad_norm": 0.7705476880073547, "tokens_per_sec": 103706.37943362484, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 103 |
+
{"event": "train", "epoch": 0, "step": 2040, "loss": 2.8529849648475647, "perplexity": 17.33946235554151, "learning_rate": 0.00029961942090331816, "grad_norm": 0.6001657843589783, "tokens_per_sec": 113197.20585263002, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 104 |
+
{"event": "train", "epoch": 0, "step": 2060, "loss": 3.194823741912842, "perplexity": 24.405871580925012, "learning_rate": 0.00029961192522316135, "grad_norm": 0.5786985158920288, "tokens_per_sec": 114397.09133525328, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 105 |
+
{"event": "train", "epoch": 0, "step": 2080, "loss": 3.242600917816162, "perplexity": 25.60021929900381, "learning_rate": 0.00029960435654125786, "grad_norm": 0.5967770218849182, "tokens_per_sec": 109205.42080004068, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 106 |
+
{"event": "train", "epoch": 0, "step": 2100, "loss": 2.986021399497986, "perplexity": 19.806722484539215, "learning_rate": 0.00029959671486130086, "grad_norm": 0.6310850381851196, "tokens_per_sec": 109433.58569071718, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 107 |
+
{"event": "train", "epoch": 0, "step": 2120, "loss": 3.0434985160827637, "perplexity": 20.978508650152907, "learning_rate": 0.0002995890001870188, "grad_norm": 0.5795332789421082, "tokens_per_sec": 102174.0045110345, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 108 |
+
{"event": "train", "epoch": 0, "step": 2140, "loss": 2.767221510410309, "perplexity": 15.91435465468213, "learning_rate": 0.0002995812125221761, "grad_norm": 0.5952848196029663, "tokens_per_sec": 110856.99747537075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 109 |
+
{"event": "train", "epoch": 0, "step": 2160, "loss": 3.160609543323517, "perplexity": 23.584967608083524, "learning_rate": 0.0002995733518705725, "grad_norm": 0.6665594577789307, "tokens_per_sec": 113447.15015683297, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 110 |
+
{"event": "train", "epoch": 0, "step": 2180, "loss": 3.224173426628113, "perplexity": 25.132791471603326, "learning_rate": 0.00029956541823604373, "grad_norm": 0.5907281637191772, "tokens_per_sec": 99320.09836117699, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 111 |
+
{"event": "train", "epoch": 0, "step": 2200, "loss": 3.19379723072052, "perplexity": 24.38083153473144, "learning_rate": 0.000299557411622461, "grad_norm": 0.6079298853874207, "tokens_per_sec": 108421.63705927736, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 112 |
+
{"event": "train", "epoch": 0, "step": 2220, "loss": 2.8439940214157104, "perplexity": 17.184262970297905, "learning_rate": 0.00029954933203373094, "grad_norm": 0.5185272693634033, "tokens_per_sec": 113961.41954688713, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 113 |
+
{"event": "train", "epoch": 0, "step": 2240, "loss": 3.0640907287597656, "perplexity": 21.414981105391753, "learning_rate": 0.00029954117947379575, "grad_norm": 0.6393017768859863, "tokens_per_sec": 114024.76595624648, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 114 |
+
{"event": "train", "epoch": 0, "step": 2260, "loss": 3.5985727310180664, "perplexity": 36.54603617820668, "learning_rate": 0.0002995329539466339, "grad_norm": 0.5479193925857544, "tokens_per_sec": 113192.63770503145, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 115 |
+
{"event": "train", "epoch": 0, "step": 2280, "loss": 3.3031176328659058, "perplexity": 27.19729807434701, "learning_rate": 0.00029952465545625846, "grad_norm": 0.6662876605987549, "tokens_per_sec": 114540.38360319923, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 116 |
+
{"event": "train", "epoch": 0, "step": 2300, "loss": 3.539451479911804, "perplexity": 34.44801857745993, "learning_rate": 0.00029951628400671886, "grad_norm": 0.7319508790969849, "tokens_per_sec": 112924.88143536636, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 117 |
+
{"event": "train", "epoch": 0, "step": 2320, "loss": 2.5574251413345337, "perplexity": 12.902552259021439, "learning_rate": 0.0002995078396020997, "grad_norm": 0.6618043780326843, "tokens_per_sec": 113693.68605777543, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 118 |
+
{"event": "train", "epoch": 0, "step": 2340, "loss": 3.195091664791107, "perplexity": 24.41241134832308, "learning_rate": 0.00029949932224652146, "grad_norm": 0.5521330237388611, "tokens_per_sec": 113106.37830684782, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 119 |
+
{"event": "train", "epoch": 0, "step": 2360, "loss": 3.0794243812561035, "perplexity": 21.74588145593026, "learning_rate": 0.00029949073194414, "grad_norm": 0.5495889186859131, "tokens_per_sec": 113145.57928853742, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 120 |
+
{"event": "train", "epoch": 0, "step": 2380, "loss": 3.329685866832733, "perplexity": 27.9295667217976, "learning_rate": 0.00029948206869914696, "grad_norm": 0.5497113466262817, "tokens_per_sec": 113288.64629752509, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 121 |
+
{"event": "train", "epoch": 0, "step": 2400, "loss": 3.46912282705307, "perplexity": 32.10856532358906, "learning_rate": 0.0002994733325157695, "grad_norm": 0.6895716190338135, "tokens_per_sec": 113380.14105969662, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 122 |
+
{"event": "train", "epoch": 0, "step": 2420, "loss": 3.491629123687744, "perplexity": 32.83940360157507, "learning_rate": 0.00029946452339827035, "grad_norm": 0.5877766609191895, "tokens_per_sec": 113186.67170015186, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 123 |
+
{"event": "train", "epoch": 0, "step": 2440, "loss": 2.815937638282776, "perplexity": 16.708835281330597, "learning_rate": 0.00029945564135094795, "grad_norm": 0.8016887307167053, "tokens_per_sec": 113890.12094453946, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|