Banaxi-Tech commited on
Commit
58aa638
·
verified ·
1 Parent(s): 0870c3e

Save 85% checkpoint at 25,501,368,320 tokens

Browse files
checkpoint_metadata.json CHANGED
@@ -50,10 +50,10 @@
50
  "model_type": "tiger_alpha",
51
  "output_attentions": false
52
  },
53
- "training_percent": 80,
54
- "step": 22889,
55
  "total_steps": 28611,
56
- "tokens_seen": 24000856064,
57
  "target_tokens": 30000000000,
58
  "trained_token_target": 30000807936,
59
  "tokens_per_step": 1048576,
@@ -69,5 +69,5 @@
69
  "dataset_id": "HuggingFaceFW/fineweb-edu",
70
  "dataset_config": "sample-100BT",
71
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
72
- "elapsed_seconds": 7330.718915462494
73
  }
 
50
  "model_type": "tiger_alpha",
51
  "output_attentions": false
52
  },
53
+ "training_percent": 85,
54
+ "step": 24320,
55
  "total_steps": 28611,
56
+ "tokens_seen": 25501368320,
57
  "target_tokens": 30000000000,
58
  "trained_token_target": 30000807936,
59
  "tokens_per_step": 1048576,
 
69
  "dataset_id": "HuggingFaceFW/fineweb-edu",
70
  "dataset_config": "sample-100BT",
71
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
72
+ "elapsed_seconds": 7783.25132393837
73
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:87b90f8f70b129276e575cc268041550cebb7bab1397137ad789db2b2d41f0f3
3
  size 12811532
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96780ce5a7a359b59f5fc9a4433d0e9b9865679baf2d652ec9bc4c907c7da710
3
  size 12811532
training_metrics.jsonl CHANGED
@@ -2287,3 +2287,147 @@
2287
  {"step": 22860, "total_steps": 28611, "tokens": 23970447360, "loss": 2.8080750107765198, "ce_loss": 2.8071916103363037, "z_loss": 8.833990097045898, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34626051783561707, "tokens_per_second": 2810334.170058564, "eta_seconds": 2145.780612230301, "data_wait_seconds": 0.30346882343292236}
2288
  {"step": 22870, "total_steps": 28611, "tokens": 23980933120, "loss": 2.820081889629364, "ce_loss": 2.8191118836402893, "z_loss": 9.700099229812622, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3906225264072418, "tokens_per_second": 4098046.93612174, "eta_seconds": 1468.9619005918503, "data_wait_seconds": 1.7583370208740234e-05}
2289
  {"step": 22880, "total_steps": 28611, "tokens": 23991418880, "loss": 2.79124915599823, "ce_loss": 2.7903621196746826, "z_loss": 8.870396137237549, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35821008682250977, "tokens_per_second": 4463639.696443234, "eta_seconds": 1346.2979686260223, "data_wait_seconds": 1.1324882507324219e-05}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2287
  {"step": 22860, "total_steps": 28611, "tokens": 23970447360, "loss": 2.8080750107765198, "ce_loss": 2.8071916103363037, "z_loss": 8.833990097045898, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34626051783561707, "tokens_per_second": 2810334.170058564, "eta_seconds": 2145.780612230301, "data_wait_seconds": 0.30346882343292236}
2288
  {"step": 22870, "total_steps": 28611, "tokens": 23980933120, "loss": 2.820081889629364, "ce_loss": 2.8191118836402893, "z_loss": 9.700099229812622, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3906225264072418, "tokens_per_second": 4098046.93612174, "eta_seconds": 1468.9619005918503, "data_wait_seconds": 1.7583370208740234e-05}
2289
  {"step": 22880, "total_steps": 28611, "tokens": 23991418880, "loss": 2.79124915599823, "ce_loss": 2.7903621196746826, "z_loss": 8.870396137237549, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35821008682250977, "tokens_per_second": 4463639.696443234, "eta_seconds": 1346.2979686260223, "data_wait_seconds": 1.1324882507324219e-05}
2290
+ {"step": 22890, "total_steps": 28611, "tokens": 24001904640, "loss": 2.786485493183136, "ce_loss": 2.7854408621788025, "z_loss": 10.44617748260498, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3955477178096771, "tokens_per_second": 8783550.11264714, "eta_seconds": 682.9702363014221, "data_wait_seconds": 2.568960189819336e-05}
2291
+ {"step": 22900, "total_steps": 28611, "tokens": 24012390400, "loss": 2.76044499874115, "ce_loss": 2.759416937828064, "z_loss": 10.280642986297607, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37307125329971313, "tokens_per_second": 3018296.7454650193, "eta_seconds": 1984.0386950016023, "data_wait_seconds": 1.9729137420654297e-05}
2292
+ {"step": 22910, "total_steps": 28611, "tokens": 24022876160, "loss": 2.7770649790763855, "ce_loss": 2.7759052515029907, "z_loss": 11.597739934921265, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34231486916542053, "tokens_per_second": 3915490.498674018, "eta_seconds": 1526.7389304161072, "data_wait_seconds": 1.5437602996826172e-05}
2293
+ {"step": 22920, "total_steps": 28611, "tokens": 24033361920, "loss": 2.7951242923736572, "ce_loss": 2.7940480709075928, "z_loss": 10.762537717819214, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3480699956417084, "tokens_per_second": 3137969.4577798806, "eta_seconds": 1901.6902797460557, "data_wait_seconds": 0.2560189962387085}
2294
+ {"step": 22930, "total_steps": 28611, "tokens": 24043847680, "loss": 2.7535558342933655, "ce_loss": 2.752236008644104, "z_loss": 13.197685956954956, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3283033072948456, "tokens_per_second": 4117447.7964575533, "eta_seconds": 1446.760359930992, "data_wait_seconds": 1.4781951904296875e-05}
2295
+ {"step": 22940, "total_steps": 28611, "tokens": 24054333440, "loss": 2.7911709547042847, "ce_loss": 2.7899685502052307, "z_loss": 12.024157285690308, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30929213762283325, "tokens_per_second": 3098458.072735781, "eta_seconds": 1919.1721677064895, "data_wait_seconds": 1.4662742614746094e-05}
2296
+ {"step": 22950, "total_steps": 28611, "tokens": 24064819200, "loss": 2.7694775462150574, "ce_loss": 2.7684786319732666, "z_loss": 9.989133358001709, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33206960558891296, "tokens_per_second": 2866624.011759422, "eta_seconds": 2070.724556708336, "data_wait_seconds": 1.9431114196777344e-05}
2297
+ {"step": 22960, "total_steps": 28611, "tokens": 24075304960, "loss": 2.7619075179100037, "ce_loss": 2.7608079314231873, "z_loss": 10.996137857437134, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.39873093366622925, "tokens_per_second": 4695888.899629693, "eta_seconds": 1261.8490561962128, "data_wait_seconds": 2.7954578399658203e-05}
2298
+ {"step": 22970, "total_steps": 28611, "tokens": 24085790720, "loss": 2.7697325944900513, "ce_loss": 2.7684386372566223, "z_loss": 12.940089702606201, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.42751049995422363, "tokens_per_second": 2976023.579969638, "eta_seconds": 1987.5572410821912, "data_wait_seconds": 1.4483928680419922e-05}
2299
+ {"step": 22980, "total_steps": 28611, "tokens": 24096276480, "loss": 2.790499985218048, "ce_loss": 2.7894675731658936, "z_loss": 10.325025081634521, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30660802125930786, "tokens_per_second": 3045278.836785028, "eta_seconds": 1938.913239955902, "data_wait_seconds": 1.9609928131103516e-05}
2300
+ {"step": 22990, "total_steps": 28611, "tokens": 24106762240, "loss": 2.7736083269119263, "ce_loss": 2.772643029689789, "z_loss": 9.652801275253296, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31868091225624084, "tokens_per_second": 3879634.262422108, "eta_seconds": 1519.2271480560303, "data_wait_seconds": 1.800060272216797e-05}
2301
+ {"step": 23000, "total_steps": 28611, "tokens": 24117248000, "loss": 2.749691128730774, "ce_loss": 2.7484692335128784, "z_loss": 12.21900200843811, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3050047755241394, "tokens_per_second": 8501686.42273773, "eta_seconds": 692.0462180614471, "data_wait_seconds": 1.341104507446289e-05}
2302
+ {"step": 23010, "total_steps": 28611, "tokens": 24127733760, "loss": 2.766323447227478, "ce_loss": 2.7652094960212708, "z_loss": 11.13914179801941, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30235669016838074, "tokens_per_second": 4230196.542494366, "eta_seconds": 1388.3691022396088, "data_wait_seconds": 1.4424324035644531e-05}
2303
+ {"step": 23020, "total_steps": 28611, "tokens": 24138219520, "loss": 2.7941039204597473, "ce_loss": 2.7928255796432495, "z_loss": 12.783050298690796, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3170047700405121, "tokens_per_second": 4174211.118606515, "eta_seconds": 1404.4781754970552, "data_wait_seconds": 1.4126300811767578e-05}
2304
+ {"step": 23030, "total_steps": 28611, "tokens": 24148705280, "loss": 2.7584911584854126, "ce_loss": 2.757317364215851, "z_loss": 11.737863302230835, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3615874648094177, "tokens_per_second": 4024599.865028171, "eta_seconds": 1454.0831019878387, "data_wait_seconds": 1.8477439880371094e-05}
2305
+ {"step": 23040, "total_steps": 28611, "tokens": 24159191040, "loss": 2.7871922850608826, "ce_loss": 2.7859935760498047, "z_loss": 11.986680507659912, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3636150658130646, "tokens_per_second": 4534887.653011141, "eta_seconds": 1288.1503011703492, "data_wait_seconds": 0.25157374143600464}
2306
+ {"step": 23050, "total_steps": 28611, "tokens": 24169676800, "loss": 2.799834728240967, "ce_loss": 2.7987576127052307, "z_loss": 10.771176099777222, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31954267621040344, "tokens_per_second": 6788249.616492649, "eta_seconds": 859.0036409139633, "data_wait_seconds": 1.4662742614746094e-05}
2307
+ {"step": 23060, "total_steps": 28611, "tokens": 24180162560, "loss": 2.8215218782424927, "ce_loss": 2.8205615282058716, "z_loss": 9.603160858154297, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.42624717950820923, "tokens_per_second": 3173188.2362570222, "eta_seconds": 1834.3208604812623, "data_wait_seconds": 1.5079975128173828e-05}
2308
+ {"step": 23070, "total_steps": 28611, "tokens": 24190648320, "loss": 2.775083541870117, "ce_loss": 2.7738779187202454, "z_loss": 12.056711673736572, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32341209053993225, "tokens_per_second": 2946082.24848407, "eta_seconds": 1972.1647686481474, "data_wait_seconds": 1.919269561767578e-05}
2309
+ {"step": 23080, "total_steps": 28611, "tokens": 24201134080, "loss": 2.769125282764435, "ce_loss": 2.7681179642677307, "z_loss": 10.073641061782837, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3053056299686432, "tokens_per_second": 4424475.674095643, "eta_seconds": 1310.8160792827607, "data_wait_seconds": 1.5556812286376953e-05}
2310
+ {"step": 23090, "total_steps": 28611, "tokens": 24211619840, "loss": 2.800717532634735, "ce_loss": 2.79966139793396, "z_loss": 10.561519145965576, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3589835464954376, "tokens_per_second": 3007188.465854379, "eta_seconds": 1925.116487288475, "data_wait_seconds": 0.27530306577682495}
2311
+ {"step": 23100, "total_steps": 28611, "tokens": 24222105600, "loss": 2.803210496902466, "ce_loss": 2.8019840717315674, "z_loss": 12.264556646347046, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.40144363045692444, "tokens_per_second": 4421659.054788381, "eta_seconds": 1306.9081682682038, "data_wait_seconds": 1.4483928680419922e-05}
2312
+ {"step": 23110, "total_steps": 28611, "tokens": 24232591360, "loss": 2.7002424001693726, "ce_loss": 2.6990705132484436, "z_loss": 11.718626976013184, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3565092980861664, "tokens_per_second": 2942472.075314713, "eta_seconds": 1960.330099439621, "data_wait_seconds": 1.531839370727539e-05}
2313
+ {"step": 23120, "total_steps": 28611, "tokens": 24243077120, "loss": 2.7867541909217834, "ce_loss": 2.7857643961906433, "z_loss": 9.897874355316162, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.449424147605896, "tokens_per_second": 2907743.4021626427, "eta_seconds": 1980.1371784448625, "data_wait_seconds": 0.5474514365196228}
2314
+ {"step": 23130, "total_steps": 28611, "tokens": 24253562880, "loss": 2.7833228707313538, "ce_loss": 2.7823328971862793, "z_loss": 9.899353504180908, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3487566113471985, "tokens_per_second": 4324567.779954788, "eta_seconds": 1328.9755990505219, "data_wait_seconds": 1.7642974853515625e-05}
2315
+ {"step": 23140, "total_steps": 28611, "tokens": 24264048640, "loss": 2.7963061332702637, "ce_loss": 2.795082747936249, "z_loss": 12.233836889266968, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3355173170566559, "tokens_per_second": 4368304.907143712, "eta_seconds": 1313.2689722776413, "data_wait_seconds": 1.2993812561035156e-05}
2316
+ {"step": 23150, "total_steps": 28611, "tokens": 24274534400, "loss": 2.801144599914551, "ce_loss": 2.800052762031555, "z_loss": 10.918463230133057, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.407394140958786, "tokens_per_second": 2941921.9451102572, "eta_seconds": 1946.4396550416946, "data_wait_seconds": 1.4841556549072266e-05}
2317
+ {"step": 23160, "total_steps": 28611, "tokens": 24285020160, "loss": 2.8105159401893616, "ce_loss": 2.80945885181427, "z_loss": 10.5712571144104, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30784955620765686, "tokens_per_second": 2650941.2767890366, "eta_seconds": 2156.135190939903, "data_wait_seconds": 0.10604768991470337}
2318
+ {"step": 23170, "total_steps": 28611, "tokens": 24295505920, "loss": 2.7671538591384888, "ce_loss": 2.765905797481537, "z_loss": 12.480711460113525, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3183639943599701, "tokens_per_second": 4092472.2487199297, "eta_seconds": 1394.0966900348662, "data_wait_seconds": 2.294778823852539e-05}
2319
+ {"step": 23180, "total_steps": 28611, "tokens": 24305991680, "loss": 2.7741856575012207, "ce_loss": 2.773166000843048, "z_loss": 10.196396827697754, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3530384302139282, "tokens_per_second": 4403718.05959295, "eta_seconds": 1293.1836641073226, "data_wait_seconds": 1.341104507446289e-05}
2320
+ {"step": 23190, "total_steps": 28611, "tokens": 24316477440, "loss": 2.7652631998062134, "ce_loss": 2.764122426509857, "z_loss": 11.4077627658844, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33675122261047363, "tokens_per_second": 4192117.9399627955, "eta_seconds": 1355.956721019745, "data_wait_seconds": 1.5795230865478516e-05}
2321
+ {"step": 23200, "total_steps": 28611, "tokens": 24326963200, "loss": 2.7702426314353943, "ce_loss": 2.7688764929771423, "z_loss": 13.66141128540039, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3058788776397705, "tokens_per_second": 2914287.465313024, "eta_seconds": 1946.9063376665115, "data_wait_seconds": 1.4781951904296875e-05}
2322
+ {"step": 23210, "total_steps": 28611, "tokens": 24337448960, "loss": 2.7945247888565063, "ce_loss": 2.793513000011444, "z_loss": 10.11855411529541, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2877140939235687, "tokens_per_second": 4270478.354646174, "eta_seconds": 1326.1650114297865, "data_wait_seconds": 1.6629695892333984e-05}
2323
+ {"step": 23220, "total_steps": 28611, "tokens": 24347934720, "loss": 2.780710220336914, "ce_loss": 2.7795392870903015, "z_loss": 11.709001541137695, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3477744162082672, "tokens_per_second": 3033121.718965969, "eta_seconds": 1863.7145949840544, "data_wait_seconds": 1.6927719116210938e-05}
2324
+ {"step": 23230, "total_steps": 28611, "tokens": 24358420480, "loss": 2.8042391538619995, "ce_loss": 2.8030863404273987, "z_loss": 11.528316974639893, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2641335129737854, "tokens_per_second": 4381849.442823547, "eta_seconds": 1287.6725979804994, "data_wait_seconds": 1.5437602996826172e-05}
2325
+ {"step": 23240, "total_steps": 28611, "tokens": 24368906240, "loss": 2.8033717274665833, "ce_loss": 2.8022576570510864, "z_loss": 11.140111446380615, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34659841656684875, "tokens_per_second": 3102911.443508101, "eta_seconds": 1815.037843823433, "data_wait_seconds": 1.4007091522216797e-05}
2326
+ {"step": 23250, "total_steps": 28611, "tokens": 24379392000, "loss": 2.8112115263938904, "ce_loss": 2.8103118538856506, "z_loss": 8.996346473693848, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3665536344051361, "tokens_per_second": 2819914.5873087747, "eta_seconds": 1993.470284986496, "data_wait_seconds": 1.430511474609375e-05}
2327
+ {"step": 23260, "total_steps": 28611, "tokens": 24389877760, "loss": 2.795059323310852, "ce_loss": 2.7939881086349487, "z_loss": 10.712098121643066, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37062355875968933, "tokens_per_second": 4036305.1725593987, "eta_seconds": 1390.115448689461, "data_wait_seconds": 1.5437602996826172e-05}
2328
+ {"step": 23270, "total_steps": 28611, "tokens": 24400363520, "loss": 2.7886054515838623, "ce_loss": 2.7875674962997437, "z_loss": 10.379583597183228, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.38265934586524963, "tokens_per_second": 3079439.463295367, "eta_seconds": 1818.6570909261704, "data_wait_seconds": 2.014636993408203e-05}
2329
+ {"step": 23280, "total_steps": 28611, "tokens": 24410849280, "loss": 2.7988938689231873, "ce_loss": 2.797650396823883, "z_loss": 12.434683084487915, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31940174102783203, "tokens_per_second": 2857155.2539554094, "eta_seconds": 1956.4770406723023, "data_wait_seconds": 2.2530555725097656e-05}
2330
+ {"step": 23290, "total_steps": 28611, "tokens": 24421335040, "loss": 2.7698240280151367, "ce_loss": 2.7687681913375854, "z_loss": 10.558617115020752, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32591748237609863, "tokens_per_second": 4124569.0242365105, "eta_seconds": 1352.7408229112625, "data_wait_seconds": 1.3768672943115234e-05}
2331
+ {"step": 23300, "total_steps": 28611, "tokens": 24431820800, "loss": 2.814791440963745, "ce_loss": 2.813551425933838, "z_loss": 12.4003586769104, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35541993379592896, "tokens_per_second": 3101866.4681794043, "eta_seconds": 1795.3664972782135, "data_wait_seconds": 1.4781951904296875e-05}
2332
+ {"step": 23310, "total_steps": 28611, "tokens": 24442306560, "loss": 2.7785382866859436, "ce_loss": 2.777260959148407, "z_loss": 12.773099899291992, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3290182054042816, "tokens_per_second": 2872550.0706072915, "eta_seconds": 1935.0407266616821, "data_wait_seconds": 1.996755599975586e-05}
2333
+ {"step": 23320, "total_steps": 28611, "tokens": 24452792320, "loss": 2.7657687067985535, "ce_loss": 2.764599919319153, "z_loss": 11.688108205795288, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33041250705718994, "tokens_per_second": 3158584.9337982535, "eta_seconds": 1756.4877096176149, "data_wait_seconds": 0.23629504442214966}
2334
+ {"step": 23330, "total_steps": 28611, "tokens": 24463278080, "loss": 2.8244171142578125, "ce_loss": 2.8235172629356384, "z_loss": 8.998570442199707, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27507612109184265, "tokens_per_second": 3698698.767385929, "eta_seconds": 1497.1562174320222, "data_wait_seconds": 1.4543533325195312e-05}
2335
+ {"step": 23340, "total_steps": 28611, "tokens": 24473763840, "loss": 2.805329978466034, "ce_loss": 2.8043810725212097, "z_loss": 9.488694906234741, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3716217577457428, "tokens_per_second": 3068139.1842672955, "eta_seconds": 1801.4319964170456, "data_wait_seconds": 1.704692840576172e-05}
2336
+ {"step": 23350, "total_steps": 28611, "tokens": 24484249600, "loss": 2.7831115126609802, "ce_loss": 2.782068610191345, "z_loss": 10.429091691970825, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34639260172843933, "tokens_per_second": 2872497.538419738, "eta_seconds": 1920.4745216369631, "data_wait_seconds": 1.4483928680419922e-05}
2337
+ {"step": 23360, "total_steps": 28611, "tokens": 24494735360, "loss": 2.802032172679901, "ce_loss": 2.801016330718994, "z_loss": 10.159053325653076, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3555278182029724, "tokens_per_second": 3157059.01822102, "eta_seconds": 1744.0512021541595, "data_wait_seconds": 1.4007091522216797e-05}
2338
+ {"step": 23370, "total_steps": 28611, "tokens": 24505221120, "loss": 2.8158161640167236, "ce_loss": 2.8147597908973694, "z_loss": 10.563305139541626, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.40496230125427246, "tokens_per_second": 2850277.264693403, "eta_seconds": 1928.08850004673, "data_wait_seconds": 1.710653305053711e-05}
2339
+ {"step": 23380, "total_steps": 28611, "tokens": 24515706880, "loss": 2.810714066028595, "ce_loss": 2.8095551133155823, "z_loss": 11.589900493621826, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37661343812942505, "tokens_per_second": 2992502.296815802, "eta_seconds": 1832.9479853153227, "data_wait_seconds": 1.4007091522216797e-05}
2340
+ {"step": 23390, "total_steps": 28611, "tokens": 24526192640, "loss": 2.7866073846817017, "ce_loss": 2.7854697704315186, "z_loss": 11.376120805740356, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3301549255847931, "tokens_per_second": 2919314.086002163, "eta_seconds": 1875.3087659358978, "data_wait_seconds": 1.430511474609375e-05}
2341
+ {"step": 23400, "total_steps": 28611, "tokens": 24536678400, "loss": 2.8479902744293213, "ce_loss": 2.8468226194381714, "z_loss": 11.676147937774658, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.40192246437072754, "tokens_per_second": 2824980.8080001995, "eta_seconds": 1934.2182858467102, "data_wait_seconds": 1.5854835510253906e-05}
2342
+ {"step": 23410, "total_steps": 28611, "tokens": 24547164160, "loss": 2.8258418440818787, "ce_loss": 2.8250556588172913, "z_loss": 7.861978530883789, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35380667448043823, "tokens_per_second": 4597974.247961592, "eta_seconds": 1186.0970683813093, "data_wait_seconds": 1.901388168334961e-05}
2343
+ {"step": 23420, "total_steps": 28611, "tokens": 24557649920, "loss": 2.831709086894989, "ce_loss": 2.830726444721222, "z_loss": 9.826154947280884, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3163912892341614, "tokens_per_second": 2864814.252569854, "eta_seconds": 1900.0038173913956, "data_wait_seconds": 1.4185905456542969e-05}
2344
+ {"step": 23430, "total_steps": 28611, "tokens": 24568135680, "loss": 2.7711713910102844, "ce_loss": 2.7699690461158752, "z_loss": 12.023261547088623, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32494625449180603, "tokens_per_second": 2862135.9306935715, "eta_seconds": 1898.1181843042373, "data_wait_seconds": 1.4901161193847656e-05}
2345
+ {"step": 23440, "total_steps": 28611, "tokens": 24578621440, "loss": 2.7949939370155334, "ce_loss": 2.7937931418418884, "z_loss": 12.007441282272339, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2829321622848511, "tokens_per_second": 4036996.1459890143, "eta_seconds": 1343.124020910263, "data_wait_seconds": 1.7642974853515625e-05}
2346
+ {"step": 23450, "total_steps": 28611, "tokens": 24589107200, "loss": 2.809710919857025, "ce_loss": 2.808863043785095, "z_loss": 8.479070901870728, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3108294904232025, "tokens_per_second": 2667469.109332507, "eta_seconds": 2028.7772844552994, "data_wait_seconds": 1.5079975128173828e-05}
2347
+ {"step": 23460, "total_steps": 28611, "tokens": 24599592960, "loss": 2.8095061779022217, "ce_loss": 2.8084622621536255, "z_loss": 10.439602375030518, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34389227628707886, "tokens_per_second": 3253734.698326941, "eta_seconds": 1660.0047258853913, "data_wait_seconds": 1.3828277587890625e-05}
2348
+ {"step": 23470, "total_steps": 28611, "tokens": 24610078720, "loss": 2.8191727995872498, "ce_loss": 2.817956566810608, "z_loss": 12.162458658218384, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33253392577171326, "tokens_per_second": 2884064.7801471422, "eta_seconds": 1869.142903137207, "data_wait_seconds": 1.895427703857422e-05}
2349
+ {"step": 23480, "total_steps": 28611, "tokens": 24620564480, "loss": 2.777974307537079, "ce_loss": 2.777060806751251, "z_loss": 9.135586977005005, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34372663497924805, "tokens_per_second": 3127953.2805531765, "eta_seconds": 1720.0523708105088, "data_wait_seconds": 1.2218952178955078e-05}
2350
+ {"step": 23490, "total_steps": 28611, "tokens": 24631050240, "loss": 2.796201169490814, "ce_loss": 2.7951819896698, "z_loss": 10.191922426223755, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3090410828590393, "tokens_per_second": 2930373.0023579327, "eta_seconds": 1832.4485284566879, "data_wait_seconds": 1.4603137969970703e-05}
2351
+ {"step": 23500, "total_steps": 28611, "tokens": 24641536000, "loss": 2.817646622657776, "ce_loss": 2.8166475892066956, "z_loss": 9.9895920753479, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3144966959953308, "tokens_per_second": 3958453.1622233996, "eta_seconds": 1353.880345773697, "data_wait_seconds": 1.3589859008789062e-05}
2352
+ {"step": 23510, "total_steps": 28611, "tokens": 24652021760, "loss": 2.7911496758461, "ce_loss": 2.790174126625061, "z_loss": 9.754876613616943, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2571480870246887, "tokens_per_second": 2956520.0601623286, "eta_seconds": 1809.1492928028106, "data_wait_seconds": 1.6868114471435547e-05}
2353
+ {"step": 23520, "total_steps": 28611, "tokens": 24662507520, "loss": 2.790510058403015, "ce_loss": 2.7892624735832214, "z_loss": 12.47588324546814, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3963746130466461, "tokens_per_second": 3238506.6526779584, "eta_seconds": 1648.3833410024642, "data_wait_seconds": 0.24434524774551392}
2354
+ {"step": 23530, "total_steps": 28611, "tokens": 24672993280, "loss": 2.79267680644989, "ce_loss": 2.7915475368499756, "z_loss": 11.29244327545166, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.38177359104156494, "tokens_per_second": 4075710.0177854025, "eta_seconds": 1307.2114141464233, "data_wait_seconds": 1.2516975402832031e-05}
2355
+ {"step": 23540, "total_steps": 28611, "tokens": 24683479040, "loss": 2.8065168261528015, "ce_loss": 2.805590808391571, "z_loss": 9.260204076766968, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35205793380737305, "tokens_per_second": 2802879.90704581, "eta_seconds": 1897.0947997570036, "data_wait_seconds": 1.4603137969970703e-05}
2356
+ {"step": 23550, "total_steps": 28611, "tokens": 24693964800, "loss": 2.8205146193504333, "ce_loss": 2.819465219974518, "z_loss": 10.493902206420898, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34254229068756104, "tokens_per_second": 2841835.505925694, "eta_seconds": 1867.399828362465, "data_wait_seconds": 0.2632169723510742}
2357
+ {"step": 23560, "total_steps": 28611, "tokens": 24704450560, "loss": 2.801864743232727, "ce_loss": 2.800811290740967, "z_loss": 10.534888982772827, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33140939474105835, "tokens_per_second": 3951454.328350163, "eta_seconds": 1340.3564702749252, "data_wait_seconds": 1.704692840576172e-05}
2358
+ {"step": 23570, "total_steps": 28611, "tokens": 24714936320, "loss": 2.7998937368392944, "ce_loss": 2.7988653779029846, "z_loss": 10.28418779373169, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3638182580471039, "tokens_per_second": 3039804.4365497297, "eta_seconds": 1738.885420536995, "data_wait_seconds": 1.3589859008789062e-05}
2359
+ {"step": 23580, "total_steps": 28611, "tokens": 24725422080, "loss": 2.841962218284607, "ce_loss": 2.8408267498016357, "z_loss": 11.354425191879272, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3200906217098236, "tokens_per_second": 2782973.816596245, "eta_seconds": 1895.5930611133576, "data_wait_seconds": 1.7642974853515625e-05}
2360
+ {"step": 23590, "total_steps": 28611, "tokens": 24735907840, "loss": 2.7921574115753174, "ce_loss": 2.791225492954254, "z_loss": 9.319166898727417, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.39235061407089233, "tokens_per_second": 1926422.5245136644, "eta_seconds": 2732.9934264183044, "data_wait_seconds": 0.3607966899871826}
2361
+ {"step": 23600, "total_steps": 28611, "tokens": 24746393600, "loss": 2.86216139793396, "ce_loss": 2.8611807823181152, "z_loss": 9.80666732788086, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.40159663558006287, "tokens_per_second": 2942475.028268001, "eta_seconds": 1785.7124650239946, "data_wait_seconds": 1.5020370483398438e-05}
2362
+ {"step": 23610, "total_steps": 28611, "tokens": 24756879360, "loss": 2.814050078392029, "ce_loss": 2.8130794763565063, "z_loss": 9.706179857254028, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3558671772480011, "tokens_per_second": 4235204.996587701, "eta_seconds": 1238.1758569478989, "data_wait_seconds": 1.4543533325195312e-05}
2363
+ {"step": 23620, "total_steps": 28611, "tokens": 24767365120, "loss": 2.7785057425498962, "ce_loss": 2.77756130695343, "z_loss": 9.444878101348877, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3073768615722656, "tokens_per_second": 3080580.0696870554, "eta_seconds": 1698.8497937440873, "data_wait_seconds": 1.7940998077392578e-05}
2364
+ {"step": 23630, "total_steps": 28611, "tokens": 24777850880, "loss": 2.809764325618744, "ce_loss": 2.808649778366089, "z_loss": 11.145405054092407, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34095096588134766, "tokens_per_second": 4212184.300081341, "eta_seconds": 1239.9640385866167, "data_wait_seconds": 1.436471939086914e-05}
2365
+ {"step": 23640, "total_steps": 28611, "tokens": 24788336640, "loss": 2.783549726009369, "ce_loss": 2.7825005650520325, "z_loss": 10.491706609725952, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.39818286895751953, "tokens_per_second": 2958072.6961908527, "eta_seconds": 1762.1173755168913, "data_wait_seconds": 1.3887882232666016e-05}
2366
+ {"step": 23650, "total_steps": 28611, "tokens": 24798822400, "loss": 2.7974682450294495, "ce_loss": 2.796446979045868, "z_loss": 10.2121901512146, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3012678623199463, "tokens_per_second": 2670167.9887667135, "eta_seconds": 1948.1866151809693, "data_wait_seconds": 1.8298625946044922e-05}
2367
+ {"step": 23660, "total_steps": 28611, "tokens": 24809308160, "loss": 2.776791214942932, "ce_loss": 2.775840401649475, "z_loss": 9.508683919906616, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.5165886282920837, "tokens_per_second": 3081653.070846721, "eta_seconds": 1684.6477058410644, "data_wait_seconds": 0.19178807735443115}
2368
+ {"step": 23670, "total_steps": 28611, "tokens": 24819793920, "loss": 2.77731853723526, "ce_loss": 2.776157796382904, "z_loss": 11.607421636581421, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35861316323280334, "tokens_per_second": 3887254.986296643, "eta_seconds": 1332.8207267761231, "data_wait_seconds": 1.3589859008789062e-05}
2369
+ {"step": 23680, "total_steps": 28611, "tokens": 24830279680, "loss": 2.845184564590454, "ce_loss": 2.844114124774933, "z_loss": 10.70432424545288, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.356754869222641, "tokens_per_second": 2342879.8254418503, "eta_seconds": 2206.91142578125, "data_wait_seconds": 0.24981731176376343}
2370
+ {"step": 23690, "total_steps": 28611, "tokens": 24840765440, "loss": 2.8124337792396545, "ce_loss": 2.81136554479599, "z_loss": 10.682424068450928, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3399955928325653, "tokens_per_second": 3951861.57947354, "eta_seconds": 1305.7245028018951, "data_wait_seconds": 1.2874603271484375e-05}
2371
+ {"step": 23700, "total_steps": 28611, "tokens": 24851251200, "loss": 2.8012478351593018, "ce_loss": 2.800027310848236, "z_loss": 12.20562481880188, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37360504269599915, "tokens_per_second": 2939379.0631433167, "eta_seconds": 1751.9199209690094, "data_wait_seconds": 2.1219253540039062e-05}
2372
+ {"step": 23710, "total_steps": 28611, "tokens": 24861736960, "loss": 2.772055208683014, "ce_loss": 2.770959436893463, "z_loss": 10.957678318023682, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3454967439174652, "tokens_per_second": 2937300.1321998155, "eta_seconds": 1749.5900128364563, "data_wait_seconds": 0.27215301990509033}
2373
+ {"step": 23720, "total_steps": 28611, "tokens": 24872222720, "loss": 2.838987171649933, "ce_loss": 2.8377782702445984, "z_loss": 12.088899612426758, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3314855992794037, "tokens_per_second": 4116591.448816137, "eta_seconds": 1245.8329372167586, "data_wait_seconds": 1.3947486877441406e-05}
2374
+ {"step": 23730, "total_steps": 28611, "tokens": 24882708480, "loss": 2.8047048449516296, "ce_loss": 2.8034574389457703, "z_loss": 12.474591255187988, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2844815254211426, "tokens_per_second": 2956191.3682168294, "eta_seconds": 1731.315337371826, "data_wait_seconds": 1.4781951904296875e-05}
2375
+ {"step": 23740, "total_steps": 28611, "tokens": 24893194240, "loss": 2.8010770678520203, "ce_loss": 2.800033450126648, "z_loss": 10.436212062835693, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.39387843012809753, "tokens_per_second": 4266331.214259911, "eta_seconds": 1197.1910851478576, "data_wait_seconds": 1.5556812286376953e-05}
2376
+ {"step": 23750, "total_steps": 28611, "tokens": 24903680000, "loss": 2.791708767414093, "ce_loss": 2.7906234860420227, "z_loss": 10.85245418548584, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35645604133605957, "tokens_per_second": 3070349.4304862227, "eta_seconds": 1660.1133035182954, "data_wait_seconds": 1.5437602996826172e-05}
2377
+ {"step": 23760, "total_steps": 28611, "tokens": 24914165760, "loss": 2.7946845293045044, "ce_loss": 2.7935112714767456, "z_loss": 11.732153177261353, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3794766366481781, "tokens_per_second": 2582638.5755354483, "eta_seconds": 1969.5524662971497, "data_wait_seconds": 1.8775463104248047e-05}
2378
+ {"step": 23770, "total_steps": 28611, "tokens": 24924651520, "loss": 2.8323238492012024, "ce_loss": 2.8311474323272705, "z_loss": 11.76402497291565, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31344515085220337, "tokens_per_second": 2826568.8953483254, "eta_seconds": 1795.8721700906754, "data_wait_seconds": 1.7940998077392578e-05}
2379
+ {"step": 23780, "total_steps": 28611, "tokens": 24935137280, "loss": 2.7801464796066284, "ce_loss": 2.778996765613556, "z_loss": 11.496746301651001, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.38642266392707825, "tokens_per_second": 2934489.9143182044, "eta_seconds": 1726.2525358438493, "data_wait_seconds": 1.4424324035644531e-05}
2380
+ {"step": 23790, "total_steps": 28611, "tokens": 24945623040, "loss": 2.850745916366577, "ce_loss": 2.8494834303855896, "z_loss": 12.625140190124512, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37250521779060364, "tokens_per_second": 2840717.2865145705, "eta_seconds": 1779.5452296495437, "data_wait_seconds": 0.2628834843635559}
2381
+ {"step": 23800, "total_steps": 28611, "tokens": 24956108800, "loss": 2.8316383957862854, "ce_loss": 2.8307301998138428, "z_loss": 9.082080364227295, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3706226348876953, "tokens_per_second": 4036292.9483495792, "eta_seconds": 1249.834737110138, "data_wait_seconds": 1.3589859008789062e-05}
2382
+ {"step": 23810, "total_steps": 28611, "tokens": 24966594560, "loss": 2.8250683546066284, "ce_loss": 2.82406222820282, "z_loss": 10.061400651931763, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32023271918296814, "tokens_per_second": 2852363.1728812787, "eta_seconds": 1764.9272097826004, "data_wait_seconds": 1.5556812286376953e-05}
2383
+ {"step": 23820, "total_steps": 28611, "tokens": 24977080320, "loss": 2.8031316995620728, "ce_loss": 2.8018998503684998, "z_loss": 12.318760871887207, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3320731520652771, "tokens_per_second": 4261087.489242272, "eta_seconds": 1178.9778146266938, "data_wait_seconds": 1.7762184143066406e-05}
2384
+ {"step": 23830, "total_steps": 28611, "tokens": 24987566080, "loss": 2.834014356136322, "ce_loss": 2.833008587360382, "z_loss": 10.057032346725464, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28644073009490967, "tokens_per_second": 3004539.150106978, "eta_seconds": 1668.5560099363329, "data_wait_seconds": 1.6927719116210938e-05}
2385
+ {"step": 23840, "total_steps": 28611, "tokens": 24998051840, "loss": 2.817663311958313, "ce_loss": 2.8167211413383484, "z_loss": 9.421175956726074, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32987770438194275, "tokens_per_second": 3039430.9199903496, "eta_seconds": 1645.9515704393386, "data_wait_seconds": 0.25241100788116455}
2386
+ {"step": 23850, "total_steps": 28611, "tokens": 25008537600, "loss": 2.7815043926239014, "ce_loss": 2.7803231477737427, "z_loss": 11.812742471694946, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.39670202136039734, "tokens_per_second": 3946834.8757092203, "eta_seconds": 1264.8794523239135, "data_wait_seconds": 1.329183578491211e-05}
2387
+ {"step": 23860, "total_steps": 28611, "tokens": 25019023360, "loss": 2.799402952194214, "ce_loss": 2.7981414198875427, "z_loss": 12.615334510803223, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2869165241718292, "tokens_per_second": 3087215.448930991, "eta_seconds": 1613.6821865558625, "data_wait_seconds": 1.823902130126953e-05}
2388
+ {"step": 23870, "total_steps": 28611, "tokens": 25029509120, "loss": 2.8073307275772095, "ce_loss": 2.806241512298584, "z_loss": 10.891721248626709, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.4159943163394928, "tokens_per_second": 4655026.80010089, "eta_seconds": 1067.9420397520064, "data_wait_seconds": 1.2993812561035156e-05}
2389
+ {"step": 23880, "total_steps": 28611, "tokens": 25039994880, "loss": 2.7726939916610718, "ce_loss": 2.7718604803085327, "z_loss": 8.335806846618652, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3575029671192169, "tokens_per_second": 3283495.2438006764, "eta_seconds": 1510.8330262899399, "data_wait_seconds": 1.3887882232666016e-05}
2390
+ {"step": 23890, "total_steps": 28611, "tokens": 25050480640, "loss": 2.7718435525894165, "ce_loss": 2.770854115486145, "z_loss": 9.893960952758789, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32428455352783203, "tokens_per_second": 3939926.442139818, "eta_seconds": 1256.4517050504685, "data_wait_seconds": 1.4424324035644531e-05}
2391
+ {"step": 23900, "total_steps": 28611, "tokens": 25060966400, "loss": 2.7674800157546997, "ce_loss": 2.7662802934646606, "z_loss": 11.996886253356934, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3682052493095398, "tokens_per_second": 3911831.306597432, "eta_seconds": 1262.7951332330704, "data_wait_seconds": 1.4960765838623047e-05}
2392
+ {"step": 23910, "total_steps": 28611, "tokens": 25071452160, "loss": 2.769427001476288, "ce_loss": 2.7682942152023315, "z_loss": 11.327844619750977, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3710947632789612, "tokens_per_second": 4301711.404876356, "eta_seconds": 1145.9057365894319, "data_wait_seconds": 0.30046284198760986}
2393
+ {"step": 23920, "total_steps": 28611, "tokens": 25081937920, "loss": 2.7636667490005493, "ce_loss": 2.762704908847809, "z_loss": 9.618651390075684, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3091283142566681, "tokens_per_second": 4650946.862810891, "eta_seconds": 1057.6061522722246, "data_wait_seconds": 1.4960765838623047e-05}
2394
+ {"step": 23930, "total_steps": 28611, "tokens": 25092423680, "loss": 2.7420926690101624, "ce_loss": 2.740802586078644, "z_loss": 12.900353193283081, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3233064115047455, "tokens_per_second": 3836052.204214327, "eta_seconds": 1279.5405262231827, "data_wait_seconds": 1.4662742614746094e-05}
2395
+ {"step": 23940, "total_steps": 28611, "tokens": 25102909440, "loss": 2.7916126251220703, "ce_loss": 2.79028457403183, "z_loss": 13.280364990234375, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27841225266456604, "tokens_per_second": 3905702.6217253343, "eta_seconds": 1254.0377418279647, "data_wait_seconds": 1.5139579772949219e-05}
2396
+ {"step": 23950, "total_steps": 28611, "tokens": 25113395200, "loss": 2.82623553276062, "ce_loss": 2.825152575969696, "z_loss": 10.830028057098389, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3682011365890503, "tokens_per_second": 4522320.790902149, "eta_seconds": 1080.7311028957367, "data_wait_seconds": 0.2696409821510315}
2397
+ {"step": 23960, "total_steps": 28611, "tokens": 25123880960, "loss": 2.771929144859314, "ce_loss": 2.7707204818725586, "z_loss": 12.087159872055054, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37973713874816895, "tokens_per_second": 6482126.75047849, "eta_seconds": 752.365259695053, "data_wait_seconds": 1.4901161193847656e-05}
2398
+ {"step": 23970, "total_steps": 28611, "tokens": 25134366720, "loss": 2.7808934450149536, "ce_loss": 2.7796508073806763, "z_loss": 12.42634916305542, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3703615069389343, "tokens_per_second": 2921341.041507513, "eta_seconds": 1665.8244096994401, "data_wait_seconds": 1.4126300811767578e-05}
2399
+ {"step": 23980, "total_steps": 28611, "tokens": 25144852480, "loss": 2.7547059059143066, "ce_loss": 2.7537059783935547, "z_loss": 9.999518156051636, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3884282410144806, "tokens_per_second": 4072434.595756805, "eta_seconds": 1192.396180176735, "data_wait_seconds": 1.71661376953125e-05}
2400
+ {"step": 23990, "total_steps": 28611, "tokens": 25155338240, "loss": 2.7291308641433716, "ce_loss": 2.7280872464179993, "z_loss": 10.436176538467407, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3485441207885742, "tokens_per_second": 3003924.3288892163, "eta_seconds": 1613.0465236425398, "data_wait_seconds": 1.621246337890625e-05}
2401
+ {"step": 24000, "total_steps": 28611, "tokens": 25165824000, "loss": 2.808784067630768, "ce_loss": 2.807703375816345, "z_loss": 10.806834697723389, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3057137131690979, "tokens_per_second": 2774302.128843713, "eta_seconds": 1742.774835419655, "data_wait_seconds": 0.28155022859573364}
2402
+ {"step": 24010, "total_steps": 28611, "tokens": 25176309760, "loss": 2.719568431377411, "ce_loss": 2.718437135219574, "z_loss": 11.312816143035889, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33096110820770264, "tokens_per_second": 4436634.584062341, "eta_seconds": 1087.4229293823241, "data_wait_seconds": 1.5676021575927734e-05}
2403
+ {"step": 24020, "total_steps": 28611, "tokens": 25186795520, "loss": 2.7711127400398254, "ce_loss": 2.7699177265167236, "z_loss": 11.950648307800293, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3901277482509613, "tokens_per_second": 3983547.6251305756, "eta_seconds": 1208.4736694574356, "data_wait_seconds": 1.3113021850585938e-05}
2404
+ {"step": 24030, "total_steps": 28611, "tokens": 25197281280, "loss": 2.763829827308655, "ce_loss": 2.762730360031128, "z_loss": 10.994549036026001, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3894810378551483, "tokens_per_second": 2773324.027076134, "eta_seconds": 1732.046673631668, "data_wait_seconds": 1.430511474609375e-05}
2405
+ {"step": 24040, "total_steps": 28611, "tokens": 25207767040, "loss": 2.742172956466675, "ce_loss": 2.741283893585205, "z_loss": 8.890764474868774, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34632226824760437, "tokens_per_second": 2899469.577451938, "eta_seconds": 1653.075077342987, "data_wait_seconds": 0.2635232210159302}
2406
+ {"step": 24050, "total_steps": 28611, "tokens": 25218252800, "loss": 2.804338812828064, "ce_loss": 2.8031694293022156, "z_loss": 11.693650722503662, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3323540687561035, "tokens_per_second": 4439492.279174691, "eta_seconds": 1077.2752457380295, "data_wait_seconds": 0.5546951293945312}
2407
+ {"step": 24060, "total_steps": 28611, "tokens": 25228738560, "loss": 2.8042436242103577, "ce_loss": 2.8032639026641846, "z_loss": 9.79741096496582, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33724185824394226, "tokens_per_second": 4103781.5705616223, "eta_seconds": 1162.846826505661, "data_wait_seconds": 1.52587890625e-05}
2408
+ {"step": 24070, "total_steps": 28611, "tokens": 25239224320, "loss": 2.7619733810424805, "ce_loss": 2.7609679102897644, "z_loss": 10.054909706115723, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35237082839012146, "tokens_per_second": 2911015.416770528, "eta_seconds": 1635.7122633457184, "data_wait_seconds": 1.4901161193847656e-05}
2409
+ {"step": 24080, "total_steps": 28611, "tokens": 25249710080, "loss": 2.7491530179977417, "ce_loss": 2.747893452644348, "z_loss": 12.595803260803223, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3254718482494354, "tokens_per_second": 3072565.6539863776, "eta_seconds": 1546.2966104030609, "data_wait_seconds": 0.2047799825668335}
2410
+ {"step": 24090, "total_steps": 28611, "tokens": 25260195840, "loss": 2.755694270133972, "ce_loss": 2.7544859647750854, "z_loss": 12.082912921905518, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35657742619514465, "tokens_per_second": 4042695.961689105, "eta_seconds": 1172.636315202713, "data_wait_seconds": 1.800060272216797e-05}
2411
+ {"step": 24100, "total_steps": 28611, "tokens": 25270681600, "loss": 2.7437599897384644, "ce_loss": 2.742568850517273, "z_loss": 11.91091275215149, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3076075613498688, "tokens_per_second": 4249837.672126269, "eta_seconds": 1113.0134139060972, "data_wait_seconds": 1.2814998626708984e-05}
2412
+ {"step": 24110, "total_steps": 28611, "tokens": 25281167360, "loss": 2.780550539493561, "ce_loss": 2.7794641852378845, "z_loss": 10.863816499710083, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.402352511882782, "tokens_per_second": 2855822.2483758577, "eta_seconds": 1652.6380725145339, "data_wait_seconds": 1.4424324035644531e-05}
2413
+ {"step": 24120, "total_steps": 28611, "tokens": 25291653120, "loss": 2.736139476299286, "ce_loss": 2.734799385070801, "z_loss": 13.400541543960571, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.36710935831069946, "tokens_per_second": 2725119.2435277863, "eta_seconds": 1728.0545896053313, "data_wait_seconds": 1.4722347259521484e-05}
2414
+ {"step": 24130, "total_steps": 28611, "tokens": 25302138880, "loss": 2.7594234347343445, "ce_loss": 2.7583211064338684, "z_loss": 11.02393913269043, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.37351346015930176, "tokens_per_second": 3825141.6571211633, "eta_seconds": 1228.3647187948227, "data_wait_seconds": 1.5795230865478516e-05}
2415
+ {"step": 24140, "total_steps": 28611, "tokens": 25312624640, "loss": 2.7691086530685425, "ce_loss": 2.768048942089081, "z_loss": 10.597534894943237, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25871023535728455, "tokens_per_second": 2972430.700183183, "eta_seconds": 1577.2220680236817, "data_wait_seconds": 0.2565252184867859}
2416
+ {"step": 24150, "total_steps": 28611, "tokens": 25323110400, "loss": 2.7591832280158997, "ce_loss": 2.7580214738845825, "z_loss": 11.617713928222656, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29463890194892883, "tokens_per_second": 3697761.1723073744, "eta_seconds": 1265.0080191850661, "data_wait_seconds": 1.2874603271484375e-05}
2417
+ {"step": 24160, "total_steps": 28611, "tokens": 25333596160, "loss": 2.766149640083313, "ce_loss": 2.7652008533477783, "z_loss": 9.487737894058228, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3664793074131012, "tokens_per_second": 2846742.7833950045, "eta_seconds": 1639.4919144868852, "data_wait_seconds": 0.5420861840248108}
2418
+ {"step": 24170, "total_steps": 28611, "tokens": 25344081920, "loss": 2.7385606169700623, "ce_loss": 2.737536132335663, "z_loss": 10.244540452957153, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3529835343360901, "tokens_per_second": 3966400.587345747, "eta_seconds": 1174.0432952880858, "data_wait_seconds": 1.4662742614746094e-05}
2419
+ {"step": 24180, "total_steps": 28611, "tokens": 25354567680, "loss": 2.796099007129669, "ce_loss": 2.795050323009491, "z_loss": 10.486486911773682, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2998255491256714, "tokens_per_second": 2719931.4762033867, "eta_seconds": 1708.2195991516114, "data_wait_seconds": 1.7940998077392578e-05}
2420
+ {"step": 24190, "total_steps": 28611, "tokens": 25365053440, "loss": 2.7868918776512146, "ce_loss": 2.7859427332878113, "z_loss": 9.491708040237427, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3295270800590515, "tokens_per_second": 4285104.441840845, "eta_seconds": 1081.8299901247024, "data_wait_seconds": 1.7642974853515625e-05}
2421
+ {"step": 24200, "total_steps": 28611, "tokens": 25375539200, "loss": 2.8181658387184143, "ce_loss": 2.817130446434021, "z_loss": 10.354123592376709, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.39476478099823, "tokens_per_second": 2681896.1006777296, "eta_seconds": 1724.626369690895, "data_wait_seconds": 1.7702579498291016e-05}
2422
+ {"step": 24210, "total_steps": 28611, "tokens": 25386024960, "loss": 2.7513890862464905, "ce_loss": 2.7503998279571533, "z_loss": 9.892839193344116, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3832443058490753, "tokens_per_second": 3984035.8627411798, "eta_seconds": 1158.3186334133147, "data_wait_seconds": 1.3709068298339844e-05}
2423
+ {"step": 24220, "total_steps": 28611, "tokens": 25396510720, "loss": 2.794589936733246, "ce_loss": 2.7936036586761475, "z_loss": 9.86250114440918, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28259703516960144, "tokens_per_second": 2828486.708795967, "eta_seconds": 1627.8305999040601, "data_wait_seconds": 1.8417835235595703e-05}
2424
+ {"step": 24230, "total_steps": 28611, "tokens": 25406996480, "loss": 2.7565749287605286, "ce_loss": 2.7556071281433105, "z_loss": 9.677725076675415, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3372759222984314, "tokens_per_second": 4322326.270565721, "eta_seconds": 1062.8099704742433, "data_wait_seconds": 1.6570091247558594e-05}
2425
+ {"step": 24240, "total_steps": 28611, "tokens": 25417482240, "loss": 2.7771692276000977, "ce_loss": 2.7760576605796814, "z_loss": 11.115618467330933, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.416414737701416, "tokens_per_second": 2756088.215387207, "eta_seconds": 1662.9822189331053, "data_wait_seconds": 1.7762184143066406e-05}
2426
+ {"step": 24250, "total_steps": 28611, "tokens": 25427968000, "loss": 2.730824053287506, "ce_loss": 2.7298033237457275, "z_loss": 10.207448720932007, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32271426916122437, "tokens_per_second": 4014076.7849757196, "eta_seconds": 1139.2009124279023, "data_wait_seconds": 1.3113021850585938e-05}
2427
+ {"step": 24260, "total_steps": 28611, "tokens": 25438453760, "loss": 2.8236056566238403, "ce_loss": 2.8226025700569153, "z_loss": 10.03083062171936, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3580639064311981, "tokens_per_second": 2792813.0029559815, "eta_seconds": 1633.6053187847137, "data_wait_seconds": 1.4066696166992188e-05}
2428
+ {"step": 24270, "total_steps": 28611, "tokens": 25448939520, "loss": 2.825295865535736, "ce_loss": 2.8240415453910828, "z_loss": 12.542971134185791, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3696603775024414, "tokens_per_second": 2816176.329387975, "eta_seconds": 1616.329335808754, "data_wait_seconds": 0.2777951955795288}
2429
+ {"step": 24280, "total_steps": 28611, "tokens": 25459425280, "loss": 2.7577514052391052, "ce_loss": 2.7566912174224854, "z_loss": 10.601874828338623, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.41465333104133606, "tokens_per_second": 4311557.529376219, "eta_seconds": 1053.3044323444367, "data_wait_seconds": 1.811981201171875e-05}
2430
+ {"step": 24290, "total_steps": 28611, "tokens": 25469911040, "loss": 2.796673536300659, "ce_loss": 2.795699715614319, "z_loss": 9.738670825958252, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3445846140384674, "tokens_per_second": 2804883.0412518806, "eta_seconds": 1615.360365962982, "data_wait_seconds": 0.2634139060974121}
2431
+ {"step": 24300, "total_steps": 28611, "tokens": 25480396800, "loss": 2.767670750617981, "ce_loss": 2.766706645488739, "z_loss": 9.641210317611694, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35753899812698364, "tokens_per_second": 3445330.601046373, "eta_seconds": 1312.0398764133454, "data_wait_seconds": 1.436471939086914e-05}
2432
+ {"step": 24310, "total_steps": 28611, "tokens": 25490882560, "loss": 2.7835773229599, "ce_loss": 2.7823426127433777, "z_loss": 12.34714150428772, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3897356688976288, "tokens_per_second": 2740980.8077307055, "eta_seconds": 1645.3691916704179, "data_wait_seconds": 1.3709068298339844e-05}
2433
+ {"step": 24320, "total_steps": 28611, "tokens": 25501368320, "loss": 2.787176012992859, "ce_loss": 2.786125659942627, "z_loss": 10.503315925598145, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.42201486229896545, "tokens_per_second": 2838732.6523054647, "eta_seconds": 1585.017036509514, "data_wait_seconds": 1.7464160919189453e-05}