Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +91 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -4407,3 +4407,94 @@
|
|
| 4407 |
{"event": "train", "epoch": 2, "step": 88080, "loss": 2.0317426919937134, "perplexity": 7.62736693473249, "learning_rate": 3.194719343562858e-07, "grad_norm": 1.0545654296875, "tokens_per_sec": 134968.22514254064, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4408 |
{"event": "train", "epoch": 2, "step": 88100, "loss": 2.1066610515117645, "perplexity": 8.2207467588052, "learning_rate": 3.12673600014296e-07, "grad_norm": 0.9421895742416382, "tokens_per_sec": 135020.06893699375, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4409 |
{"event": "train", "epoch": 2, "step": 88120, "loss": 2.8083614706993103, "perplexity": 16.582724665903807, "learning_rate": 3.059483042103238e-07, "grad_norm": 1.1604013442993164, "tokens_per_sec": 134987.84419706961, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4407 |
{"event": "train", "epoch": 2, "step": 88080, "loss": 2.0317426919937134, "perplexity": 7.62736693473249, "learning_rate": 3.194719343562858e-07, "grad_norm": 1.0545654296875, "tokens_per_sec": 134968.22514254064, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4408 |
{"event": "train", "epoch": 2, "step": 88100, "loss": 2.1066610515117645, "perplexity": 8.2207467588052, "learning_rate": 3.12673600014296e-07, "grad_norm": 0.9421895742416382, "tokens_per_sec": 135020.06893699375, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4409 |
{"event": "train", "epoch": 2, "step": 88120, "loss": 2.8083614706993103, "perplexity": 16.582724665903807, "learning_rate": 3.059483042103238e-07, "grad_norm": 1.1604013442993164, "tokens_per_sec": 134987.84419706961, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4410 |
+
{"event": "train", "epoch": 2, "step": 88140, "loss": 2.0194091498851776, "perplexity": 7.5338722290793445, "learning_rate": 2.9929605022592767e-07, "grad_norm": 0.8540683388710022, "tokens_per_sec": 134926.88473205853, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4411 |
+
{"event": "train", "epoch": 2, "step": 88160, "loss": 1.8109159171581268, "perplexity": 6.1160466597004435, "learning_rate": 2.9271684130697783e-07, "grad_norm": 0.7459234595298767, "tokens_per_sec": 134973.39447512294, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4412 |
+
{"event": "train", "epoch": 2, "step": 88180, "loss": 1.9245347082614899, "perplexity": 6.851959763801109, "learning_rate": 2.8621068066378965e-07, "grad_norm": 0.8038586378097534, "tokens_per_sec": 134982.67375763235, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4413 |
+
{"event": "train", "epoch": 2, "step": 88200, "loss": 1.7806071043014526, "perplexity": 5.933457552951341, "learning_rate": 2.7977757147094027e-07, "grad_norm": 0.728468656539917, "tokens_per_sec": 134954.83973696176, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4414 |
+
{"event": "train", "epoch": 2, "step": 88220, "loss": 1.586144894361496, "perplexity": 4.884880850767171, "learning_rate": 2.734175168674188e-07, "grad_norm": 0.7508091330528259, "tokens_per_sec": 134919.0700015805, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4415 |
+
{"event": "train", "epoch": 2, "step": 88240, "loss": 2.8639771342277527, "perplexity": 17.531112055985933, "learning_rate": 2.671305199565595e-07, "grad_norm": 1.2302143573760986, "tokens_per_sec": 134927.41457706256, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4416 |
+
{"event": "train", "epoch": 2, "step": 88260, "loss": 2.3574650585651398, "perplexity": 10.564138010373494, "learning_rate": 2.60916583806042e-07, "grad_norm": 0.9367653727531433, "tokens_per_sec": 135016.22237765536, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4417 |
+
{"event": "train", "epoch": 2, "step": 88280, "loss": 2.011819839477539, "perplexity": 7.476911752927401, "learning_rate": 2.547757114478908e-07, "grad_norm": 0.7996217012405396, "tokens_per_sec": 135016.22237765536, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4418 |
+
{"event": "train", "epoch": 2, "step": 88300, "loss": 1.8998144567012787, "perplexity": 6.684654034447785, "learning_rate": 2.4870790587849237e-07, "grad_norm": 0.7698596715927124, "tokens_per_sec": 134703.39688799853, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4419 |
+
{"event": "train", "epoch": 2, "step": 88320, "loss": 2.1738370656967163, "perplexity": 8.791954708967378, "learning_rate": 2.4271317005857814e-07, "grad_norm": 0.8689727187156677, "tokens_per_sec": 134914.699445377, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4420 |
+
{"event": "train", "epoch": 2, "step": 88340, "loss": 1.8529677093029022, "perplexity": 6.378721652303861, "learning_rate": 2.3679150691324191e-07, "grad_norm": 0.7414506077766418, "tokens_per_sec": 134990.89361626896, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4421 |
+
{"event": "train", "epoch": 2, "step": 88360, "loss": 2.054734915494919, "perplexity": 7.804768675150395, "learning_rate": 2.3094291933185563e-07, "grad_norm": 0.9870826601982117, "tokens_per_sec": 134971.40622342337, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4422 |
+
{"event": "train", "epoch": 2, "step": 88380, "loss": 1.9244248270988464, "perplexity": 6.851206903859197, "learning_rate": 2.2516741016823647e-07, "grad_norm": 0.8545060157775879, "tokens_per_sec": 135006.27539564585, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4423 |
+
{"event": "train", "epoch": 2, "step": 88400, "loss": 1.8843434154987335, "perplexity": 6.5820313631571885, "learning_rate": 2.1946498224044695e-07, "grad_norm": 0.8015924096107483, "tokens_per_sec": 134924.36802511592, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4424 |
+
{"event": "train", "epoch": 2, "step": 88420, "loss": 1.7996081709861755, "perplexity": 6.047277501353445, "learning_rate": 2.1383563833097793e-07, "grad_norm": 0.7876927256584167, "tokens_per_sec": 135042.62214208863, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4425 |
+
{"event": "train", "epoch": 2, "step": 88440, "loss": 2.05060076713562, "perplexity": 7.772569208084854, "learning_rate": 2.082793811865656e-07, "grad_norm": 0.9270812273025513, "tokens_per_sec": 134930.19633160645, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4426 |
+
{"event": "train", "epoch": 2, "step": 88460, "loss": 1.966462880373001, "perplexity": 7.145357751897976, "learning_rate": 2.0279621351837457e-07, "grad_norm": 0.880328893661499, "tokens_per_sec": 134889.67353192024, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4427 |
+
{"event": "train", "epoch": 2, "step": 88480, "loss": 2.062799423933029, "perplexity": 7.867964778238949, "learning_rate": 1.9738613800184808e-07, "grad_norm": 1.014369010925293, "tokens_per_sec": 134846.66125601437, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4428 |
+
{"event": "train", "epoch": 2, "step": 88500, "loss": 1.8863459527492523, "perplexity": 6.595225332444329, "learning_rate": 1.9204915727677446e-07, "grad_norm": 0.843732476234436, "tokens_per_sec": 134825.23153349172, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4429 |
+
{"event": "train", "epoch": 2, "step": 88520, "loss": 2.206128180027008, "perplexity": 9.080490218950068, "learning_rate": 1.8678527394730397e-07, "grad_norm": 1.068782925605774, "tokens_per_sec": 135020.9974496713, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4430 |
+
{"event": "train", "epoch": 2, "step": 88540, "loss": 2.0959249138832092, "perplexity": 8.13295977996293, "learning_rate": 1.8159449058188213e-07, "grad_norm": 0.9412415027618408, "tokens_per_sec": 135079.78502651196, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4431 |
+
{"event": "train", "epoch": 2, "step": 88560, "loss": 2.04272198677063, "perplexity": 7.71157145192135, "learning_rate": 1.764768097133161e-07, "grad_norm": 0.8255587816238403, "tokens_per_sec": 135022.58921540118, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4432 |
+
{"event": "train", "epoch": 2, "step": 88580, "loss": 2.060060352087021, "perplexity": 7.846443345260472, "learning_rate": 1.714322338387085e-07, "grad_norm": 0.8901656866073608, "tokens_per_sec": 135004.94924211391, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4433 |
+
{"event": "train", "epoch": 2, "step": 88600, "loss": 2.7591058015823364, "perplexity": 15.785721068518967, "learning_rate": 1.66460765419557e-07, "grad_norm": 1.289744257926941, "tokens_per_sec": 135031.60993016502, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4434 |
+
{"event": "train", "epoch": 2, "step": 88620, "loss": 2.4684447646141052, "perplexity": 11.80407445435921, "learning_rate": 1.6156240688158783e-07, "grad_norm": 0.9098497033119202, "tokens_per_sec": 134965.83943027956, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4435 |
+
{"event": "train", "epoch": 2, "step": 88640, "loss": 2.58322811126709, "perplexity": 13.239808825819143, "learning_rate": 1.5673716061495575e-07, "grad_norm": 0.9482001662254333, "tokens_per_sec": 134998.05366370652, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4436 |
+
{"event": "train", "epoch": 2, "step": 88660, "loss": 1.637957125902176, "perplexity": 5.144648899770429, "learning_rate": 1.5198502897406085e-07, "grad_norm": 0.7522266507148743, "tokens_per_sec": 135001.63397226474, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4437 |
+
{"event": "train", "epoch": 2, "step": 88680, "loss": 2.4773033559322357, "perplexity": 11.909106456770019, "learning_rate": 1.4730601427769814e-07, "grad_norm": 0.9010107517242432, "tokens_per_sec": 135096.7805869638, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4438 |
+
{"event": "train", "epoch": 2, "step": 88700, "loss": 1.7389581799507141, "perplexity": 5.691410906910106, "learning_rate": 1.427001188089248e-07, "grad_norm": 0.7414847612380981, "tokens_per_sec": 134991.55655171737, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4439 |
+
{"event": "train", "epoch": 2, "step": 88720, "loss": 2.0333397686481476, "perplexity": 7.6395581569656645, "learning_rate": 1.381673448151265e-07, "grad_norm": 0.8721235394477844, "tokens_per_sec": 134903.84050129956, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4440 |
+
{"event": "train", "epoch": 2, "step": 88740, "loss": 1.9010854959487915, "perplexity": 6.693155894035629, "learning_rate": 1.3370769450806717e-07, "grad_norm": 0.8035126328468323, "tokens_per_sec": 134971.2736753928, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4441 |
+
{"event": "train", "epoch": 2, "step": 88760, "loss": 1.6227423250675201, "perplexity": 5.066966550683585, "learning_rate": 1.2932117006377293e-07, "grad_norm": 0.7023391723632812, "tokens_per_sec": 134986.91614031125, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4442 |
+
{"event": "train", "epoch": 2, "step": 88780, "loss": 2.732225179672241, "perplexity": 15.367043436408927, "learning_rate": 1.2500777362259833e-07, "grad_norm": 1.216907024383545, "tokens_per_sec": 134857.64332118913, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4443 |
+
{"event": "train", "epoch": 2, "step": 88800, "loss": 1.6475732326507568, "perplexity": 5.194359018624015, "learning_rate": 1.2076750728924314e-07, "grad_norm": 0.7284180521965027, "tokens_per_sec": 135051.38012280886, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4444 |
+
{"event": "train", "epoch": 2, "step": 88820, "loss": 1.9510034322738647, "perplexity": 7.03574393224399, "learning_rate": 1.1660037313270238e-07, "grad_norm": 0.9050215482711792, "tokens_per_sec": 134974.85256359895, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4445 |
+
{"event": "train", "epoch": 2, "step": 88840, "loss": 1.6398335099220276, "perplexity": 5.154311299104596, "learning_rate": 1.1250637318628284e-07, "grad_norm": 0.7921112179756165, "tokens_per_sec": 135089.21172328517, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4446 |
+
{"event": "train", "epoch": 2, "step": 88860, "loss": 1.7508487701416016, "perplexity": 5.759489084359308, "learning_rate": 1.084855094476199e-07, "grad_norm": 0.7502493858337402, "tokens_per_sec": 135019.0077952851, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4447 |
+
{"event": "train", "epoch": 2, "step": 88880, "loss": 1.8948379456996918, "perplexity": 6.651470417904444, "learning_rate": 1.0453778387866087e-07, "grad_norm": 0.8147830367088318, "tokens_per_sec": 134991.68913958842, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4448 |
+
{"event": "train", "epoch": 2, "step": 88900, "loss": 1.8923231959342957, "perplexity": 6.634764648450281, "learning_rate": 1.0066319840563157e-07, "grad_norm": 0.7627758979797363, "tokens_per_sec": 134981.0829328753, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4449 |
+
{"event": "train", "epoch": 2, "step": 88920, "loss": 2.2237913608551025, "perplexity": 9.24230543842485, "learning_rate": 9.68617549191363e-08, "grad_norm": 0.9955180287361145, "tokens_per_sec": 135017.68139160436, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4450 |
+
{"event": "train", "epoch": 2, "step": 88940, "loss": 2.1959145069122314, "perplexity": 8.988217086092854, "learning_rate": 9.313345527404136e-08, "grad_norm": 0.8549285531044006, "tokens_per_sec": 134822.05733904906, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4451 |
+
{"event": "train", "epoch": 2, "step": 88960, "loss": 2.4166274070739746, "perplexity": 11.20799549482503, "learning_rate": 8.947830128954159e-08, "grad_norm": 0.9891444444656372, "tokens_per_sec": 134975.38278540087, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4452 |
+
{"event": "train", "epoch": 2, "step": 88980, "loss": 2.868027687072754, "perplexity": 17.602266762463604, "learning_rate": 8.589629474911033e-08, "grad_norm": 1.4488589763641357, "tokens_per_sec": 134899.47093171056, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4453 |
+
{"event": "train", "epoch": 2, "step": 89000, "loss": 2.5233217775821686, "perplexity": 12.469950343189957, "learning_rate": 8.238743740059941e-08, "grad_norm": 1.0177851915359497, "tokens_per_sec": 134949.4068150755, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4454 |
+
{"event": "train", "epoch": 2, "step": 89020, "loss": 1.8689685463905334, "perplexity": 6.48160747279809, "learning_rate": 7.895173095607271e-08, "grad_norm": 0.7893440127372742, "tokens_per_sec": 135024.18101866223, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4455 |
+
{"event": "train", "epoch": 2, "step": 89040, "loss": 1.8252780437469482, "perplexity": 6.204519907259599, "learning_rate": 7.558917709198908e-08, "grad_norm": 0.8275688886642456, "tokens_per_sec": 135013.03918394292, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4456 |
+
{"event": "train", "epoch": 2, "step": 89060, "loss": 2.2250813245773315, "perplexity": 9.254235370085414, "learning_rate": 7.229977744905289e-08, "grad_norm": 0.9887278079986572, "tokens_per_sec": 134985.59036707834, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4457 |
+
{"event": "train", "epoch": 2, "step": 89080, "loss": 1.7493182718753815, "perplexity": 5.7506809384467745, "learning_rate": 6.908353363231356e-08, "grad_norm": 0.7360363006591797, "tokens_per_sec": 134937.34994143574, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4458 |
+
{"event": "train", "epoch": 2, "step": 89100, "loss": 1.7341846525669098, "perplexity": 5.66430754188708, "learning_rate": 6.594044721109903e-08, "grad_norm": 0.7796377539634705, "tokens_per_sec": 134933.77304170767, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4459 |
+
{"event": "train", "epoch": 2, "step": 89120, "loss": 1.9771372079849243, "perplexity": 7.2220381693072575, "learning_rate": 6.287051971903257e-08, "grad_norm": 0.7679601907730103, "tokens_per_sec": 134970.21330052018, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4460 |
+
{"event": "train", "epoch": 2, "step": 89140, "loss": 2.562023788690567, "perplexity": 12.96202318495626, "learning_rate": 5.987375265408257e-08, "grad_norm": 1.0373823642730713, "tokens_per_sec": 134969.4180302996, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4461 |
+
{"event": "train", "epoch": 2, "step": 89160, "loss": 2.243604987859726, "perplexity": 9.427255243682136, "learning_rate": 5.6950147478496046e-08, "grad_norm": 0.8920223116874695, "tokens_per_sec": 135014.63076203686, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4462 |
+
{"event": "train", "epoch": 2, "step": 89180, "loss": 1.9221079647541046, "perplexity": 6.835351974504252, "learning_rate": 5.4099705618798546e-08, "grad_norm": 0.7622512578964233, "tokens_per_sec": 134914.30213612667, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4463 |
+
{"event": "train", "epoch": 2, "step": 89200, "loss": 2.3816067278385162, "perplexity": 10.822277356457818, "learning_rate": 5.132242846584415e-08, "grad_norm": 1.0184663534164429, "tokens_per_sec": 134878.8186158525, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4464 |
+
{"event": "train", "epoch": 2, "step": 89220, "loss": 2.1102153062820435, "perplexity": 8.250017373985788, "learning_rate": 4.861831737476558e-08, "grad_norm": 0.9283671975135803, "tokens_per_sec": 134870.3477000525, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4465 |
+
{"event": "train", "epoch": 2, "step": 89240, "loss": 1.8972191214561462, "perplexity": 6.667327609896381, "learning_rate": 4.598737366504064e-08, "grad_norm": 0.7736440300941467, "tokens_per_sec": 135007.86681427533, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4466 |
+
{"event": "train", "epoch": 2, "step": 89260, "loss": 1.7163139283657074, "perplexity": 5.56398138458144, "learning_rate": 4.34295986203925e-08, "grad_norm": 0.7491029500961304, "tokens_per_sec": 135021.79332784493, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4467 |
+
{"event": "train", "epoch": 2, "step": 89280, "loss": 1.824800193309784, "perplexity": 6.201555782969967, "learning_rate": 4.094499348885613e-08, "grad_norm": 0.8498821258544922, "tokens_per_sec": 135012.24340896722, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4468 |
+
{"event": "train", "epoch": 2, "step": 89300, "loss": 2.2679757475852966, "perplexity": 9.659827081373686, "learning_rate": 3.853355948279508e-08, "grad_norm": 0.943860650062561, "tokens_per_sec": 135035.19201887207, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4469 |
+
{"event": "train", "epoch": 2, "step": 89320, "loss": 2.6055575013160706, "perplexity": 13.538771081857428, "learning_rate": 3.619529777883485e-08, "grad_norm": 1.0121357440948486, "tokens_per_sec": 134960.1405701481, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4470 |
+
{"event": "train", "epoch": 2, "step": 89340, "loss": 2.167461097240448, "perplexity": 8.736075813415246, "learning_rate": 3.393020951791277e-08, "grad_norm": 1.1991899013519287, "tokens_per_sec": 134950.33435613845, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4471 |
+
{"event": "train", "epoch": 2, "step": 89360, "loss": 2.078522562980652, "perplexity": 7.992651547459366, "learning_rate": 3.173829580524473e-08, "grad_norm": 0.9740650057792664, "tokens_per_sec": 135005.4797004003, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4472 |
+
{"event": "train", "epoch": 2, "step": 89380, "loss": 2.5386828184127808, "perplexity": 12.662980536278749, "learning_rate": 2.961955771035855e-08, "grad_norm": 1.2886024713516235, "tokens_per_sec": 135091.4690149276, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4473 |
+
{"event": "train", "epoch": 2, "step": 89400, "loss": 2.3317618668079376, "perplexity": 10.296065858719931, "learning_rate": 2.7573996267077288e-08, "grad_norm": 0.9705244898796082, "tokens_per_sec": 134728.617866382, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4474 |
+
{"event": "train", "epoch": 2, "step": 89420, "loss": 2.034703403711319, "perplexity": 7.649982832446753, "learning_rate": 2.5601612473519153e-08, "grad_norm": 1.0336111783981323, "tokens_per_sec": 135077.793639187, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4475 |
+
{"event": "train", "epoch": 2, "step": 89440, "loss": 2.292974978685379, "perplexity": 9.904359152719937, "learning_rate": 2.3702407292080998e-08, "grad_norm": 1.1166934967041016, "tokens_per_sec": 134965.971967637, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4476 |
+
{"event": "train", "epoch": 2, "step": 89460, "loss": 1.727269560098648, "perplexity": 5.625273449192327, "learning_rate": 2.1876381649454878e-08, "grad_norm": 0.7740680575370789, "tokens_per_sec": 135071.42159429519, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4477 |
+
{"event": "train", "epoch": 2, "step": 89480, "loss": 2.0860145688056946, "perplexity": 8.05275741501972, "learning_rate": 2.012353643666137e-08, "grad_norm": 0.861927330493927, "tokens_per_sec": 134979.88983891465, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4478 |
+
{"event": "train", "epoch": 2, "step": 89500, "loss": 1.9138035774230957, "perplexity": 6.778823605761547, "learning_rate": 1.8443872508949688e-08, "grad_norm": 0.8385021090507507, "tokens_per_sec": 135070.0941606112, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4479 |
+
{"event": "train", "epoch": 2, "step": 89520, "loss": 1.9901905357837677, "perplexity": 7.316927766068229, "learning_rate": 1.6837390685930876e-08, "grad_norm": 0.8111202716827393, "tokens_per_sec": 135030.81393626073, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4480 |
+
{"event": "train", "epoch": 2, "step": 89540, "loss": 1.9936281740665436, "perplexity": 7.3421240000256525, "learning_rate": 1.5304091751444593e-08, "grad_norm": 0.8949370384216309, "tokens_per_sec": 135090.93788069562, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4481 |
+
{"event": "train", "epoch": 2, "step": 89560, "loss": 2.2212426364421844, "perplexity": 9.21877934242535, "learning_rate": 1.3843976453675693e-08, "grad_norm": 0.8154850602149963, "tokens_per_sec": 135043.81634529977, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4482 |
+
{"event": "train", "epoch": 2, "step": 89580, "loss": 1.7502898573875427, "perplexity": 5.7562709318702066, "learning_rate": 1.2457045505037634e-08, "grad_norm": 0.7991673946380615, "tokens_per_sec": 135050.4511922145, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4483 |
+
{"event": "train", "epoch": 2, "step": 89600, "loss": 2.331788122653961, "perplexity": 10.296336194188694, "learning_rate": 1.1143299582305712e-08, "grad_norm": 0.953007698059082, "tokens_per_sec": 135076.06781766558, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4484 |
+
{"event": "train", "epoch": 2, "step": 89620, "loss": 2.0170161426067352, "perplexity": 7.5158651721025995, "learning_rate": 9.902739326500505e-09, "grad_norm": 0.8688393235206604, "tokens_per_sec": 135000.83833172897, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4485 |
+
{"event": "train", "epoch": 2, "step": 89640, "loss": 1.8941733539104462, "perplexity": 6.647051373871136, "learning_rate": 8.735365342921154e-09, "grad_norm": 0.8296386003494263, "tokens_per_sec": 135129.8545179965, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4486 |
+
{"event": "train", "epoch": 2, "step": 89660, "loss": 1.9137374758720398, "perplexity": 6.778375529816292, "learning_rate": 7.641178201211993e-09, "grad_norm": 0.8136295080184937, "tokens_per_sec": 135049.78767818524, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4487 |
+
{"event": "train", "epoch": 2, "step": 89680, "loss": 1.9445903599262238, "perplexity": 6.990767569866956, "learning_rate": 6.620178435229306e-09, "grad_norm": 0.806309163570404, "tokens_per_sec": 135012.24340896722, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4488 |
+
{"event": "train", "epoch": 2, "step": 89700, "loss": 2.4811633825302124, "perplexity": 11.955164760401525, "learning_rate": 5.672366543207886e-09, "grad_norm": 1.0556854009628296, "tokens_per_sec": 134951.65943691815, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4489 |
+
{"event": "train", "epoch": 2, "step": 89720, "loss": 1.9908719062805176, "perplexity": 7.321915003660205, "learning_rate": 4.797742987577828e-09, "grad_norm": 0.8580937385559082, "tokens_per_sec": 135070.3596452607, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4490 |
+
{"event": "train", "epoch": 2, "step": 89740, "loss": 1.9088431298732758, "perplexity": 6.7452808690881225, "learning_rate": 3.996308195147719e-09, "grad_norm": 0.886562168598175, "tokens_per_sec": 134907.68065978054, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4491 |
+
{"event": "train", "epoch": 2, "step": 89760, "loss": 2.0854081511497498, "perplexity": 8.047875561115017, "learning_rate": 3.2680625569381097e-09, "grad_norm": 0.8783664703369141, "tokens_per_sec": 134871.67121050798, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4492 |
+
{"event": "train", "epoch": 2, "step": 89780, "loss": 1.8361603021621704, "perplexity": 6.272407813141088, "learning_rate": 2.6130064282980824e-09, "grad_norm": 0.8052548170089722, "tokens_per_sec": 135026.83410750356, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4493 |
+
{"event": "train", "epoch": 2, "step": 89800, "loss": 2.1799951791763306, "perplexity": 8.846263612341078, "learning_rate": 2.0311401288552954e-09, "grad_norm": 0.9683517217636108, "tokens_per_sec": 135077.1298564598, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4494 |
+
{"event": "train", "epoch": 2, "step": 89820, "loss": 2.202281206846237, "perplexity": 9.045624922537838, "learning_rate": 1.5224639425326317e-09, "grad_norm": 0.8907625079154968, "tokens_per_sec": 135117.23403949928, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4495 |
+
{"event": "train", "epoch": 2, "step": 89840, "loss": 1.9739172160625458, "perplexity": 7.198820664887916, "learning_rate": 1.0869781175148992e-09, "grad_norm": 0.9165233969688416, "tokens_per_sec": 135068.76675301805, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4496 |
+
{"event": "train", "epoch": 2, "step": 89860, "loss": 2.2911266088485718, "perplexity": 9.886069142567958, "learning_rate": 7.246828663320911e-10, "grad_norm": 0.9431777000427246, "tokens_per_sec": 135055.09597298093, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4497 |
+
{"event": "train", "epoch": 2, "step": 89880, "loss": 1.9226097762584686, "perplexity": 6.838782893526758, "learning_rate": 4.3557836572616197e-10, "grad_norm": 0.8639533519744873, "tokens_per_sec": 134978.16651820656, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4498 |
+
{"event": "train", "epoch": 2, "step": 89900, "loss": 2.201013594865799, "perplexity": 9.034165844382265, "learning_rate": 2.1966475676760052e-10, "grad_norm": 0.8885666131973267, "tokens_per_sec": 135037.3148277878, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4499 |
+
{"event": "train", "epoch": 2, "step": 89920, "loss": 2.180094212293625, "perplexity": 8.84713972878457, "learning_rate": 7.69421448387761e-11, "grad_norm": 0.9647793173789978, "tokens_per_sec": 134914.699445377, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 4500 |
+
{"event": "train", "epoch": 2, "step": 89940, "loss": 2.309117615222931, "perplexity": 10.065539056895059, "learning_rate": 7.410599534018987e-12, "grad_norm": 1.5659576654434204, "tokens_per_sec": 135034.13063944218, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|