Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +94 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -254,3 +254,97 @@
|
|
| 254 |
{"event": "train", "epoch": 0, "step": 5060, "loss": 2.959497392177582, "perplexity": 19.288274881008256, "learning_rate": 0.0002976636455865192, "grad_norm": 0.576752245426178, "tokens_per_sec": 135255.12423608417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 255 |
{"event": "train", "epoch": 0, "step": 5080, "loss": 2.556455075740814, "perplexity": 12.890042005867912, "learning_rate": 0.000297645188452761, "grad_norm": 0.4632514715194702, "tokens_per_sec": 134223.6264992114, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 256 |
{"event": "train", "epoch": 0, "step": 5100, "loss": 2.8122610449790955, "perplexity": 16.647516480641194, "learning_rate": 0.0002976266592769068, "grad_norm": 0.6443740725517273, "tokens_per_sec": 134261.7826474312, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 254 |
{"event": "train", "epoch": 0, "step": 5060, "loss": 2.959497392177582, "perplexity": 19.288274881008256, "learning_rate": 0.0002976636455865192, "grad_norm": 0.576752245426178, "tokens_per_sec": 135255.12423608417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 255 |
{"event": "train", "epoch": 0, "step": 5080, "loss": 2.556455075740814, "perplexity": 12.890042005867912, "learning_rate": 0.000297645188452761, "grad_norm": 0.4632514715194702, "tokens_per_sec": 134223.6264992114, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 256 |
{"event": "train", "epoch": 0, "step": 5100, "loss": 2.8122610449790955, "perplexity": 16.647516480641194, "learning_rate": 0.0002976266592769068, "grad_norm": 0.6443740725517273, "tokens_per_sec": 134261.7826474312, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 257 |
+
{"event": "train", "epoch": 0, "step": 5120, "loss": 2.4498719573020935, "perplexity": 11.586863011036211, "learning_rate": 0.000297608058067998, "grad_norm": 0.511711835861206, "tokens_per_sec": 134235.16284048613, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 258 |
+
{"event": "train", "epoch": 0, "step": 5140, "loss": 2.988963782787323, "perplexity": 19.86508727748164, "learning_rate": 0.0002975893848351109, "grad_norm": 0.6102117300033569, "tokens_per_sec": 134339.07921902108, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 259 |
+
{"event": "train", "epoch": 0, "step": 5160, "loss": 2.424449384212494, "perplexity": 11.296007946305116, "learning_rate": 0.0002975706395873567, "grad_norm": 0.48475968837738037, "tokens_per_sec": 134930.19633160645, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 260 |
+
{"event": "train", "epoch": 0, "step": 5180, "loss": 3.142603635787964, "perplexity": 23.164099290948077, "learning_rate": 0.0002975518223338822, "grad_norm": 0.5839901566505432, "tokens_per_sec": 134094.37165420092, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 261 |
+
{"event": "train", "epoch": 0, "step": 5200, "loss": 2.9571946263313293, "perplexity": 19.24390960140921, "learning_rate": 0.00029753293308386905, "grad_norm": 0.5119471549987793, "tokens_per_sec": 134054.74152227028, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 262 |
+
{"event": "train", "epoch": 0, "step": 5220, "loss": 2.8433881998062134, "perplexity": 17.17385552529307, "learning_rate": 0.000297513971846534, "grad_norm": 0.4762398898601532, "tokens_per_sec": 133699.77194987776, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 263 |
+
{"event": "train", "epoch": 0, "step": 5240, "loss": 3.597591459751129, "perplexity": 36.51019219219749, "learning_rate": 0.00029749493863112897, "grad_norm": 1.3235771656036377, "tokens_per_sec": 134913.50752464615, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 264 |
+
{"event": "train", "epoch": 0, "step": 5260, "loss": 2.650114119052887, "perplexity": 14.155653983029186, "learning_rate": 0.0002974758334469412, "grad_norm": 0.5243156552314758, "tokens_per_sec": 134692.04392812971, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 265 |
+
{"event": "train", "epoch": 0, "step": 5280, "loss": 2.3612019419670105, "perplexity": 10.603688814706729, "learning_rate": 0.00029745665630329283, "grad_norm": 0.5233219861984253, "tokens_per_sec": 134932.0508983594, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 266 |
+
{"event": "train", "epoch": 0, "step": 5300, "loss": 2.4568322002887726, "perplexity": 11.667791708028707, "learning_rate": 0.00029743740720954116, "grad_norm": 0.45642051100730896, "tokens_per_sec": 135088.14949449676, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 267 |
+
{"event": "train", "epoch": 0, "step": 5320, "loss": 2.806243419647217, "perplexity": 16.54763877844572, "learning_rate": 0.00029741808617507844, "grad_norm": 0.5300256013870239, "tokens_per_sec": 134334.4835569978, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 268 |
+
{"event": "train", "epoch": 0, "step": 5340, "loss": 3.1043291091918945, "perplexity": 22.294256944834846, "learning_rate": 0.0002973986932093325, "grad_norm": 0.5303938984870911, "tokens_per_sec": 134673.698422295, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 269 |
+
{"event": "train", "epoch": 0, "step": 5360, "loss": 2.7469165325164795, "perplexity": 15.594472624329763, "learning_rate": 0.0002973792283217657, "grad_norm": 1.1522241830825806, "tokens_per_sec": 133908.06559830392, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 270 |
+
{"event": "train", "epoch": 0, "step": 5380, "loss": 2.4340319335460663, "perplexity": 11.40477278968307, "learning_rate": 0.00029735969152187597, "grad_norm": 0.46629437804222107, "tokens_per_sec": 134402.13718236447, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 271 |
+
{"event": "train", "epoch": 0, "step": 5400, "loss": 2.861873507499695, "perplexity": 17.49427190264356, "learning_rate": 0.00029734008281919605, "grad_norm": 0.5416823625564575, "tokens_per_sec": 134629.50496244372, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 272 |
+
{"event": "train", "epoch": 0, "step": 5420, "loss": 2.940413534641266, "perplexity": 18.92367028760494, "learning_rate": 0.0002973204022232936, "grad_norm": 0.7814084887504578, "tokens_per_sec": 134202.1322443974, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 273 |
+
{"event": "train", "epoch": 0, "step": 5440, "loss": 3.469117224216461, "perplexity": 32.10838542504778, "learning_rate": 0.000297300649743772, "grad_norm": 0.7917582988739014, "tokens_per_sec": 134712.37500196032, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 274 |
+
{"event": "train", "epoch": 0, "step": 5460, "loss": 2.090118706226349, "perplexity": 8.08587495103977, "learning_rate": 0.0002972808253902689, "grad_norm": 0.44121894240379333, "tokens_per_sec": 134839.3849691792, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 275 |
+
{"event": "train", "epoch": 0, "step": 5480, "loss": 2.1968494057655334, "perplexity": 8.996624089176308, "learning_rate": 0.0002972609291724576, "grad_norm": 0.44608959555625916, "tokens_per_sec": 134130.49078626596, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 276 |
+
{"event": "train", "epoch": 0, "step": 5500, "loss": 2.534905970096588, "perplexity": 12.61524458203597, "learning_rate": 0.0002972409611000462, "grad_norm": 0.45778605341911316, "tokens_per_sec": 134829.33175586522, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 277 |
+
{"event": "train", "epoch": 0, "step": 5520, "loss": 3.062117338180542, "perplexity": 21.372762653872208, "learning_rate": 0.0002972209211827778, "grad_norm": 0.6406384706497192, "tokens_per_sec": 134527.2457646695, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 278 |
+
{"event": "train", "epoch": 0, "step": 5540, "loss": 2.5919994115829468, "perplexity": 13.35644996456205, "learning_rate": 0.0002972008094304311, "grad_norm": 0.5069247484207153, "tokens_per_sec": 133967.45481536933, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 279 |
+
{"event": "train", "epoch": 0, "step": 5560, "loss": 2.4703235030174255, "perplexity": 11.826272067577309, "learning_rate": 0.0002971806258528191, "grad_norm": 0.5301553010940552, "tokens_per_sec": 134239.88282441144, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 280 |
+
{"event": "train", "epoch": 0, "step": 5580, "loss": 2.5781930685043335, "perplexity": 13.17331336643197, "learning_rate": 0.0002971603704597902, "grad_norm": 0.4883482754230499, "tokens_per_sec": 123283.2059187871, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 281 |
+
{"event": "train", "epoch": 0, "step": 5600, "loss": 2.5299007296562195, "perplexity": 12.552260007651245, "learning_rate": 0.0002971400432612281, "grad_norm": 0.4818657636642456, "tokens_per_sec": 134732.3159763511, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 282 |
+
{"event": "train", "epoch": 0, "step": 5620, "loss": 2.6286816596984863, "perplexity": 13.855491603290341, "learning_rate": 0.00029711964426705094, "grad_norm": 0.4457741677761078, "tokens_per_sec": 134934.70036787266, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 283 |
+
{"event": "train", "epoch": 0, "step": 5640, "loss": 2.4363393783569336, "perplexity": 11.431119058066256, "learning_rate": 0.0002970991734872125, "grad_norm": 0.5037310719490051, "tokens_per_sec": 134880.40703102842, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 284 |
+
{"event": "train", "epoch": 0, "step": 5660, "loss": 2.6908987164497375, "perplexity": 14.744921471030048, "learning_rate": 0.0002970786309317013, "grad_norm": 0.4707108438014984, "tokens_per_sec": 133632.9533282579, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 285 |
+
{"event": "train", "epoch": 0, "step": 5680, "loss": 2.607206702232361, "perplexity": 13.561117657462585, "learning_rate": 0.0002970580166105406, "grad_norm": 0.4742436707019806, "tokens_per_sec": 134554.90366680568, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 286 |
+
{"event": "train", "epoch": 0, "step": 5700, "loss": 2.437864899635315, "perplexity": 11.448570781528296, "learning_rate": 0.00029703733053378914, "grad_norm": 0.49854958057403564, "tokens_per_sec": 134966.63465832872, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 287 |
+
{"event": "train", "epoch": 0, "step": 5720, "loss": 2.6121312975883484, "perplexity": 13.628065384433905, "learning_rate": 0.00029701657271154056, "grad_norm": 0.4902406632900238, "tokens_per_sec": 133768.74053786267, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 288 |
+
{"event": "train", "epoch": 0, "step": 5740, "loss": 2.7865962386131287, "perplexity": 16.225697300956273, "learning_rate": 0.0002969957431539232, "grad_norm": 0.4907446503639221, "tokens_per_sec": 129867.300264669, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 289 |
+
{"event": "train", "epoch": 0, "step": 5760, "loss": 2.298746556043625, "perplexity": 9.961688208151767, "learning_rate": 0.0002969748418711009, "grad_norm": 0.5116310119628906, "tokens_per_sec": 134091.62425570484, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 290 |
+
{"event": "train", "epoch": 0, "step": 5780, "loss": 2.58929306268692, "perplexity": 13.320351620335641, "learning_rate": 0.00029695386887327235, "grad_norm": 0.5090847015380859, "tokens_per_sec": 135236.22582575018, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 291 |
+
{"event": "train", "epoch": 0, "step": 5800, "loss": 2.1041520535945892, "perplexity": 8.20014677577352, "learning_rate": 0.00029693282417067075, "grad_norm": 0.484286904335022, "tokens_per_sec": 134271.48904541857, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 292 |
+
{"event": "train", "epoch": 0, "step": 5820, "loss": 2.87812203168869, "perplexity": 17.78084993281404, "learning_rate": 0.0002969117077735648, "grad_norm": 0.7979270219802856, "tokens_per_sec": 134780.6742340812, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 293 |
+
{"event": "train", "epoch": 0, "step": 5840, "loss": 3.0130395889282227, "perplexity": 20.349159095130698, "learning_rate": 0.00029689051969225837, "grad_norm": 0.5532850623130798, "tokens_per_sec": 130018.78160575932, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 294 |
+
{"event": "train", "epoch": 0, "step": 5860, "loss": 2.4208860397338867, "perplexity": 11.25582800875379, "learning_rate": 0.0002968692599370896, "grad_norm": 0.47452738881111145, "tokens_per_sec": 134011.7373022327, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 295 |
+
{"event": "train", "epoch": 0, "step": 5880, "loss": 2.7266148924827576, "perplexity": 15.281071299167456, "learning_rate": 0.00029684792851843216, "grad_norm": 0.4834570288658142, "tokens_per_sec": 134824.57023065732, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 296 |
+
{"event": "train", "epoch": 0, "step": 5900, "loss": 2.3996610045433044, "perplexity": 11.019440207238894, "learning_rate": 0.0002968265254466944, "grad_norm": 0.45717793703079224, "tokens_per_sec": 132778.81538831626, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 297 |
+
{"event": "train", "epoch": 0, "step": 5920, "loss": 2.4858857095241547, "perplexity": 12.011754470059085, "learning_rate": 0.0002968050507323198, "grad_norm": 0.4466097056865692, "tokens_per_sec": 133798.4319374149, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 298 |
+
{"event": "train", "epoch": 0, "step": 5940, "loss": 2.868861138820648, "perplexity": 17.616943517797782, "learning_rate": 0.00029678350438578667, "grad_norm": 0.5672624707221985, "tokens_per_sec": 134230.836480125, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 299 |
+
{"event": "train", "epoch": 0, "step": 5960, "loss": 2.351817309856415, "perplexity": 10.504642579332588, "learning_rate": 0.0002967618864176087, "grad_norm": 0.4251001179218292, "tokens_per_sec": 133932.46753935222, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 300 |
+
{"event": "train", "epoch": 0, "step": 5980, "loss": 2.325473129749298, "perplexity": 10.231519777186849, "learning_rate": 0.0002967401968383338, "grad_norm": 0.49882251024246216, "tokens_per_sec": 134407.00035792592, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 301 |
+
{"event": "train", "epoch": 0, "step": 6000, "loss": 2.6340309977531433, "perplexity": 13.9298079060872, "learning_rate": 0.00029671843565854533, "grad_norm": 0.4530165493488312, "tokens_per_sec": 134720.16580587171, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 302 |
+
{"event": "train", "epoch": 0, "step": 6020, "loss": 2.101533740758896, "perplexity": 8.178704310007626, "learning_rate": 0.00029669660288886144, "grad_norm": 0.4476483166217804, "tokens_per_sec": 134708.54595405722, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 303 |
+
{"event": "train", "epoch": 0, "step": 6040, "loss": 2.513592779636383, "perplexity": 12.349218474274679, "learning_rate": 0.0002966746985399354, "grad_norm": 0.47127240896224976, "tokens_per_sec": 134494.3345121901, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 304 |
+
{"event": "train", "epoch": 0, "step": 6060, "loss": 2.922949969768524, "perplexity": 18.59606445273539, "learning_rate": 0.00029665272262245493, "grad_norm": 0.5516300201416016, "tokens_per_sec": 133795.43615463693, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 305 |
+
{"event": "train", "epoch": 0, "step": 6080, "loss": 2.607621669769287, "perplexity": 13.566746248815445, "learning_rate": 0.0002966306751471434, "grad_norm": 0.47290879487991333, "tokens_per_sec": 133825.4000206426, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 306 |
+
{"event": "train", "epoch": 0, "step": 6100, "loss": 2.3813690543174744, "perplexity": 10.819705493336805, "learning_rate": 0.0002966085561247582, "grad_norm": 0.4657699167728424, "tokens_per_sec": 134183.13391078863, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 307 |
+
{"event": "train", "epoch": 0, "step": 6120, "loss": 3.3925670385360718, "perplexity": 29.742203741415686, "learning_rate": 0.00029658636556609204, "grad_norm": 0.720697820186615, "tokens_per_sec": 134947.02176803153, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 308 |
+
{"event": "train", "epoch": 0, "step": 6140, "loss": 2.6970620155334473, "perplexity": 14.836079460435734, "learning_rate": 0.00029656410348197297, "grad_norm": 0.598620593547821, "tokens_per_sec": 121573.27620713966, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 309 |
+
{"event": "train", "epoch": 0, "step": 6160, "loss": 2.4979013800621033, "perplexity": 12.156954344331773, "learning_rate": 0.0002965417698832635, "grad_norm": 0.5381540060043335, "tokens_per_sec": 134315.97278869015, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 310 |
+
{"event": "train", "epoch": 0, "step": 6180, "loss": 2.8182130455970764, "perplexity": 16.746897974986556, "learning_rate": 0.000296519364780861, "grad_norm": 0.472140371799469, "tokens_per_sec": 134408.0519016185, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 311 |
+
{"event": "train", "epoch": 0, "step": 6200, "loss": 3.1519338488578796, "perplexity": 23.381236666695425, "learning_rate": 0.0002964968881856979, "grad_norm": 0.5627263784408569, "tokens_per_sec": 133590.8685047215, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 312 |
+
{"event": "train", "epoch": 0, "step": 6220, "loss": 2.489430695772171, "perplexity": 12.054411539159315, "learning_rate": 0.0002964743401087415, "grad_norm": 0.49713075160980225, "tokens_per_sec": 134085.47572413928, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 313 |
+
{"event": "train", "epoch": 0, "step": 6240, "loss": 2.702902376651764, "perplexity": 14.922981043361448, "learning_rate": 0.0002964517205609938, "grad_norm": 0.4908139705657959, "tokens_per_sec": 133390.67973278958, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 314 |
+
{"event": "train", "epoch": 0, "step": 6260, "loss": 3.266832947731018, "perplexity": 26.228141771369092, "learning_rate": 0.0002964290295534919, "grad_norm": 1.3085947036743164, "tokens_per_sec": 134752.39497341504, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 315 |
+
{"event": "train", "epoch": 0, "step": 6280, "loss": 2.3373910784721375, "perplexity": 10.354188030268906, "learning_rate": 0.00029640626709730747, "grad_norm": 0.6734078526496887, "tokens_per_sec": 134466.96461996943, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 316 |
+
{"event": "train", "epoch": 0, "step": 6300, "loss": 3.0702144503593445, "perplexity": 21.54652283915226, "learning_rate": 0.00029638343320354747, "grad_norm": 0.5207536220550537, "tokens_per_sec": 135212.41070492205, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 317 |
+
{"event": "train", "epoch": 0, "step": 6320, "loss": 2.700285851955414, "perplexity": 14.883985733261763, "learning_rate": 0.0002963605278833533, "grad_norm": 0.4782971143722534, "tokens_per_sec": 134390.0464873417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 318 |
+
{"event": "train", "epoch": 0, "step": 6340, "loss": 2.410338968038559, "perplexity": 11.137735841750663, "learning_rate": 0.0002963375511479015, "grad_norm": 0.48620015382766724, "tokens_per_sec": 134821.26381379142, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 319 |
+
{"event": "train", "epoch": 0, "step": 6360, "loss": 2.645766496658325, "perplexity": 14.094244134908168, "learning_rate": 0.00029631450300840337, "grad_norm": 0.6486227512359619, "tokens_per_sec": 134839.25268032338, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 320 |
+
{"event": "train", "epoch": 0, "step": 6380, "loss": 2.94451767206192, "perplexity": 19.00149522384509, "learning_rate": 0.000296291383476105, "grad_norm": 0.5201399326324463, "tokens_per_sec": 125626.1965812669, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 321 |
+
{"event": "train", "epoch": 0, "step": 6400, "loss": 2.8127677142620087, "perplexity": 16.655953403062952, "learning_rate": 0.0002962681925622873, "grad_norm": 0.9973184466362, "tokens_per_sec": 135339.69939468897, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 322 |
+
{"event": "train", "epoch": 0, "step": 6420, "loss": 2.5570201873779297, "perplexity": 12.897328377221047, "learning_rate": 0.0002962449302782661, "grad_norm": 0.6173967123031616, "tokens_per_sec": 134275.9492123538, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 323 |
+
{"event": "train", "epoch": 0, "step": 6440, "loss": 2.830825686454773, "perplexity": 16.959458239956643, "learning_rate": 0.0002962215966353922, "grad_norm": 0.5163636207580566, "tokens_per_sec": 133888.88956517927, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 324 |
+
{"event": "train", "epoch": 0, "step": 6460, "loss": 2.381280481815338, "perplexity": 10.818747207388402, "learning_rate": 0.00029619819164505064, "grad_norm": 0.5077759027481079, "tokens_per_sec": 132094.667417618, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 325 |
+
{"event": "train", "epoch": 0, "step": 6480, "loss": 2.5304166674613953, "perplexity": 12.558737864071404, "learning_rate": 0.0002961747153186622, "grad_norm": 0.44933298230171204, "tokens_per_sec": 135250.99807612257, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 326 |
+
{"event": "train", "epoch": 0, "step": 6500, "loss": 2.2993077635765076, "perplexity": 9.967280351644165, "learning_rate": 0.0002961511676676815, "grad_norm": 0.49738889932632446, "tokens_per_sec": 134911.65346759162, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 327 |
+
{"event": "train", "epoch": 0, "step": 6520, "loss": 2.254355937242508, "perplexity": 9.52915396011951, "learning_rate": 0.0002961275487035987, "grad_norm": 0.5270805954933167, "tokens_per_sec": 135025.9055145495, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 328 |
+
{"event": "train", "epoch": 0, "step": 6540, "loss": 2.611161768436432, "perplexity": 13.61485898079062, "learning_rate": 0.00029610385843793817, "grad_norm": 0.5748497247695923, "tokens_per_sec": 134778.69165609207, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 329 |
+
{"event": "train", "epoch": 0, "step": 6560, "loss": 2.4672847986221313, "perplexity": 11.790390067671401, "learning_rate": 0.00029608009688225943, "grad_norm": 0.475819855928421, "tokens_per_sec": 134643.089714087, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 330 |
+
{"event": "train", "epoch": 0, "step": 6580, "loss": 3.038722038269043, "perplexity": 20.878544198069452, "learning_rate": 0.00029605626404815666, "grad_norm": 0.5069053173065186, "tokens_per_sec": 133884.846341031, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 331 |
+
{"event": "train", "epoch": 0, "step": 6600, "loss": 2.547275424003601, "perplexity": 12.772257347550067, "learning_rate": 0.0002960323599472589, "grad_norm": 0.4761556386947632, "tokens_per_sec": 134490.12303434886, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 332 |
+
{"event": "train", "epoch": 0, "step": 6620, "loss": 2.5598740577697754, "perplexity": 12.934188252446651, "learning_rate": 0.0002960083845912302, "grad_norm": 0.45484763383865356, "tokens_per_sec": 134299.17310400156, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 333 |
+
{"event": "train", "epoch": 0, "step": 6640, "loss": 2.039873868227005, "perplexity": 7.689639229878573, "learning_rate": 0.00029598433799176875, "grad_norm": 0.46897658705711365, "tokens_per_sec": 135299.1966765602, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 334 |
+
{"event": "train", "epoch": 0, "step": 6660, "loss": 2.2812221944332123, "perplexity": 9.788636718879772, "learning_rate": 0.0002959602201606081, "grad_norm": 0.4648001790046692, "tokens_per_sec": 135004.81662819366, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 335 |
+
{"event": "train", "epoch": 0, "step": 6680, "loss": 2.6110113859176636, "perplexity": 13.612811697946048, "learning_rate": 0.0002959360311095163, "grad_norm": 0.511810839176178, "tokens_per_sec": 134006.51070675714, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 336 |
+
{"event": "train", "epoch": 0, "step": 6700, "loss": 2.496724486351013, "perplexity": 12.14265531708598, "learning_rate": 0.000295911770850296, "grad_norm": 0.5543397068977356, "tokens_per_sec": 134652.85132522968, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 337 |
+
{"event": "train", "epoch": 0, "step": 6720, "loss": 2.188087224960327, "perplexity": 8.91813839733134, "learning_rate": 0.000295887439394785, "grad_norm": 0.4284917414188385, "tokens_per_sec": 134954.44219125892, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 338 |
+
{"event": "train", "epoch": 0, "step": 6740, "loss": 3.0125972628593445, "perplexity": 20.340160121970108, "learning_rate": 0.0002958630367548554, "grad_norm": 0.525662899017334, "tokens_per_sec": 134300.74789591614, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 339 |
+
{"event": "train", "epoch": 0, "step": 6760, "loss": 2.907151758670807, "perplexity": 18.304588363593446, "learning_rate": 0.00029583856294241445, "grad_norm": 0.6326534152030945, "tokens_per_sec": 134337.37222213534, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 340 |
+
{"event": "train", "epoch": 0, "step": 6780, "loss": 2.4545068740844727, "perplexity": 11.640691806490388, "learning_rate": 0.00029581401796940367, "grad_norm": 0.4885016083717346, "tokens_per_sec": 135280.41910476438, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 341 |
+
{"event": "train", "epoch": 0, "step": 6800, "loss": 2.688436985015869, "perplexity": 14.708668075724626, "learning_rate": 0.00029578940184779974, "grad_norm": 0.4786076247692108, "tokens_per_sec": 134680.0329960754, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 342 |
+
{"event": "train", "epoch": 0, "step": 6820, "loss": 2.7291504740715027, "perplexity": 15.319866866087244, "learning_rate": 0.0002957647145896136, "grad_norm": 0.48833197355270386, "tokens_per_sec": 130485.63495644116, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 343 |
+
{"event": "train", "epoch": 0, "step": 6840, "loss": 2.6794556081295013, "perplexity": 14.577155450505334, "learning_rate": 0.00029573995620689155, "grad_norm": 0.5164268612861633, "tokens_per_sec": 134491.04427258912, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 344 |
+
{"event": "train", "epoch": 0, "step": 6860, "loss": 2.4876415133476257, "perplexity": 12.032863280526337, "learning_rate": 0.0002957151267117141, "grad_norm": 0.4964078962802887, "tokens_per_sec": 134616.58220939778, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 345 |
+
{"event": "train", "epoch": 0, "step": 6880, "loss": 3.0409411787986755, "perplexity": 20.924928068827768, "learning_rate": 0.00029569022611619644, "grad_norm": 0.5106762051582336, "tokens_per_sec": 133938.47151885473, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 346 |
+
{"event": "train", "epoch": 0, "step": 6900, "loss": 3.477058529853821, "perplexity": 32.364383059600804, "learning_rate": 0.0002956652544324886, "grad_norm": 0.6291273832321167, "tokens_per_sec": 134907.01854789315, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 347 |
+
{"event": "train", "epoch": 0, "step": 6920, "loss": 2.317229062318802, "perplexity": 10.14751717555011, "learning_rate": 0.0002956402116727754, "grad_norm": 0.4948063790798187, "tokens_per_sec": 134350.5040841144, "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}
|