Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +87 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -1380,3 +1380,90 @@
|
|
| 1380 |
{"event": "train", "epoch": 0, "step": 27580, "loss": 2.488200694322586, "perplexity": 12.039593710335884, "learning_rate": 0.00023562579072735926, "grad_norm": 0.5174441933631897, "tokens_per_sec": 134356.020231762, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1381 |
{"event": "train", "epoch": 0, "step": 27600, "loss": 2.7255293130874634, "perplexity": 15.264491484008166, "learning_rate": 0.00023553973965161122, "grad_norm": 0.8017266392707825, "tokens_per_sec": 133591.90731699255, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1382 |
{"event": "train", "epoch": 0, "step": 27620, "loss": 2.438509523868561, "perplexity": 11.455953186873153, "learning_rate": 0.0002354536468375433, "grad_norm": 0.6045737862586975, "tokens_per_sec": 134000.63127413692, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1380 |
{"event": "train", "epoch": 0, "step": 27580, "loss": 2.488200694322586, "perplexity": 12.039593710335884, "learning_rate": 0.00023562579072735926, "grad_norm": 0.5174441933631897, "tokens_per_sec": 134356.020231762, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1381 |
{"event": "train", "epoch": 0, "step": 27600, "loss": 2.7255293130874634, "perplexity": 15.264491484008166, "learning_rate": 0.00023553973965161122, "grad_norm": 0.8017266392707825, "tokens_per_sec": 133591.90731699255, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1382 |
{"event": "train", "epoch": 0, "step": 27620, "loss": 2.438509523868561, "perplexity": 11.455953186873153, "learning_rate": 0.0002354536468375433, "grad_norm": 0.6045737862586975, "tokens_per_sec": 134000.63127413692, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1383 |
+
{"event": "train", "epoch": 0, "step": 27640, "loss": 2.2154150009155273, "perplexity": 9.16521189380211, "learning_rate": 0.00023536751232716347, "grad_norm": 0.5105199217796326, "tokens_per_sec": 134181.16887293404, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1384 |
+
{"event": "train", "epoch": 0, "step": 27660, "loss": 2.92266309261322, "perplexity": 18.5907304318063, "learning_rate": 0.00023528133616250042, "grad_norm": 0.6390305161476135, "tokens_per_sec": 133852.63969660876, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1385 |
+
{"event": "train", "epoch": 0, "step": 27680, "loss": 2.2538874745368958, "perplexity": 9.52469095233113, "learning_rate": 0.0002351951183856029, "grad_norm": 0.4750637114048004, "tokens_per_sec": 134295.36750934133, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1386 |
+
{"event": "train", "epoch": 0, "step": 27700, "loss": 2.3390634059906006, "perplexity": 10.371518110589486, "learning_rate": 0.00023510885903854027, "grad_norm": 0.5861057639122009, "tokens_per_sec": 133996.84257980983, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1387 |
+
{"event": "train", "epoch": 0, "step": 27720, "loss": 1.6399566233158112, "perplexity": 5.154945902924559, "learning_rate": 0.00023502255816340174, "grad_norm": 0.42753756046295166, "tokens_per_sec": 133978.16352986018, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1388 |
+
{"event": "train", "epoch": 0, "step": 27740, "loss": 2.236361086368561, "perplexity": 9.359211882609834, "learning_rate": 0.00023493621580229714, "grad_norm": 0.5534217953681946, "tokens_per_sec": 134090.44683366962, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1389 |
+
{"event": "train", "epoch": 0, "step": 27760, "loss": 2.528014361858368, "perplexity": 12.528604147423751, "learning_rate": 0.00023484983199735643, "grad_norm": 0.5278375148773193, "tokens_per_sec": 134708.54595405722, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1390 |
+
{"event": "train", "epoch": 0, "step": 27780, "loss": 2.7659162878990173, "perplexity": 15.893596430731746, "learning_rate": 0.00023476340679072977, "grad_norm": 0.616127073764801, "tokens_per_sec": 134771.4227024504, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1391 |
+
{"event": "train", "epoch": 0, "step": 27800, "loss": 2.052763879299164, "perplexity": 7.789400344335807, "learning_rate": 0.00023467694022458757, "grad_norm": 0.4841892123222351, "tokens_per_sec": 134868.4948290531, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1392 |
+
{"event": "train", "epoch": 0, "step": 27820, "loss": 2.9308100938796997, "perplexity": 18.742807782202597, "learning_rate": 0.00023459043234112028, "grad_norm": 0.7197891473770142, "tokens_per_sec": 134135.0725012224, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1393 |
+
{"event": "train", "epoch": 0, "step": 27840, "loss": 2.828152298927307, "perplexity": 16.914179586486302, "learning_rate": 0.0002345038831825386, "grad_norm": 0.6360902190208435, "tokens_per_sec": 134491.4390928413, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1394 |
+
{"event": "train", "epoch": 0, "step": 27860, "loss": 2.5350935459136963, "perplexity": 12.617611118792059, "learning_rate": 0.00023441729279107356, "grad_norm": 0.4692329466342926, "tokens_per_sec": 133864.1123516492, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1395 |
+
{"event": "train", "epoch": 0, "step": 27880, "loss": 1.927246630191803, "perplexity": 6.870566962982027, "learning_rate": 0.00023433066120897607, "grad_norm": 0.46477967500686646, "tokens_per_sec": 134804.07305996984, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1396 |
+
{"event": "train", "epoch": 0, "step": 27900, "loss": 2.1841831505298615, "perplexity": 8.883389197115664, "learning_rate": 0.00023424398847851698, "grad_norm": 0.6935025453567505, "tokens_per_sec": 134870.2153504356, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1397 |
+
{"event": "train", "epoch": 0, "step": 27920, "loss": 1.834309309720993, "perplexity": 6.260808372241503, "learning_rate": 0.0002341572746419877, "grad_norm": 0.494891494512558, "tokens_per_sec": 134431.453005364, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1398 |
+
{"event": "train", "epoch": 0, "step": 27940, "loss": 2.767745018005371, "perplexity": 15.922688121340594, "learning_rate": 0.00023407051974169945, "grad_norm": 0.5883350372314453, "tokens_per_sec": 133966.14899168845, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1399 |
+
{"event": "train", "epoch": 0, "step": 27960, "loss": 1.9772624671459198, "perplexity": 7.222942852407762, "learning_rate": 0.0002339837238199834, "grad_norm": 0.4658470153808594, "tokens_per_sec": 134445.39236558782, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1400 |
+
{"event": "train", "epoch": 0, "step": 27980, "loss": 2.7654828429222107, "perplexity": 15.886708923980887, "learning_rate": 0.0002338968869191907, "grad_norm": 0.8604710698127747, "tokens_per_sec": 131975.9413555292, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1401 |
+
{"event": "train", "epoch": 0, "step": 28000, "loss": 2.6701542735099792, "perplexity": 14.442197069380999, "learning_rate": 0.00023381000908169282, "grad_norm": 0.5738292932510376, "tokens_per_sec": 133615.80445901424, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1402 |
+
{"event": "train", "epoch": 0, "step": 28020, "loss": 2.5217655897140503, "perplexity": 12.450559849288727, "learning_rate": 0.00023372309034988095, "grad_norm": 0.6270379424095154, "tokens_per_sec": 134578.61949328962, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1403 |
+
{"event": "train", "epoch": 0, "step": 28040, "loss": 2.5894041657447815, "perplexity": 13.321831634347967, "learning_rate": 0.00023363613076616622, "grad_norm": 0.6103567481040955, "tokens_per_sec": 133970.58889606965, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1404 |
+
{"event": "train", "epoch": 0, "step": 28060, "loss": 2.5439032912254333, "perplexity": 12.729260136793515, "learning_rate": 0.0002335491303729797, "grad_norm": 0.5052766799926758, "tokens_per_sec": 134275.9492123538, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1405 |
+
{"event": "train", "epoch": 0, "step": 28080, "loss": 2.250934511423111, "perplexity": 9.496606378042767, "learning_rate": 0.0002334620892127724, "grad_norm": 0.502346932888031, "tokens_per_sec": 135036.9167960322, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1406 |
+
{"event": "train", "epoch": 0, "step": 28100, "loss": 2.1962429583072662, "perplexity": 8.991169763412604, "learning_rate": 0.0002333750073280153, "grad_norm": 0.4877512454986572, "tokens_per_sec": 132592.9464646114, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1407 |
+
{"event": "train", "epoch": 0, "step": 28120, "loss": 2.614003688097, "perplexity": 13.653606348578537, "learning_rate": 0.00023328788476119927, "grad_norm": 0.5996014475822449, "tokens_per_sec": 132859.5504858536, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1408 |
+
{"event": "train", "epoch": 0, "step": 28140, "loss": 2.680246204137802, "perplexity": 14.588684648285236, "learning_rate": 0.00023320072155483496, "grad_norm": 0.6157772541046143, "tokens_per_sec": 134727.29716202003, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1409 |
+
{"event": "train", "epoch": 0, "step": 28160, "loss": 2.3548507690429688, "perplexity": 10.536556363975714, "learning_rate": 0.00023311351775145276, "grad_norm": 0.5583117008209229, "tokens_per_sec": 133694.95981727698, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1410 |
+
{"event": "train", "epoch": 0, "step": 28180, "loss": 2.1423847675323486, "perplexity": 8.519730999655017, "learning_rate": 0.00023302627339360295, "grad_norm": 0.4545435309410095, "tokens_per_sec": 134828.53814497514, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1411 |
+
{"event": "train", "epoch": 0, "step": 28200, "loss": 2.8122612833976746, "perplexity": 16.647520449718893, "learning_rate": 0.00023293898852385552, "grad_norm": 0.6272359490394592, "tokens_per_sec": 134163.87902121316, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1412 |
+
{"event": "train", "epoch": 0, "step": 28220, "loss": 2.4195228815078735, "perplexity": 11.24049498725458, "learning_rate": 0.00023285166318480052, "grad_norm": 0.514761209487915, "tokens_per_sec": 134975.51534150224, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1413 |
+
{"event": "train", "epoch": 0, "step": 28240, "loss": 2.443494439125061, "perplexity": 11.513202716166923, "learning_rate": 0.0002327642974190474, "grad_norm": 0.6522575616836548, "tokens_per_sec": 134231.88526914947, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1414 |
+
{"event": "train", "epoch": 0, "step": 28260, "loss": 2.1567131876945496, "perplexity": 8.642684042967819, "learning_rate": 0.00023267689126922562, "grad_norm": 0.44610223174095154, "tokens_per_sec": 134676.60169464123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1415 |
+
{"event": "train", "epoch": 0, "step": 28280, "loss": 2.219123274087906, "perplexity": 9.199262097896732, "learning_rate": 0.00023258944477798398, "grad_norm": 0.4858779013156891, "tokens_per_sec": 134090.05436425278, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1416 |
+
{"event": "train", "epoch": 0, "step": 28300, "loss": 1.9557819664478302, "perplexity": 7.069444931623651, "learning_rate": 0.00023250195798799132, "grad_norm": 0.4873845875263214, "tokens_per_sec": 133954.91995879178, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1417 |
+
{"event": "train", "epoch": 0, "step": 28320, "loss": 2.446384370326996, "perplexity": 11.546523203698635, "learning_rate": 0.00023241443094193596, "grad_norm": 0.5340095162391663, "tokens_per_sec": 135047.92987358812, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1418 |
+
{"event": "train", "epoch": 0, "step": 28340, "loss": 1.9740183055400848, "perplexity": 7.199548426691737, "learning_rate": 0.0002323268636825259, "grad_norm": 0.4784976840019226, "tokens_per_sec": 133725.3991379394, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1419 |
+
{"event": "train", "epoch": 0, "step": 28360, "loss": 3.5331227779388428, "perplexity": 34.23069574540134, "learning_rate": 0.00023223925625248876, "grad_norm": 1.4685428142547607, "tokens_per_sec": 134561.75404696775, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1420 |
+
{"event": "train", "epoch": 0, "step": 28380, "loss": 2.6175360679626465, "perplexity": 13.701921355979024, "learning_rate": 0.00023215160869457185, "grad_norm": 0.6509571075439453, "tokens_per_sec": 134933.1106736791, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1421 |
+
{"event": "train", "epoch": 0, "step": 28400, "loss": 2.0401431918144226, "perplexity": 7.6917105100116885, "learning_rate": 0.0002320639210515421, "grad_norm": 0.47923922538757324, "tokens_per_sec": 134087.17632539413, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1422 |
+
{"event": "train", "epoch": 0, "step": 28420, "loss": 2.0051503479480743, "perplexity": 7.4272104786732065, "learning_rate": 0.00023197619336618562, "grad_norm": 0.4992538392543793, "tokens_per_sec": 134190.47056066737, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1423 |
+
{"event": "train", "epoch": 0, "step": 28440, "loss": 2.8721237778663635, "perplexity": 17.67451511218109, "learning_rate": 0.00023188842568130834, "grad_norm": 1.2495945692062378, "tokens_per_sec": 133891.1069208123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1424 |
+
{"event": "train", "epoch": 0, "step": 28460, "loss": 2.4828308522701263, "perplexity": 11.97511626551785, "learning_rate": 0.0002318006180397357, "grad_norm": 0.5249966382980347, "tokens_per_sec": 134154.0572267459, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1425 |
+
{"event": "train", "epoch": 0, "step": 28480, "loss": 2.7949913144111633, "perplexity": 16.36248663470877, "learning_rate": 0.00023171277048431274, "grad_norm": 0.5322928428649902, "tokens_per_sec": 134200.95288088988, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1426 |
+
{"event": "train", "epoch": 0, "step": 28500, "loss": 2.1728314757347107, "perplexity": 8.783118051337382, "learning_rate": 0.00023162488305790375, "grad_norm": 0.4957251250743866, "tokens_per_sec": 135373.9590998903, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1427 |
+
{"event": "train", "epoch": 0, "step": 28520, "loss": 2.426966905593872, "perplexity": 11.32448171445553, "learning_rate": 0.0002315369558033927, "grad_norm": 0.47933048009872437, "tokens_per_sec": 134490.12303434886, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1428 |
+
{"event": "train", "epoch": 0, "step": 28540, "loss": 2.657597303390503, "perplexity": 14.261980686190062, "learning_rate": 0.00023144898876368273, "grad_norm": 0.5545516014099121, "tokens_per_sec": 133573.99006934353, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1429 |
+
{"event": "train", "epoch": 0, "step": 28560, "loss": 2.018953263759613, "perplexity": 7.5304384240298425, "learning_rate": 0.00023136098198169665, "grad_norm": 0.5015084147453308, "tokens_per_sec": 132581.30696607014, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1430 |
+
{"event": "train", "epoch": 0, "step": 28580, "loss": 3.629830300807953, "perplexity": 37.70641732566465, "learning_rate": 0.00023127293550037641, "grad_norm": 0.780419111251831, "tokens_per_sec": 134417.5165353191, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1431 |
+
{"event": "train", "epoch": 0, "step": 28600, "loss": 2.1998548209667206, "perplexity": 9.023703351804224, "learning_rate": 0.00023118484936268354, "grad_norm": 0.5396973490715027, "tokens_per_sec": 134988.1093584873, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1432 |
+
{"event": "train", "epoch": 0, "step": 28620, "loss": 2.3695948123931885, "perplexity": 10.693058711346607, "learning_rate": 0.00023109672361159902, "grad_norm": 0.5067847967147827, "tokens_per_sec": 134659.71168258318, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1433 |
+
{"event": "train", "epoch": 0, "step": 28640, "loss": 1.8245516419410706, "perplexity": 6.200014569335271, "learning_rate": 0.00023100855829012279, "grad_norm": 0.4644889533519745, "tokens_per_sec": 134433.16239403185, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1434 |
+
{"event": "train", "epoch": 0, "step": 28660, "loss": 2.4303001165390015, "perplexity": 11.362291579976308, "learning_rate": 0.00023092035344127442, "grad_norm": 0.5231790542602539, "tokens_per_sec": 134621.85650282196, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1435 |
+
{"event": "train", "epoch": 0, "step": 28680, "loss": 2.1175650358200073, "perplexity": 8.310876144160318, "learning_rate": 0.0002308321091080926, "grad_norm": 0.673210084438324, "tokens_per_sec": 134780.01336827077, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1436 |
+
{"event": "train", "epoch": 0, "step": 28700, "loss": 2.7710460424423218, "perplexity": 15.97533615234178, "learning_rate": 0.00023074382533363528, "grad_norm": 0.5912601947784424, "tokens_per_sec": 134869.68595456553, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1437 |
+
{"event": "train", "epoch": 0, "step": 28720, "loss": 2.652205526828766, "perplexity": 14.185290207754996, "learning_rate": 0.0002306555021609797, "grad_norm": 0.536240816116333, "tokens_per_sec": 134315.57899798095, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1438 |
+
{"event": "train", "epoch": 0, "step": 28740, "loss": 2.296539545059204, "perplexity": 9.939726896195635, "learning_rate": 0.0002305671396332223, "grad_norm": 0.5616986155509949, "tokens_per_sec": 135343.96424938377, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1439 |
+
{"event": "train", "epoch": 0, "step": 28760, "loss": 2.216973900794983, "perplexity": 9.17951068381385, "learning_rate": 0.00023047873779347895, "grad_norm": 0.49697941541671753, "tokens_per_sec": 134308.35991083828, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1440 |
+
{"event": "train", "epoch": 0, "step": 28780, "loss": 2.2262399196624756, "perplexity": 9.264963495278453, "learning_rate": 0.0002303902966848843, "grad_norm": 0.5492908358573914, "tokens_per_sec": 133910.54453407176, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1441 |
+
{"event": "train", "epoch": 0, "step": 28800, "loss": 2.168167382478714, "perplexity": 8.742248154262663, "learning_rate": 0.00023030181635059257, "grad_norm": 0.5061967968940735, "tokens_per_sec": 134928.20935237108, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1442 |
+
{"event": "train", "epoch": 0, "step": 28820, "loss": 2.9357746243476868, "perplexity": 18.836088378086043, "learning_rate": 0.00023021329683377674, "grad_norm": 0.647043764591217, "tokens_per_sec": 135349.56227909756, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1443 |
+
{"event": "train", "epoch": 0, "step": 28840, "loss": 2.0530540347099304, "perplexity": 7.791660808919499, "learning_rate": 0.00023012473817762922, "grad_norm": 0.45762941241264343, "tokens_per_sec": 133531.16395874726, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1444 |
+
{"event": "train", "epoch": 0, "step": 28860, "loss": 2.004845768213272, "perplexity": 7.424948645347074, "learning_rate": 0.0002300361404253612, "grad_norm": 0.46982327103614807, "tokens_per_sec": 134990.76102996065, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1445 |
+
{"event": "train", "epoch": 0, "step": 28880, "loss": 2.2018361687660217, "perplexity": 9.04160017063832, "learning_rate": 0.00022994750362020317, "grad_norm": 0.5469425320625305, "tokens_per_sec": 134492.36034911132, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1446 |
+
{"event": "train", "epoch": 0, "step": 28900, "loss": 2.1553778648376465, "perplexity": 8.631150971321048, "learning_rate": 0.00022985882780540484, "grad_norm": 0.5133631229400635, "tokens_per_sec": 134696.92810653526, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1447 |
+
{"event": "train", "epoch": 0, "step": 28920, "loss": 2.369564414024353, "perplexity": 10.69273366474439, "learning_rate": 0.00022977011302423457, "grad_norm": 0.5439782738685608, "tokens_per_sec": 134082.2054583502, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1448 |
+
{"event": "train", "epoch": 0, "step": 28940, "loss": 2.8058557510375977, "perplexity": 16.54122502161442, "learning_rate": 0.00022968135931997983, "grad_norm": 0.6047022342681885, "tokens_per_sec": 134472.09562619918, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1449 |
+
{"event": "train", "epoch": 0, "step": 28960, "loss": 2.0186552703380585, "perplexity": 7.528194737236615, "learning_rate": 0.00022959256673594735, "grad_norm": 0.5285494923591614, "tokens_per_sec": 134378.48351005843, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1450 |
+
{"event": "train", "epoch": 0, "step": 28980, "loss": 2.0074075758457184, "perplexity": 7.443994320719342, "learning_rate": 0.00022950373531546253, "grad_norm": 0.47101029753685, "tokens_per_sec": 134570.31800312732, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1451 |
+
{"event": "train", "epoch": 0, "step": 29000, "loss": 2.7014829516410828, "perplexity": 14.901814016891851, "learning_rate": 0.00022941486510187, "grad_norm": 0.534988522529602, "tokens_per_sec": 132711.1198281231, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1452 |
+
{"event": "train", "epoch": 0, "step": 29020, "loss": 2.196074366569519, "perplexity": 8.98965405424942, "learning_rate": 0.00022932595613853294, "grad_norm": 0.5309546589851379, "tokens_per_sec": 134817.95755909366, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1453 |
+
{"event": "train", "epoch": 0, "step": 29040, "loss": 3.06206876039505, "perplexity": 21.371724437609963, "learning_rate": 0.00022923700846883365, "grad_norm": 0.7975141406059265, "tokens_per_sec": 134907.15096975068, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1454 |
+
{"event": "train", "epoch": 0, "step": 29060, "loss": 2.3615682125091553, "perplexity": 10.60757334490869, "learning_rate": 0.00022914802213617355, "grad_norm": 0.5179206132888794, "tokens_per_sec": 133915.3721987076, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1455 |
+
{"event": "train", "epoch": 0, "step": 29080, "loss": 2.2465667128562927, "perplexity": 9.455217569105423, "learning_rate": 0.00022905899718397247, "grad_norm": 0.45268553495407104, "tokens_per_sec": 133753.37911060723, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1456 |
+
{"event": "train", "epoch": 0, "step": 29100, "loss": 2.342695116996765, "perplexity": 10.40925294659723, "learning_rate": 0.00022896993365566942, "grad_norm": 0.5007821917533875, "tokens_per_sec": 134448.68037184968, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1457 |
+
{"event": "train", "epoch": 0, "step": 29120, "loss": 2.4044233858585358, "perplexity": 11.072044143966234, "learning_rate": 0.00022888083159472203, "grad_norm": 1.1751930713653564, "tokens_per_sec": 134737.59934277667, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1458 |
+
{"event": "train", "epoch": 0, "step": 29140, "loss": 2.126679450273514, "perplexity": 8.386971167818443, "learning_rate": 0.00022879169104460693, "grad_norm": 0.5025606751441956, "tokens_per_sec": 134156.93813740648, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1459 |
+
{"event": "train", "epoch": 0, "step": 29160, "loss": 2.11116486787796, "perplexity": 8.257854994212446, "learning_rate": 0.00022870251204881942, "grad_norm": 0.4665128290653229, "tokens_per_sec": 135031.74259672812, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1460 |
+
{"event": "train", "epoch": 0, "step": 29180, "loss": 2.7341710329055786, "perplexity": 15.396974558908116, "learning_rate": 0.0002286132946508734, "grad_norm": 0.8170921802520752, "tokens_per_sec": 134293.26796322933, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1461 |
+
{"event": "train", "epoch": 0, "step": 29200, "loss": 3.0257920622825623, "perplexity": 20.61032290723536, "learning_rate": 0.0002285240388943019, "grad_norm": 0.9345505833625793, "tokens_per_sec": 133588.5312361796, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1462 |
+
{"event": "train", "epoch": 0, "step": 29220, "loss": 2.388871043920517, "perplexity": 10.901180039942341, "learning_rate": 0.00022843474482265637, "grad_norm": 1.1859937906265259, "tokens_per_sec": 134278.04821697247, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1463 |
+
{"event": "train", "epoch": 0, "step": 29240, "loss": 2.170736163854599, "perplexity": 8.764733946682462, "learning_rate": 0.0002283454124795068, "grad_norm": 0.4780159890651703, "tokens_per_sec": 134390.0464873417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1464 |
+
{"event": "train", "epoch": 0, "step": 29260, "loss": 2.9996771812438965, "perplexity": 20.07905398160632, "learning_rate": 0.00022825604190844232, "grad_norm": 0.834141731262207, "tokens_per_sec": 133985.08592697445, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1465 |
+
{"event": "train", "epoch": 0, "step": 29280, "loss": 2.345316231250763, "perplexity": 10.436572576163227, "learning_rate": 0.00022816663315307044, "grad_norm": 0.608173668384552, "tokens_per_sec": 133927.11661999061, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1466 |
+
{"event": "train", "epoch": 0, "step": 29300, "loss": 2.802604854106903, "perplexity": 16.48753851589145, "learning_rate": 0.00022807718625701717, "grad_norm": 0.5954622030258179, "tokens_per_sec": 134400.56001181295, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1467 |
+
{"event": "train", "epoch": 0, "step": 29320, "loss": 2.7037989497184753, "perplexity": 14.936366585901776, "learning_rate": 0.00022798770126392746, "grad_norm": 0.7145785689353943, "tokens_per_sec": 133596.45231064133, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1468 |
+
{"event": "train", "epoch": 0, "step": 29340, "loss": 2.052047997713089, "perplexity": 7.783826051567568, "learning_rate": 0.00022789817821746464, "grad_norm": 0.5364621877670288, "tokens_per_sec": 134002.98298527458, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 1469 |
+
{"event": "train", "epoch": 0, "step": 29360, "loss": 2.2093048989772797, "perplexity": 9.109382250942037, "learning_rate": 0.0002278086171613106, "grad_norm": 0.6387125253677368, "tokens_per_sec": 134636.49483207046, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|