Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +87 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -348,3 +348,90 @@
|
|
| 348 |
{"event": "train", "epoch": 0, "step": 6940, "loss": 2.8565449118614197, "perplexity": 17.401299926707615, "learning_rate": 0.0002956150978492763, "grad_norm": 0.8099679946899414, "tokens_per_sec": 135084.1662853144, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 349 |
{"event": "train", "epoch": 0, "step": 6960, "loss": 2.299078106880188, "perplexity": 9.964991561795332, "learning_rate": 0.000295589912974245, "grad_norm": 0.42535004019737244, "tokens_per_sec": 134954.17716209157, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 350 |
{"event": "train", "epoch": 0, "step": 6980, "loss": 2.853616416454315, "perplexity": 17.350414844525563, "learning_rate": 0.0002955646570599705, "grad_norm": 1.3053666353225708, "tokens_per_sec": 134851.02740506933, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 348 |
{"event": "train", "epoch": 0, "step": 6940, "loss": 2.8565449118614197, "perplexity": 17.401299926707615, "learning_rate": 0.0002956150978492763, "grad_norm": 0.8099679946899414, "tokens_per_sec": 135084.1662853144, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 349 |
{"event": "train", "epoch": 0, "step": 6960, "loss": 2.299078106880188, "perplexity": 9.964991561795332, "learning_rate": 0.000295589912974245, "grad_norm": 0.42535004019737244, "tokens_per_sec": 134954.17716209157, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 350 |
{"event": "train", "epoch": 0, "step": 6980, "loss": 2.853616416454315, "perplexity": 17.350414844525563, "learning_rate": 0.0002955646570599705, "grad_norm": 1.3053666353225708, "tokens_per_sec": 134851.02740506933, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 351 |
+
{"event": "train", "epoch": 0, "step": 7000, "loss": 2.4062060117721558, "perplexity": 11.091799059354702, "learning_rate": 0.00029553933011877606, "grad_norm": 0.5070578455924988, "tokens_per_sec": 135082.44030094994, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 352 |
+
{"event": "train", "epoch": 0, "step": 7020, "loss": 3.0852232575416565, "perplexity": 21.872349463985476, "learning_rate": 0.00029551393216301975, "grad_norm": 0.5308393836021423, "tokens_per_sec": 134835.54869774397, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 353 |
+
{"event": "train", "epoch": 0, "step": 7040, "loss": 2.5913652777671814, "perplexity": 13.34798287289925, "learning_rate": 0.00029548846320509445, "grad_norm": 0.48368656635284424, "tokens_per_sec": 134433.95135731006, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 354 |
+
{"event": "train", "epoch": 0, "step": 7060, "loss": 2.481366455554962, "perplexity": 11.957592778394938, "learning_rate": 0.0002954629232574272, "grad_norm": 0.5330314040184021, "tokens_per_sec": 134501.968488094, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 355 |
+
{"event": "train", "epoch": 0, "step": 7080, "loss": 3.2685586810112, "perplexity": 26.27344362671799, "learning_rate": 0.00029543731233248004, "grad_norm": 0.5262284278869629, "tokens_per_sec": 135447.46933043463, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 356 |
+
{"event": "train", "epoch": 0, "step": 7100, "loss": 2.8895803689956665, "perplexity": 17.98576063497192, "learning_rate": 0.00029541163044274996, "grad_norm": 0.5334199070930481, "tokens_per_sec": 127574.15440822927, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 357 |
+
{"event": "train", "epoch": 0, "step": 7120, "loss": 2.6594191789627075, "perplexity": 14.28798792419563, "learning_rate": 0.0002953858776007679, "grad_norm": 0.49170982837677, "tokens_per_sec": 134906.75370495793, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 358 |
+
{"event": "train", "epoch": 0, "step": 7140, "loss": 2.348553717136383, "perplexity": 10.470415586146448, "learning_rate": 0.0002953600538190997, "grad_norm": 0.46368491649627686, "tokens_per_sec": 133976.33509155866, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 359 |
+
{"event": "train", "epoch": 0, "step": 7160, "loss": 2.038433402776718, "perplexity": 7.678570544186328, "learning_rate": 0.0002953341591103458, "grad_norm": 0.48968055844306946, "tokens_per_sec": 135336.23439815585, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 360 |
+
{"event": "train", "epoch": 0, "step": 7180, "loss": 2.2867568731307983, "perplexity": 9.842963881017566, "learning_rate": 0.0002953081934871414, "grad_norm": 0.4285159409046173, "tokens_per_sec": 133072.44726005578, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 361 |
+
{"event": "train", "epoch": 0, "step": 7200, "loss": 2.724501609802246, "perplexity": 15.248812174181996, "learning_rate": 0.000295282156962156, "grad_norm": 0.6686787605285645, "tokens_per_sec": 134451.44242139655, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 362 |
+
{"event": "train", "epoch": 0, "step": 7220, "loss": 2.4598798155784607, "perplexity": 11.703404888507185, "learning_rate": 0.0002952560495480941, "grad_norm": 0.564836323261261, "tokens_per_sec": 133532.72079070937, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 363 |
+
{"event": "train", "epoch": 0, "step": 7240, "loss": 3.2447729110717773, "perplexity": 25.6558832316235, "learning_rate": 0.0002952298712576947, "grad_norm": 0.9134393334388733, "tokens_per_sec": 134122.506259716, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 364 |
+
{"event": "train", "epoch": 0, "step": 7260, "loss": 3.616989314556122, "perplexity": 37.22532519850114, "learning_rate": 0.0002952036221037308, "grad_norm": 1.6203144788742065, "tokens_per_sec": 134409.49780106187, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 365 |
+
{"event": "train", "epoch": 0, "step": 7280, "loss": 2.5245394706726074, "perplexity": 12.4851441643891, "learning_rate": 0.00029517730209901056, "grad_norm": 0.490701824426651, "tokens_per_sec": 129876.87256679818, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 366 |
+
{"event": "train", "epoch": 0, "step": 7300, "loss": 2.8993363678455353, "perplexity": 18.162088423327045, "learning_rate": 0.00029515091125637693, "grad_norm": 0.5544854402542114, "tokens_per_sec": 127647.85258715488, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 367 |
+
{"event": "train", "epoch": 0, "step": 7320, "loss": 2.7973774075508118, "perplexity": 16.40157566830697, "learning_rate": 0.0002951244495887069, "grad_norm": 0.44284114241600037, "tokens_per_sec": 130768.82786889146, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 368 |
+
{"event": "train", "epoch": 0, "step": 7340, "loss": 2.249237596988678, "perplexity": 9.480505114698484, "learning_rate": 0.00029509791710891195, "grad_norm": 0.43650883436203003, "tokens_per_sec": 134711.71480575745, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 369 |
+
{"event": "train", "epoch": 0, "step": 7360, "loss": 2.773635685443878, "perplexity": 16.01676018338225, "learning_rate": 0.00029507131382993854, "grad_norm": 0.5346018671989441, "tokens_per_sec": 133620.4809868391, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 370 |
+
{"event": "train", "epoch": 0, "step": 7380, "loss": 3.3100183308124542, "perplexity": 27.385627468103262, "learning_rate": 0.0002950446397647676, "grad_norm": 0.6422619819641113, "tokens_per_sec": 134589.16254428728, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 371 |
+
{"event": "train", "epoch": 0, "step": 7400, "loss": 2.6244680881500244, "perplexity": 13.797233321993033, "learning_rate": 0.00029501789492641424, "grad_norm": 0.5490100383758545, "tokens_per_sec": 134289.72512802275, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 372 |
+
{"event": "train", "epoch": 0, "step": 7420, "loss": 3.118928015232086, "perplexity": 22.622116076559973, "learning_rate": 0.0002949910793279286, "grad_norm": 0.5458810329437256, "tokens_per_sec": 134369.6812137163, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 373 |
+
{"event": "train", "epoch": 0, "step": 7440, "loss": 2.8415797352790833, "perplexity": 17.142825283784184, "learning_rate": 0.0002949641929823949, "grad_norm": 0.5814038515090942, "tokens_per_sec": 133934.68633863592, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 374 |
+
{"event": "train", "epoch": 0, "step": 7460, "loss": 2.4952054023742676, "perplexity": 12.124223607160697, "learning_rate": 0.00029493723590293233, "grad_norm": 0.5592000484466553, "tokens_per_sec": 134317.54797461786, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 375 |
+
{"event": "train", "epoch": 0, "step": 7480, "loss": 2.909779369831085, "perplexity": 18.352748950199594, "learning_rate": 0.000294910208102694, "grad_norm": 0.6167746782302856, "tokens_per_sec": 134174.3571872764, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 376 |
+
{"event": "train", "epoch": 0, "step": 7500, "loss": 3.025047540664673, "perplexity": 20.59498378714105, "learning_rate": 0.0002948831095948682, "grad_norm": 0.5176410675048828, "tokens_per_sec": 135085.75954079928, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 377 |
+
{"event": "train", "epoch": 0, "step": 7520, "loss": 2.8269914388656616, "perplexity": 16.8945559832612, "learning_rate": 0.00029485594039267737, "grad_norm": 0.5081061124801636, "tokens_per_sec": 134979.62470978766, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 378 |
+
{"event": "train", "epoch": 0, "step": 7540, "loss": 2.3309198021888733, "perplexity": 10.287399555250598, "learning_rate": 0.00029482870050937833, "grad_norm": 0.49117016792297363, "tokens_per_sec": 133106.47094907844, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 379 |
+
{"event": "train", "epoch": 0, "step": 7560, "loss": 2.422105073928833, "perplexity": 11.269557614718238, "learning_rate": 0.00029480138995826254, "grad_norm": 0.4651361405849457, "tokens_per_sec": 134807.64308659278, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 380 |
+
{"event": "train", "epoch": 0, "step": 7580, "loss": 2.4463911056518555, "perplexity": 11.546600973545313, "learning_rate": 0.000294774008752656, "grad_norm": 0.5217791795730591, "tokens_per_sec": 134337.2409164794, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 381 |
+
{"event": "train", "epoch": 0, "step": 7600, "loss": 2.5241847336292267, "perplexity": 12.480716006724258, "learning_rate": 0.00029474655690591903, "grad_norm": 0.49159660935401917, "tokens_per_sec": 134831.18355094123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 382 |
+
{"event": "train", "epoch": 0, "step": 7620, "loss": 2.340264618396759, "perplexity": 10.383983992401637, "learning_rate": 0.00029471903443144644, "grad_norm": 0.5450993180274963, "tokens_per_sec": 134162.30743745505, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 383 |
+
{"event": "train", "epoch": 0, "step": 7640, "loss": 2.293130785226822, "perplexity": 9.905902436888494, "learning_rate": 0.0002946914413426678, "grad_norm": 0.44311031699180603, "tokens_per_sec": 134633.4614032785, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 384 |
+
{"event": "train", "epoch": 0, "step": 7660, "loss": 2.5720155239105225, "perplexity": 13.0921854792335, "learning_rate": 0.0002946637776530468, "grad_norm": 0.5294637680053711, "tokens_per_sec": 134303.37263117984, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 385 |
+
{"event": "train", "epoch": 0, "step": 7680, "loss": 2.8418617844581604, "perplexity": 17.147661085517438, "learning_rate": 0.00029463604337608135, "grad_norm": 0.4467974305152893, "tokens_per_sec": 134708.14985807706, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 386 |
+
{"event": "train", "epoch": 0, "step": 7700, "loss": 2.6919615864753723, "perplexity": 14.760601737658485, "learning_rate": 0.0002946082385253045, "grad_norm": 0.5872318744659424, "tokens_per_sec": 134812.93236251065, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 387 |
+
{"event": "train", "epoch": 0, "step": 7720, "loss": 2.9615346789360046, "perplexity": 19.327610683565602, "learning_rate": 0.0002945803631142834, "grad_norm": 0.7430490851402283, "tokens_per_sec": 134540.54647919862, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 388 |
+
{"event": "train", "epoch": 0, "step": 7740, "loss": 2.3296769559383392, "perplexity": 10.274621841297323, "learning_rate": 0.0002945524171566194, "grad_norm": 0.5000406503677368, "tokens_per_sec": 134675.9418490254, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 389 |
+
{"event": "train", "epoch": 0, "step": 7760, "loss": 2.828247219324112, "perplexity": 16.915785163324067, "learning_rate": 0.0002945244006659487, "grad_norm": 0.49270689487457275, "tokens_per_sec": 134116.22402185845, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 390 |
+
{"event": "train", "epoch": 0, "step": 7780, "loss": 2.697409987449646, "perplexity": 14.841242897747945, "learning_rate": 0.00029449631365594144, "grad_norm": 0.5628951787948608, "tokens_per_sec": 135329.17168542423, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 391 |
+
{"event": "train", "epoch": 0, "step": 7800, "loss": 2.9445497393608093, "perplexity": 19.00210456024161, "learning_rate": 0.00029446815614030244, "grad_norm": 0.5572422742843628, "tokens_per_sec": 134947.28676909543, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 392 |
+
{"event": "train", "epoch": 0, "step": 7820, "loss": 3.599599301815033, "perplexity": 36.58357253525963, "learning_rate": 0.0002944399281327712, "grad_norm": 0.5984124541282654, "tokens_per_sec": 135109.53006416376, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 393 |
+
{"event": "train", "epoch": 0, "step": 7840, "loss": 3.2988542914390564, "perplexity": 27.081593526019898, "learning_rate": 0.00029441162964712097, "grad_norm": 0.8210545778274536, "tokens_per_sec": 134975.51534150224, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 394 |
+
{"event": "train", "epoch": 0, "step": 7860, "loss": 3.319634795188904, "perplexity": 27.6502507097851, "learning_rate": 0.00029438326069716, "grad_norm": 0.7878577709197998, "tokens_per_sec": 135268.0367501961, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 395 |
+
{"event": "train", "epoch": 0, "step": 7880, "loss": 3.1673664450645447, "perplexity": 23.744868525805828, "learning_rate": 0.0002943548212967305, "grad_norm": 1.1468292474746704, "tokens_per_sec": 135237.8226702195, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 396 |
+
{"event": "train", "epoch": 0, "step": 7900, "loss": 2.3047143816947937, "perplexity": 10.02131557245777, "learning_rate": 0.0002943263114597091, "grad_norm": 0.44652536511421204, "tokens_per_sec": 134787.94418575458, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 397 |
+
{"event": "train", "epoch": 0, "step": 7920, "loss": 2.402333438396454, "perplexity": 11.04892831725311, "learning_rate": 0.00029429773120000707, "grad_norm": 0.5666265487670898, "tokens_per_sec": 134836.60695771608, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 398 |
+
{"event": "train", "epoch": 0, "step": 7940, "loss": 2.6710326075553894, "perplexity": 14.454887715254852, "learning_rate": 0.0002942690805315699, "grad_norm": 0.5165908336639404, "tokens_per_sec": 134905.1646691834, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 399 |
+
{"event": "train", "epoch": 0, "step": 7960, "loss": 3.044100761413574, "perplexity": 20.991146664245335, "learning_rate": 0.00029424035946837746, "grad_norm": 0.5363592505455017, "tokens_per_sec": 134845.73513953167, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 400 |
+
{"event": "train", "epoch": 0, "step": 7980, "loss": 2.4784480929374695, "perplexity": 11.922747057591039, "learning_rate": 0.0002942115680244438, "grad_norm": 0.590940535068512, "tokens_per_sec": 134712.24296220206, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 401 |
+
{"event": "train", "epoch": 0, "step": 8000, "loss": 2.5466060042381287, "perplexity": 12.763710207163454, "learning_rate": 0.00029418270621381744, "grad_norm": 0.5239678025245667, "tokens_per_sec": 133860.07062380202, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 402 |
+
{"event": "train", "epoch": 0, "step": 8020, "loss": 3.590390205383301, "perplexity": 36.248217416825305, "learning_rate": 0.0002941537740505812, "grad_norm": 0.5908644199371338, "tokens_per_sec": 134890.99742170147, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 403 |
+
{"event": "train", "epoch": 0, "step": 8040, "loss": 2.2353702187538147, "perplexity": 9.349942735662957, "learning_rate": 0.00029412477154885236, "grad_norm": 0.5089201927185059, "tokens_per_sec": 131845.15891472498, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 404 |
+
{"event": "train", "epoch": 0, "step": 8060, "loss": 2.2262316048145294, "perplexity": 9.264886458836036, "learning_rate": 0.00029409569872278235, "grad_norm": 0.48652294278144836, "tokens_per_sec": 134865.84795879415, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 405 |
+
{"event": "train", "epoch": 0, "step": 8080, "loss": 2.625719666481018, "perplexity": 13.814512451082704, "learning_rate": 0.00029406655558655707, "grad_norm": 0.640375018119812, "tokens_per_sec": 134966.50211966978, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 406 |
+
{"event": "train", "epoch": 0, "step": 8100, "loss": 2.4849225282669067, "perplexity": 12.000190543259642, "learning_rate": 0.00029403734215439637, "grad_norm": 0.4607117176055908, "tokens_per_sec": 134153.53343810092, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 407 |
+
{"event": "train", "epoch": 0, "step": 8120, "loss": 2.3537555932998657, "perplexity": 10.525023299548991, "learning_rate": 0.00029400805844055503, "grad_norm": 0.45522192120552063, "tokens_per_sec": 135257.12085548026, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 408 |
+
{"event": "train", "epoch": 0, "step": 8140, "loss": 2.7365848422050476, "perplexity": 15.434184810440641, "learning_rate": 0.0002939787044593217, "grad_norm": 0.45822247862815857, "tokens_per_sec": 134447.49667106871, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 409 |
+
{"event": "train", "epoch": 0, "step": 8160, "loss": 2.2167763710021973, "perplexity": 9.177697636041973, "learning_rate": 0.00029394928022501946, "grad_norm": 0.46568408608436584, "tokens_per_sec": 134877.36260147282, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 410 |
+
{"event": "train", "epoch": 0, "step": 8180, "loss": 2.1823671460151672, "perplexity": 8.867271561506923, "learning_rate": 0.0002939197857520051, "grad_norm": 0.4862081706523895, "tokens_per_sec": 135150.45088712825, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 411 |
+
{"event": "train", "epoch": 0, "step": 8200, "loss": 2.544679880142212, "perplexity": 12.739149378575313, "learning_rate": 0.00029389022105467066, "grad_norm": 0.527452290058136, "tokens_per_sec": 134870.3477000525, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 412 |
+
{"event": "train", "epoch": 0, "step": 8220, "loss": 2.635549545288086, "perplexity": 13.950977050653382, "learning_rate": 0.00029386058614744174, "grad_norm": 0.5054888129234314, "tokens_per_sec": 134243.42303018938, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 413 |
+
{"event": "train", "epoch": 0, "step": 8240, "loss": 2.2233987748622894, "perplexity": 9.238677750904671, "learning_rate": 0.0002938308810447785, "grad_norm": 0.5040379166603088, "tokens_per_sec": 134454.33611883412, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 414 |
+
{"event": "train", "epoch": 0, "step": 8260, "loss": 2.7090484499931335, "perplexity": 15.014981209607724, "learning_rate": 0.00029380110576117516, "grad_norm": 0.537735641002655, "tokens_per_sec": 134486.30661036933, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 415 |
+
{"event": "train", "epoch": 0, "step": 8280, "loss": 2.7649188935756683, "perplexity": 15.877752150685106, "learning_rate": 0.0002937712603111605, "grad_norm": 0.5832128524780273, "tokens_per_sec": 134139.2617716785, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 416 |
+
{"event": "train", "epoch": 0, "step": 8300, "loss": 2.7880536317825317, "perplexity": 16.24936176139439, "learning_rate": 0.00029374134470929743, "grad_norm": 0.5147441029548645, "tokens_per_sec": 134960.53814942398, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 417 |
+
{"event": "train", "epoch": 0, "step": 8320, "loss": 3.252035915851593, "perplexity": 25.84290036521805, "learning_rate": 0.0002937113589701825, "grad_norm": 0.5514777898788452, "tokens_per_sec": 134346.958228374, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 418 |
+
{"event": "train", "epoch": 0, "step": 8340, "loss": 2.3373461961746216, "perplexity": 10.353723320949886, "learning_rate": 0.00029368130310844737, "grad_norm": 0.49737513065338135, "tokens_per_sec": 135278.02234309146, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 419 |
+
{"event": "train", "epoch": 0, "step": 8360, "loss": 2.63423228263855, "perplexity": 13.932612048081543, "learning_rate": 0.00029365117713875735, "grad_norm": 0.45526123046875, "tokens_per_sec": 135142.87600861365, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 420 |
+
{"event": "train", "epoch": 0, "step": 8380, "loss": 2.2097485661506653, "perplexity": 9.113424681497161, "learning_rate": 0.0002936209810758122, "grad_norm": 0.47269517183303833, "tokens_per_sec": 134131.66891164557, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 421 |
+
{"event": "train", "epoch": 0, "step": 8400, "loss": 2.5631623566150665, "perplexity": 12.976789733555364, "learning_rate": 0.00029359071493434576, "grad_norm": 0.6967554092407227, "tokens_per_sec": 134889.54114437138, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 422 |
+
{"event": "train", "epoch": 0, "step": 8420, "loss": 2.526161849498749, "perplexity": 12.505416237967664, "learning_rate": 0.0002935603787291262, "grad_norm": 0.48178982734680176, "tokens_per_sec": 134466.43838505985, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 423 |
+
{"event": "train", "epoch": 0, "step": 8440, "loss": 2.578611135482788, "perplexity": 13.178821845123599, "learning_rate": 0.00029352997247495565, "grad_norm": 0.4507233500480652, "tokens_per_sec": 133248.29460662176, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 424 |
+
{"event": "train", "epoch": 0, "step": 8460, "loss": 3.137813091278076, "perplexity": 23.053396018454094, "learning_rate": 0.00029349949618667065, "grad_norm": 0.6049101948738098, "tokens_per_sec": 134793.6285124433, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 425 |
+
{"event": "train", "epoch": 0, "step": 8480, "loss": 2.35531485080719, "perplexity": 10.541447322456463, "learning_rate": 0.00029346894987914183, "grad_norm": 0.47761544585227966, "tokens_per_sec": 134722.14666240592, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 426 |
+
{"event": "train", "epoch": 0, "step": 8500, "loss": 2.726278603076935, "perplexity": 15.275933300755561, "learning_rate": 0.00029343833356727404, "grad_norm": 0.7445896863937378, "tokens_per_sec": 134525.27063530983, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 427 |
+
{"event": "train", "epoch": 0, "step": 8520, "loss": 3.168813943862915, "perplexity": 23.77926408222207, "learning_rate": 0.0002934076472660061, "grad_norm": 0.6050772666931152, "tokens_per_sec": 134882.65734990462, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 428 |
+
{"event": "train", "epoch": 0, "step": 8540, "loss": 2.6578986048698425, "perplexity": 14.26627848950386, "learning_rate": 0.00029337689099031113, "grad_norm": 0.7195141911506653, "tokens_per_sec": 135080.3160730485, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 429 |
+
{"event": "train", "epoch": 0, "step": 8560, "loss": 2.601744532585144, "perplexity": 13.487246464370006, "learning_rate": 0.0002933460647551964, "grad_norm": 0.5092797875404358, "tokens_per_sec": 134812.00670926328, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 430 |
+
{"event": "train", "epoch": 0, "step": 8580, "loss": 2.7348965406417847, "perplexity": 15.408149236232036, "learning_rate": 0.0002933151685757033, "grad_norm": 0.5889442563056946, "tokens_per_sec": 134991.29137675665, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 431 |
+
{"event": "train", "epoch": 0, "step": 8600, "loss": 2.6379841566085815, "perplexity": 13.984983636925472, "learning_rate": 0.00029328420246690716, "grad_norm": 0.5093480348587036, "tokens_per_sec": 134289.85634080804, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 432 |
+
{"event": "train", "epoch": 0, "step": 8620, "loss": 2.987229883670807, "perplexity": 19.830673064210618, "learning_rate": 0.00029325316644391767, "grad_norm": 0.8535952568054199, "tokens_per_sec": 134978.2990797757, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 433 |
+
{"event": "train", "epoch": 0, "step": 8640, "loss": 2.613136887550354, "perplexity": 13.641776522922259, "learning_rate": 0.0002932220605218788, "grad_norm": 0.6142433285713196, "tokens_per_sec": 133884.45507423417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 434 |
+
{"event": "train", "epoch": 0, "step": 8660, "loss": 2.877089202404022, "perplexity": 17.76249483077227, "learning_rate": 0.00029319088471596805, "grad_norm": 0.5173980593681335, "tokens_per_sec": 134270.57081253658, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 435 |
+
{"event": "train", "epoch": 0, "step": 8680, "loss": 2.6080304384231567, "perplexity": 13.572293043017527, "learning_rate": 0.0002931596390413976, "grad_norm": 0.9497238993644714, "tokens_per_sec": 134492.75517709064, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 436 |
+
{"event": "train", "epoch": 0, "step": 8700, "loss": 3.1955469250679016, "perplexity": 24.42352787973632, "learning_rate": 0.0002931283235134135, "grad_norm": 0.5306492447853088, "tokens_per_sec": 134992.8824421483, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 437 |
+
{"event": "train", "epoch": 0, "step": 8720, "loss": 2.7364795804023743, "perplexity": 15.432560265827448, "learning_rate": 0.00029309693814729576, "grad_norm": 0.7610920071601868, "tokens_per_sec": 134331.20113377023, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|