Banaxi-Tech commited on
Commit
8a67b8d
·
verified ·
1 Parent(s): 860a774

Save 25% checkpoint at 7,500,464,128 tokens

Browse files
checkpoint_metadata.json CHANGED
@@ -49,10 +49,10 @@
49
  "model_type": "bird_alpha",
50
  "output_attentions": false
51
  },
52
- "training_percent": 20,
53
- "step": 5723,
54
  "total_steps": 28611,
55
- "tokens_seen": 6001000448,
56
  "target_tokens": 30000000000,
57
  "trained_token_target": 30000807936,
58
  "tokens_per_step": 1048576,
@@ -68,5 +68,5 @@
68
  "dataset_id": "HuggingFaceFW/fineweb-edu",
69
  "dataset_config": "sample-100BT",
70
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
71
- "elapsed_seconds": 1795.5749864578247
72
  }
 
49
  "model_type": "bird_alpha",
50
  "output_attentions": false
51
  },
52
+ "training_percent": 25,
53
+ "step": 7153,
54
  "total_steps": 28611,
55
+ "tokens_seen": 7500464128,
56
  "target_tokens": 30000000000,
57
  "trained_token_target": 30000807936,
58
  "tokens_per_step": 1048576,
 
68
  "dataset_id": "HuggingFaceFW/fineweb-edu",
69
  "dataset_config": "sample-100BT",
70
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
71
+ "elapsed_seconds": 2224.6207723617554
72
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e63d34c2ca0bfe6feb963cbcd50446e9db7239635cf5e9e97608c4f69c4233ab
3
  size 13860292
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63f565d6032701e9011943e9cb321ca2198a43318b5747e5722e3985e03b2e5b
3
  size 13860292
training_metrics.jsonl CHANGED
@@ -571,3 +571,146 @@
571
  {"step": 5700, "total_steps": 28611, "tokens": 5976883200, "loss": 2.8146555423736572, "ce_loss": 2.8128944039344788, "z_loss": 17.6112961769104, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29659146070480347, "tokens_per_second": 2857028.4860553085, "eta_seconds": 8408.710257267952, "data_wait_seconds": 2.0623207092285156e-05}
572
  {"step": 5710, "total_steps": 28611, "tokens": 5987368960, "loss": 2.815598785877228, "ce_loss": 2.813680648803711, "z_loss": 19.181795597076416, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29509371519088745, "tokens_per_second": 4464471.598225356, "eta_seconds": 5378.786368703842, "data_wait_seconds": 1.5020370483398438e-05}
573
  {"step": 5720, "total_steps": 28611, "tokens": 5997854720, "loss": 2.8120274543762207, "ce_loss": 2.810309886932373, "z_loss": 17.176077365875244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2933826446533203, "tokens_per_second": 3011024.430394211, "eta_seconds": 7971.689958310127, "data_wait_seconds": 1.8417835235595703e-05}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
571
  {"step": 5700, "total_steps": 28611, "tokens": 5976883200, "loss": 2.8146555423736572, "ce_loss": 2.8128944039344788, "z_loss": 17.6112961769104, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29659146070480347, "tokens_per_second": 2857028.4860553085, "eta_seconds": 8408.710257267952, "data_wait_seconds": 2.0623207092285156e-05}
572
  {"step": 5710, "total_steps": 28611, "tokens": 5987368960, "loss": 2.815598785877228, "ce_loss": 2.813680648803711, "z_loss": 19.181795597076416, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29509371519088745, "tokens_per_second": 4464471.598225356, "eta_seconds": 5378.786368703842, "data_wait_seconds": 1.5020370483398438e-05}
573
  {"step": 5720, "total_steps": 28611, "tokens": 5997854720, "loss": 2.8120274543762207, "ce_loss": 2.810309886932373, "z_loss": 17.176077365875244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2933826446533203, "tokens_per_second": 3011024.430394211, "eta_seconds": 7971.689958310127, "data_wait_seconds": 1.8417835235595703e-05}
574
+ {"step": 5730, "total_steps": 28611, "tokens": 6008340480, "loss": 2.8192367553710938, "ce_loss": 2.8172525763511658, "z_loss": 19.842087745666504, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2992396950721741, "tokens_per_second": 9420453.78338877, "eta_seconds": 2546.8483798844472, "data_wait_seconds": 1.2218952178955078e-05}
575
+ {"step": 5740, "total_steps": 28611, "tokens": 6018826240, "loss": 2.8038220405578613, "ce_loss": 2.8018970489501953, "z_loss": 19.249648094177246, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2914619743824005, "tokens_per_second": 3039787.4183326685, "eta_seconds": 7889.361457109451, "data_wait_seconds": 1.5854835510253906e-05}
576
+ {"step": 5750, "total_steps": 28611, "tokens": 6029312000, "loss": 2.867118716239929, "ce_loss": 2.8655086755752563, "z_loss": 16.099973917007446, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2980847656726837, "tokens_per_second": 4649764.29084761, "eta_seconds": 5155.421745395661, "data_wait_seconds": 1.6868114471435547e-05}
577
+ {"step": 5760, "total_steps": 28611, "tokens": 6039797760, "loss": 2.8344058990478516, "ce_loss": 2.8324458599090576, "z_loss": 19.60045623779297, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2632685899734497, "tokens_per_second": 3138710.489707879, "eta_seconds": 7634.030043411255, "data_wait_seconds": 1.3470649719238281e-05}
578
+ {"step": 5770, "total_steps": 28611, "tokens": 6050283520, "loss": 2.834263503551483, "ce_loss": 2.8327330350875854, "z_loss": 15.305086612701416, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2713818848133087, "tokens_per_second": 2822906.8387076897, "eta_seconds": 8484.348150491714, "data_wait_seconds": 0.5733647346496582}
579
+ {"step": 5780, "total_steps": 28611, "tokens": 6060769280, "loss": 2.8680981397628784, "ce_loss": 2.8664366006851196, "z_loss": 16.616079568862915, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25046810507774353, "tokens_per_second": 4179593.773159176, "eta_seconds": 5727.838626265526, "data_wait_seconds": 1.329183578491211e-05}
580
+ {"step": 5790, "total_steps": 28611, "tokens": 6071255040, "loss": 2.835805356502533, "ce_loss": 2.834206283092499, "z_loss": 15.990609645843506, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29892244935035706, "tokens_per_second": 3131518.5512059457, "eta_seconds": 7641.517207932473, "data_wait_seconds": 1.2934207916259766e-05}
581
+ {"step": 5800, "total_steps": 28611, "tokens": 6081740800, "loss": 2.941807508468628, "ce_loss": 2.9402633905410767, "z_loss": 15.44070029258728, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2970503866672516, "tokens_per_second": 2822489.8018604876, "eta_seconds": 8474.456531333923, "data_wait_seconds": 1.3887882232666016e-05}
582
+ {"step": 5810, "total_steps": 28611, "tokens": 6092226560, "loss": 2.8379628658294678, "ce_loss": 2.8359365463256836, "z_loss": 20.263611316680908, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30414220690727234, "tokens_per_second": 2859671.593211011, "eta_seconds": 8360.603865408897, "data_wait_seconds": 0.294294536113739}
583
+ {"step": 5820, "total_steps": 28611, "tokens": 6102712320, "loss": 2.8347859978675842, "ce_loss": 2.832552969455719, "z_loss": 22.330461502075195, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.294763445854187, "tokens_per_second": 4342136.295732885, "eta_seconds": 5503.764504003524, "data_wait_seconds": 1.8596649169921875e-05}
584
+ {"step": 5830, "total_steps": 28611, "tokens": 6113198080, "loss": 2.8254036903381348, "ce_loss": 2.8231107592582703, "z_loss": 22.929068565368652, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2494674026966095, "tokens_per_second": 4321651.383109361, "eta_seconds": 5527.426379036903, "data_wait_seconds": 1.3053417205810547e-05}
585
+ {"step": 5840, "total_steps": 28611, "tokens": 6123683840, "loss": 2.865261733531952, "ce_loss": 2.863686740398407, "z_loss": 15.749350786209106, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25071701407432556, "tokens_per_second": 2941092.5134365647, "eta_seconds": 8118.453937411308, "data_wait_seconds": 1.3649463653564453e-05}
586
+ {"step": 5850, "total_steps": 28611, "tokens": 6134169600, "loss": 2.8401493430137634, "ce_loss": 2.8385164737701416, "z_loss": 16.32957935333252, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32901930809020996, "tokens_per_second": 2780501.9904301926, "eta_seconds": 8583.571750044823, "data_wait_seconds": 0.5552079081535339}
587
+ {"step": 5860, "total_steps": 28611, "tokens": 6144655360, "loss": 2.8605454564094543, "ce_loss": 2.85869437456131, "z_loss": 18.51100492477417, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27359622716903687, "tokens_per_second": 4630937.760353506, "eta_seconds": 5151.473375487328, "data_wait_seconds": 1.531839370727539e-05}
588
+ {"step": 5870, "total_steps": 28611, "tokens": 6155141120, "loss": 2.8323974609375, "ce_loss": 2.8306667804718018, "z_loss": 17.307067394256592, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2550800144672394, "tokens_per_second": 3065940.6671494413, "eta_seconds": 7777.602179813385, "data_wait_seconds": 0.28158092498779297}
589
+ {"step": 5880, "total_steps": 28611, "tokens": 6165626880, "loss": 2.863917350769043, "ce_loss": 2.861719071865082, "z_loss": 21.983006477355957, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3219502866268158, "tokens_per_second": 4271526.87513142, "eta_seconds": 5580.0143023252485, "data_wait_seconds": 1.3053417205810547e-05}
590
+ {"step": 5890, "total_steps": 28611, "tokens": 6176112640, "loss": 2.8364126086235046, "ce_loss": 2.8347395062446594, "z_loss": 16.73126459121704, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3294468820095062, "tokens_per_second": 3038019.402987838, "eta_seconds": 7842.1800968647, "data_wait_seconds": 1.424551010131836e-05}
591
+ {"step": 5900, "total_steps": 28611, "tokens": 6186598400, "loss": 2.860374093055725, "ce_loss": 2.858469545841217, "z_loss": 19.046013832092285, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28511932492256165, "tokens_per_second": 3083141.0966612697, "eta_seconds": 7724.008986091614, "data_wait_seconds": 0.30848056077957153}
592
+ {"step": 5910, "total_steps": 28611, "tokens": 6197084160, "loss": 2.8542943000793457, "ce_loss": 2.8526735305786133, "z_loss": 16.20806121826172, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.288903146982193, "tokens_per_second": 4811209.618186749, "eta_seconds": 4947.554911351204, "data_wait_seconds": 1.4185905456542969e-05}
593
+ {"step": 5920, "total_steps": 28611, "tokens": 6207569920, "loss": 2.830465257167816, "ce_loss": 2.8286882042884827, "z_loss": 17.77110528945923, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2576541006565094, "tokens_per_second": 2852711.7378921406, "eta_seconds": 8340.568624567986, "data_wait_seconds": 1.8596649169921875e-05}
594
+ {"step": 5930, "total_steps": 28611, "tokens": 6218055680, "loss": 2.834673047065735, "ce_loss": 2.832833707332611, "z_loss": 18.393677711486816, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3161380887031555, "tokens_per_second": 2777337.7390764887, "eta_seconds": 8563.147334003448, "data_wait_seconds": 1.5735626220703125e-05}
595
+ {"step": 5940, "total_steps": 28611, "tokens": 6228541440, "loss": 2.8962756395339966, "ce_loss": 2.894761621952057, "z_loss": 15.140166282653809, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.302314430475235, "tokens_per_second": 4165285.366482517, "eta_seconds": 5707.23597650528, "data_wait_seconds": 1.5974044799804688e-05}
596
+ {"step": 5950, "total_steps": 28611, "tokens": 6239027200, "loss": 2.846462607383728, "ce_loss": 2.84472793340683, "z_loss": 17.346679210662842, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2518426179885864, "tokens_per_second": 4192213.8421128686, "eta_seconds": 5668.074585628509, "data_wait_seconds": 1.7583370208740234e-05}
597
+ {"step": 5960, "total_steps": 28611, "tokens": 6249512960, "loss": 2.8305144906044006, "ce_loss": 2.8290008902549744, "z_loss": 15.136045217514038, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2900860011577606, "tokens_per_second": 2938481.5621278496, "eta_seconds": 8082.846352386475, "data_wait_seconds": 1.436471939086914e-05}
598
+ {"step": 5970, "total_steps": 28611, "tokens": 6259998720, "loss": 2.807462513446808, "ce_loss": 2.805699408054352, "z_loss": 17.630911350250244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2628151476383209, "tokens_per_second": 2831084.71825469, "eta_seconds": 8385.764319562912, "data_wait_seconds": 2.47955322265625e-05}
599
+ {"step": 5980, "total_steps": 28611, "tokens": 6270484480, "loss": 2.8641071915626526, "ce_loss": 2.861889362335205, "z_loss": 22.178240299224854, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28001758456230164, "tokens_per_second": 4177772.6144524706, "eta_seconds": 5680.137634563446, "data_wait_seconds": 1.4781951904296875e-05}
600
+ {"step": 5990, "total_steps": 28611, "tokens": 6280970240, "loss": 2.840282440185547, "ce_loss": 2.8383830189704895, "z_loss": 18.99453353881836, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2830638885498047, "tokens_per_second": 2793785.3804141963, "eta_seconds": 8490.214696621895, "data_wait_seconds": 0.2912030816078186}
601
+ {"step": 6000, "total_steps": 28611, "tokens": 6291456000, "loss": 2.872944414615631, "ce_loss": 2.870855450630188, "z_loss": 20.8891019821167, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2945840060710907, "tokens_per_second": 4339654.720971441, "eta_seconds": 5463.418972349167, "data_wait_seconds": 1.4662742614746094e-05}
602
+ {"step": 6010, "total_steps": 28611, "tokens": 6301941760, "loss": 2.855299234390259, "ce_loss": 2.8536999225616455, "z_loss": 15.992786169052124, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3430062532424927, "tokens_per_second": 4134332.723690839, "eta_seconds": 5732.2106757879255, "data_wait_seconds": 1.6748905181884766e-05}
603
+ {"step": 6020, "total_steps": 28611, "tokens": 6312427520, "loss": 2.8321973085403442, "ce_loss": 2.830668032169342, "z_loss": 15.292386531829834, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2788487374782562, "tokens_per_second": 2894797.923523776, "eta_seconds": 8183.086019062996, "data_wait_seconds": 1.3947486877441406e-05}
604
+ {"step": 6030, "total_steps": 28611, "tokens": 6322913280, "loss": 2.802766978740692, "ce_loss": 2.8009210228919983, "z_loss": 18.459800720214844, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2596016228199005, "tokens_per_second": 2876178.4971661964, "eta_seconds": 8232.414879441261, "data_wait_seconds": 1.4781951904296875e-05}
605
+ {"step": 6040, "total_steps": 28611, "tokens": 6333399040, "loss": 2.824723720550537, "ce_loss": 2.8232834935188293, "z_loss": 14.402743577957153, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29942747950553894, "tokens_per_second": 2724976.7382534486, "eta_seconds": 8685.361810159682, "data_wait_seconds": 1.919269561767578e-05}
606
+ {"step": 6050, "total_steps": 28611, "tokens": 6343884800, "loss": 2.8453420400619507, "ce_loss": 2.8432950377464294, "z_loss": 20.469443798065186, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27730223536491394, "tokens_per_second": 4286073.275391719, "eta_seconds": 5519.486396050453, "data_wait_seconds": 1.7642974853515625e-05}
607
+ {"step": 6060, "total_steps": 28611, "tokens": 6354370560, "loss": 2.8456119894981384, "ce_loss": 2.8437539935112, "z_loss": 18.579882621765137, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2555290460586548, "tokens_per_second": 2802340.733314732, "eta_seconds": 8438.102153277398, "data_wait_seconds": 1.4960765838623047e-05}
608
+ {"step": 6070, "total_steps": 28611, "tokens": 6364856320, "loss": 2.868148386478424, "ce_loss": 2.866561472415924, "z_loss": 15.86960768699646, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24034327268600464, "tokens_per_second": 4151734.483532241, "eta_seconds": 5693.030638098717, "data_wait_seconds": 1.7523765563964844e-05}
609
+ {"step": 6080, "total_steps": 28611, "tokens": 6375342080, "loss": 2.8941214084625244, "ce_loss": 2.8920058012008667, "z_loss": 21.155527114868164, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3541170656681061, "tokens_per_second": 2916428.6971586077, "eta_seconds": 8100.820664334296, "data_wait_seconds": 1.621246337890625e-05}
610
+ {"step": 6090, "total_steps": 28611, "tokens": 6385827840, "loss": 2.781602621078491, "ce_loss": 2.779686748981476, "z_loss": 19.158141613006592, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30648961663246155, "tokens_per_second": 4543176.4660410965, "eta_seconds": 5197.900691843033, "data_wait_seconds": 1.5497207641601562e-05}
611
+ {"step": 6100, "total_steps": 28611, "tokens": 6396313600, "loss": 2.8187502026557922, "ce_loss": 2.8169321417808533, "z_loss": 18.181160926818848, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2590930163860321, "tokens_per_second": 2935079.380847615, "eta_seconds": 8042.199638628959, "data_wait_seconds": 1.621246337890625e-05}
612
+ {"step": 6110, "total_steps": 28611, "tokens": 6406799360, "loss": 2.8465240597724915, "ce_loss": 2.845044732093811, "z_loss": 14.793220281600952, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2698443531990051, "tokens_per_second": 4376858.576421401, "eta_seconds": 5390.626213765145, "data_wait_seconds": 1.621246337890625e-05}
613
+ {"step": 6120, "total_steps": 28611, "tokens": 6417285120, "loss": 2.8026572465896606, "ce_loss": 2.800860583782196, "z_loss": 17.966111183166504, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2976178228855133, "tokens_per_second": 2912758.4498434328, "eta_seconds": 8096.628409838676, "data_wait_seconds": 1.6808509826660156e-05}
614
+ {"step": 6130, "total_steps": 28611, "tokens": 6427770880, "loss": 2.8684027194976807, "ce_loss": 2.866555154323578, "z_loss": 18.476004600524902, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2881656587123871, "tokens_per_second": 4437438.540515867, "eta_seconds": 5312.307278347016, "data_wait_seconds": 1.627206802368164e-05}
615
+ {"step": 6140, "total_steps": 28611, "tokens": 6438256640, "loss": 2.8722853660583496, "ce_loss": 2.8703060150146484, "z_loss": 19.793614864349365, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2810173034667969, "tokens_per_second": 3058726.518526555, "eta_seconds": 7703.386083483696, "data_wait_seconds": 1.5556812286376953e-05}
616
+ {"step": 6150, "total_steps": 28611, "tokens": 6448742400, "loss": 2.8386171460151672, "ce_loss": 2.8368635177612305, "z_loss": 17.535988807678223, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31035614013671875, "tokens_per_second": 4329792.9539163485, "eta_seconds": 5439.53620569706, "data_wait_seconds": 1.4543533325195312e-05}
617
+ {"step": 6160, "total_steps": 28611, "tokens": 6459228160, "loss": 2.8606244325637817, "ce_loss": 2.859045088291168, "z_loss": 15.793697834014893, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2742542624473572, "tokens_per_second": 3012431.8082201257, "eta_seconds": 7814.809188961983, "data_wait_seconds": 1.5795230865478516e-05}
618
+ {"step": 6170, "total_steps": 28611, "tokens": 6469713920, "loss": 2.858353078365326, "ce_loss": 2.8567066192626953, "z_loss": 16.464667081832886, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30020013451576233, "tokens_per_second": 3202773.8663873384, "eta_seconds": 7347.098171043396, "data_wait_seconds": 0.2539411783218384}
619
+ {"step": 6180, "total_steps": 28611, "tokens": 6480199680, "loss": 2.8436787724494934, "ce_loss": 2.8416380882263184, "z_loss": 20.406986236572266, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27579835057258606, "tokens_per_second": 4070352.975929642, "eta_seconds": 5778.518078184127, "data_wait_seconds": 1.3649463653564453e-05}
620
+ {"step": 6190, "total_steps": 28611, "tokens": 6490685440, "loss": 2.8619362711906433, "ce_loss": 2.8598062992095947, "z_loss": 21.29956865310669, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3136455714702606, "tokens_per_second": 2768921.7638688744, "eta_seconds": 8490.713895487785, "data_wait_seconds": 1.800060272216797e-05}
621
+ {"step": 6200, "total_steps": 28611, "tokens": 6501171200, "loss": 2.84975266456604, "ce_loss": 2.847584843635559, "z_loss": 21.67775869369507, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26562976837158203, "tokens_per_second": 2822476.216720498, "eta_seconds": 8325.893623757363, "data_wait_seconds": 1.7642974853515625e-05}
622
+ {"step": 6210, "total_steps": 28611, "tokens": 6511656960, "loss": 2.841853618621826, "ce_loss": 2.8398106694221497, "z_loss": 20.42940616607666, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3234143555164337, "tokens_per_second": 4150573.5442182403, "eta_seconds": 5659.254251432419, "data_wait_seconds": 1.8358230590820312e-05}
623
+ {"step": 6220, "total_steps": 28611, "tokens": 6522142720, "loss": 2.8545108437538147, "ce_loss": 2.852638006210327, "z_loss": 18.72850751876831, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30644071102142334, "tokens_per_second": 4203621.854190536, "eta_seconds": 5585.341886210441, "data_wait_seconds": 1.2516975402832031e-05}
624
+ {"step": 6230, "total_steps": 28611, "tokens": 6532628480, "loss": 2.8650148510932922, "ce_loss": 2.863184630870819, "z_loss": 18.3023362159729, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2881375849246979, "tokens_per_second": 2886435.145975746, "eta_seconds": 8130.506409859658, "data_wait_seconds": 1.5676021575927734e-05}
625
+ {"step": 6240, "total_steps": 28611, "tokens": 6543114240, "loss": 2.845982551574707, "ce_loss": 2.844110429286957, "z_loss": 18.72065496444702, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27130451798439026, "tokens_per_second": 4125254.177975353, "eta_seconds": 5686.363235807418, "data_wait_seconds": 2.1636486053466797e-05}
626
+ {"step": 6250, "total_steps": 28611, "tokens": 6553600000, "loss": 2.8590697646141052, "ce_loss": 2.8574990034103394, "z_loss": 15.707717657089233, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2559467554092407, "tokens_per_second": 2927927.614448604, "eta_seconds": 8008.124183225632, "data_wait_seconds": 1.6450881958007812e-05}
627
+ {"step": 6260, "total_steps": 28611, "tokens": 6564085760, "loss": 2.8623331785202026, "ce_loss": 2.8605039715766907, "z_loss": 18.29211950302124, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28594866394996643, "tokens_per_second": 2890516.446689825, "eta_seconds": 8108.143512845039, "data_wait_seconds": 2.0444393157958984e-05}
628
+ {"step": 6270, "total_steps": 28611, "tokens": 6574571520, "loss": 2.880484640598297, "ce_loss": 2.878678798675537, "z_loss": 18.058632850646973, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3051489591598511, "tokens_per_second": 4457842.675399, "eta_seconds": 5255.061275553703, "data_wait_seconds": 1.7642974853515625e-05}
629
+ {"step": 6280, "total_steps": 28611, "tokens": 6585057280, "loss": 2.8701597452163696, "ce_loss": 2.8687350749969482, "z_loss": 14.246171474456787, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25346601009368896, "tokens_per_second": 2890117.3692666143, "eta_seconds": 8102.006826782227, "data_wait_seconds": 1.9311904907226562e-05}
630
+ {"step": 6290, "total_steps": 28611, "tokens": 6595543040, "loss": 2.893525183200836, "ce_loss": 2.8918155431747437, "z_loss": 17.09632110595703, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2798636257648468, "tokens_per_second": 4366151.77240657, "eta_seconds": 5360.616422891617, "data_wait_seconds": 1.4483928680419922e-05}
631
+ {"step": 6300, "total_steps": 28611, "tokens": 6606028800, "loss": 2.8805745244026184, "ce_loss": 2.8785901069641113, "z_loss": 19.84480094909668, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2818256616592407, "tokens_per_second": 4773630.508520862, "eta_seconds": 4900.835767292976, "data_wait_seconds": 1.9788742065429688e-05}
632
+ {"step": 6310, "total_steps": 28611, "tokens": 6616514560, "loss": 2.824944317340851, "ce_loss": 2.822709321975708, "z_loss": 22.35071611404419, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3019845187664032, "tokens_per_second": 3036991.037383183, "eta_seconds": 7699.822978782653, "data_wait_seconds": 1.3887882232666016e-05}
633
+ {"step": 6320, "total_steps": 28611, "tokens": 6627000320, "loss": 2.878734767436981, "ce_loss": 2.8769540786743164, "z_loss": 17.806623935699463, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30063357949256897, "tokens_per_second": 4292292.577353897, "eta_seconds": 5445.529910826684, "data_wait_seconds": 1.4066696166992188e-05}
634
+ {"step": 6330, "total_steps": 28611, "tokens": 6637486080, "loss": 2.866278290748596, "ce_loss": 2.864203691482544, "z_loss": 20.745704650878906, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31197869777679443, "tokens_per_second": 4505833.251820755, "eta_seconds": 5185.127933120728, "data_wait_seconds": 1.436471939086914e-05}
635
+ {"step": 6340, "total_steps": 28611, "tokens": 6647971840, "loss": 2.850071430206299, "ce_loss": 2.8483142852783203, "z_loss": 17.57119655609131, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27762535214424133, "tokens_per_second": 4434163.213897517, "eta_seconds": 5266.571158862113, "data_wait_seconds": 0.28777068853378296}
636
+ {"step": 6350, "total_steps": 28611, "tokens": 6658457600, "loss": 2.8543149828910828, "ce_loss": 2.852648437023163, "z_loss": 16.665247678756714, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2565165162086487, "tokens_per_second": 9813598.2971735, "eta_seconds": 2378.572021102905, "data_wait_seconds": 1.2636184692382812e-05}
637
+ {"step": 6360, "total_steps": 28611, "tokens": 6668943360, "loss": 2.8281288743019104, "ce_loss": 2.8259559273719788, "z_loss": 21.72959613800049, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2714237868785858, "tokens_per_second": 4051930.97261622, "eta_seconds": 5758.208798146247, "data_wait_seconds": 1.3470649719238281e-05}
638
+ {"step": 6370, "total_steps": 28611, "tokens": 6679429120, "loss": 2.8601144552230835, "ce_loss": 2.8584324717521667, "z_loss": 16.819871425628662, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2756486237049103, "tokens_per_second": 2764551.1794878757, "eta_seconds": 8435.86437792778, "data_wait_seconds": 0.30339908599853516}
639
+ {"step": 6380, "total_steps": 28611, "tokens": 6689914880, "loss": 2.8856403827667236, "ce_loss": 2.883824348449707, "z_loss": 18.160310745239258, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30156293511390686, "tokens_per_second": 3041423.7191640725, "eta_seconds": 7664.46743645668, "data_wait_seconds": 0.18364781141281128}
640
+ {"step": 6390, "total_steps": 28611, "tokens": 6700400640, "loss": 2.883876919746399, "ce_loss": 2.8821625113487244, "z_loss": 17.14425253868103, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3431142270565033, "tokens_per_second": 3140968.6615658486, "eta_seconds": 7418.2234229564665, "data_wait_seconds": 0.24225670099258423}
641
+ {"step": 6400, "total_steps": 28611, "tokens": 6710886400, "loss": 2.8644763231277466, "ce_loss": 2.8626564741134644, "z_loss": 18.198955535888672, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2837863564491272, "tokens_per_second": 3656472.2541738464, "eta_seconds": 6369.505883550643, "data_wait_seconds": 1.5020370483398438e-05}
642
+ {"step": 6410, "total_steps": 28611, "tokens": 6721372160, "loss": 2.859845995903015, "ce_loss": 2.8579035997390747, "z_loss": 19.423771381378174, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2723844647407532, "tokens_per_second": 4181036.103695824, "eta_seconds": 5567.862892985344, "data_wait_seconds": 1.615285873413086e-05}
643
+ {"step": 6420, "total_steps": 28611, "tokens": 6731857920, "loss": 2.8728801608085632, "ce_loss": 2.8715348839759827, "z_loss": 13.453296184539795, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29101407527923584, "tokens_per_second": 2860574.058135382, "eta_seconds": 8134.36378262043, "data_wait_seconds": 1.6868114471435547e-05}
644
+ {"step": 6430, "total_steps": 28611, "tokens": 6742343680, "loss": 2.8488303422927856, "ce_loss": 2.846981644630432, "z_loss": 18.486732006072998, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.317402720451355, "tokens_per_second": 4335850.470445645, "eta_seconds": 5364.221947813035, "data_wait_seconds": 1.52587890625e-05}
645
+ {"step": 6440, "total_steps": 28611, "tokens": 6752829440, "loss": 2.8569400906562805, "ce_loss": 2.855209708213806, "z_loss": 17.3036847114563, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3013807237148285, "tokens_per_second": 2818019.752138626, "eta_seconds": 8249.757113432883, "data_wait_seconds": 1.5437602996826172e-05}
646
+ {"step": 6450, "total_steps": 28611, "tokens": 6763315200, "loss": 2.859123170375824, "ce_loss": 2.8573246598243713, "z_loss": 17.98517608642578, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27207109332084656, "tokens_per_second": 4366444.369976296, "eta_seconds": 5321.83414399624, "data_wait_seconds": 1.3589859008789062e-05}
647
+ {"step": 6460, "total_steps": 28611, "tokens": 6773800960, "loss": 2.8480125069618225, "ce_loss": 2.8460773229599, "z_loss": 19.352033615112305, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30453744530677795, "tokens_per_second": 2868730.929454426, "eta_seconds": 8096.613989663124, "data_wait_seconds": 1.424551010131836e-05}
648
+ {"step": 6470, "total_steps": 28611, "tokens": 6784286720, "loss": 2.874234676361084, "ce_loss": 2.8725122213363647, "z_loss": 17.224722862243652, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29854223132133484, "tokens_per_second": 2930876.4374780627, "eta_seconds": 7921.357897973061, "data_wait_seconds": 0.29107820987701416}
649
+ {"step": 6480, "total_steps": 28611, "tokens": 6794772480, "loss": 2.88510525226593, "ce_loss": 2.8832152485847473, "z_loss": 18.899527072906494, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33666086196899414, "tokens_per_second": 4627741.206468205, "eta_seconds": 5014.549090075493, "data_wait_seconds": 1.6748905181884766e-05}
650
+ {"step": 6490, "total_steps": 28611, "tokens": 6805258240, "loss": 2.8792752623558044, "ce_loss": 2.877772867679596, "z_loss": 15.023543357849121, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2497827708721161, "tokens_per_second": 4660614.337398629, "eta_seconds": 4976.929652786255, "data_wait_seconds": 1.245737075805664e-05}
651
+ {"step": 6500, "total_steps": 28611, "tokens": 6815744000, "loss": 2.86605304479599, "ce_loss": 2.864539861679077, "z_loss": 15.132322072982788, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3232664465904236, "tokens_per_second": 2959914.7859801664, "eta_seconds": 7833.0173712492, "data_wait_seconds": 1.3530254364013672e-05}
652
+ {"step": 6510, "total_steps": 28611, "tokens": 6826229760, "loss": 2.86559134721756, "ce_loss": 2.863804340362549, "z_loss": 17.870097160339355, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2791512608528137, "tokens_per_second": 4303019.068301536, "eta_seconds": 5385.655468440055, "data_wait_seconds": 1.6748905181884766e-05}
653
+ {"step": 6520, "total_steps": 28611, "tokens": 6836715520, "loss": 2.863633453845978, "ce_loss": 2.8619801998138428, "z_loss": 16.53262186050415, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26991426944732666, "tokens_per_second": 4162366.2575773336, "eta_seconds": 5565.125936198235, "data_wait_seconds": 1.817941665649414e-05}
654
+ {"step": 6530, "total_steps": 28611, "tokens": 6847201280, "loss": 2.880589544773102, "ce_loss": 2.8782657384872437, "z_loss": 23.237850189208984, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27499350905418396, "tokens_per_second": 4537094.857192375, "eta_seconds": 5103.178880929947, "data_wait_seconds": 0.27460962533950806}
655
+ {"step": 6540, "total_steps": 28611, "tokens": 6857687040, "loss": 2.84041166305542, "ce_loss": 2.8386918902397156, "z_loss": 17.197572231292725, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2868141829967499, "tokens_per_second": 4711841.994800101, "eta_seconds": 4911.692905139924, "data_wait_seconds": 1.5795230865478516e-05}
656
+ {"step": 6550, "total_steps": 28611, "tokens": 6868172800, "loss": 2.8419299125671387, "ce_loss": 2.8402501940727234, "z_loss": 16.796772956848145, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28880444169044495, "tokens_per_second": 2890116.2297459077, "eta_seconds": 8004.050113248825, "data_wait_seconds": 1.4722347259521484e-05}
657
+ {"step": 6560, "total_steps": 28611, "tokens": 6878658560, "loss": 2.880751311779022, "ce_loss": 2.8790008425712585, "z_loss": 17.504765033721924, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2762497365474701, "tokens_per_second": 4371896.449679885, "eta_seconds": 5288.814509248733, "data_wait_seconds": 1.6033649444580078e-05}
658
+ {"step": 6570, "total_steps": 28611, "tokens": 6889144320, "loss": 2.8992104530334473, "ce_loss": 2.8976680636405945, "z_loss": 15.424383163452148, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32041507959365845, "tokens_per_second": 2936205.1165964203, "eta_seconds": 7871.270125293731, "data_wait_seconds": 1.4126300811767578e-05}
659
+ {"step": 6580, "total_steps": 28611, "tokens": 6899630080, "loss": 2.8304975628852844, "ce_loss": 2.828680455684662, "z_loss": 18.171209812164307, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2845434844493866, "tokens_per_second": 2984077.2941829427, "eta_seconds": 7741.481060504913, "data_wait_seconds": 0.5102185606956482}
660
+ {"step": 6590, "total_steps": 28611, "tokens": 6910115840, "loss": 2.8517594933509827, "ce_loss": 2.849869191646576, "z_loss": 18.9029221534729, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26716509461402893, "tokens_per_second": 4167557.2564989706, "eta_seconds": 5540.581850433349, "data_wait_seconds": 1.3470649719238281e-05}
661
+ {"step": 6600, "total_steps": 28611, "tokens": 6920601600, "loss": 2.8025189638137817, "ce_loss": 2.8011253476142883, "z_loss": 13.93643045425415, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23903363943099976, "tokens_per_second": 3022492.1834587995, "eta_seconds": 7636.150876522064, "data_wait_seconds": 0.30850666761398315}
662
+ {"step": 6610, "total_steps": 28611, "tokens": 6931087360, "loss": 2.8383166790008545, "ce_loss": 2.836561143398285, "z_loss": 17.555927276611328, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29575347900390625, "tokens_per_second": 4355176.764143826, "eta_seconds": 5297.0801933765415, "data_wait_seconds": 1.7940998077392578e-05}
663
+ {"step": 6620, "total_steps": 28611, "tokens": 6941573120, "loss": 2.8842564821243286, "ce_loss": 2.882566213607788, "z_loss": 16.90271258354187, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30975160002708435, "tokens_per_second": 3079489.9185610265, "eta_seconds": 7488.004645514488, "data_wait_seconds": 1.7344951629638672e-05}
664
+ {"step": 6630, "total_steps": 28611, "tokens": 6952058880, "loss": 2.913656771183014, "ce_loss": 2.911603033542633, "z_loss": 20.53724431991577, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27378955483436584, "tokens_per_second": 4407347.336187556, "eta_seconds": 5229.619382786751, "data_wait_seconds": 1.6510486602783203e-05}
665
+ {"step": 6640, "total_steps": 28611, "tokens": 6962544640, "loss": 2.8648797273635864, "ce_loss": 2.863372325897217, "z_loss": 15.074422836303711, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28037431836128235, "tokens_per_second": 4682015.428881066, "eta_seconds": 4920.586795568466, "data_wait_seconds": 0.24967330694198608}
666
+ {"step": 6650, "total_steps": 28611, "tokens": 6973030400, "loss": 2.853874146938324, "ce_loss": 2.8518906235694885, "z_loss": 19.835282802581787, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3369040787220001, "tokens_per_second": 7547243.150801693, "eta_seconds": 3051.150874018669, "data_wait_seconds": 1.6689300537109375e-05}
667
+ {"step": 6660, "total_steps": 28611, "tokens": 6983516160, "loss": 2.8336102962493896, "ce_loss": 2.8318580985069275, "z_loss": 17.52221918106079, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29770559072494507, "tokens_per_second": 4274939.37203185, "eta_seconds": 5384.238178110123, "data_wait_seconds": 1.4126300811767578e-05}
668
+ {"step": 6670, "total_steps": 28611, "tokens": 6994001920, "loss": 2.8383217453956604, "ce_loss": 2.836701989173889, "z_loss": 16.196824312210083, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2916512191295624, "tokens_per_second": 3943855.542622333, "eta_seconds": 5833.582332658768, "data_wait_seconds": 1.5914440155029297e-05}
669
+ {"step": 6680, "total_steps": 28611, "tokens": 7004487680, "loss": 2.8327826857566833, "ce_loss": 2.8308410048484802, "z_loss": 19.41679811477661, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2622925937175751, "tokens_per_second": 4328982.359439734, "eta_seconds": 5312.176938271523, "data_wait_seconds": 1.4066696166992188e-05}
670
+ {"step": 6690, "total_steps": 28611, "tokens": 7014973440, "loss": 2.88395231962204, "ce_loss": 2.8819385170936584, "z_loss": 20.138166904449463, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33273065090179443, "tokens_per_second": 3027009.483981556, "eta_seconds": 7593.578618645668, "data_wait_seconds": 0.2963806986808777}
671
+ {"step": 6700, "total_steps": 28611, "tokens": 7025459200, "loss": 2.8455746173858643, "ce_loss": 2.8438432812690735, "z_loss": 17.313640117645264, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2548736035823822, "tokens_per_second": 4293510.269809806, "eta_seconds": 5351.180570721626, "data_wait_seconds": 1.2934207916259766e-05}
672
+ {"step": 6710, "total_steps": 28611, "tokens": 7035944960, "loss": 2.842646062374115, "ce_loss": 2.840782940387726, "z_loss": 18.631669998168945, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30880439281463623, "tokens_per_second": 3012579.3455579653, "eta_seconds": 7622.990249156952, "data_wait_seconds": 1.436471939086914e-05}
673
+ {"step": 6720, "total_steps": 28611, "tokens": 7046430720, "loss": 2.884640336036682, "ce_loss": 2.8830212354660034, "z_loss": 16.191219091415405, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29471835494041443, "tokens_per_second": 4454899.041694295, "eta_seconds": 5152.614459085464, "data_wait_seconds": 1.2755393981933594e-05}
674
+ {"step": 6730, "total_steps": 28611, "tokens": 7056916480, "loss": 2.869576334953308, "ce_loss": 2.867846429347992, "z_loss": 17.298936367034912, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2911228537559509, "tokens_per_second": 2921710.1636399473, "eta_seconds": 7852.8978478193285, "data_wait_seconds": 1.424551010131836e-05}
675
+ {"step": 6740, "total_steps": 28611, "tokens": 7067402240, "loss": 2.840177834033966, "ce_loss": 2.8383558988571167, "z_loss": 18.21904754638672, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3237078785896301, "tokens_per_second": 3158272.375135022, "eta_seconds": 7261.376781988144, "data_wait_seconds": 0.4793297052383423}
676
+ {"step": 6750, "total_steps": 28611, "tokens": 7077888000, "loss": 2.85284960269928, "ce_loss": 2.850959539413452, "z_loss": 18.901082038879395, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30999869108200073, "tokens_per_second": 3857781.5631955042, "eta_seconds": 5941.995304942131, "data_wait_seconds": 1.6868114471435547e-05}
677
+ {"step": 6760, "total_steps": 28611, "tokens": 7088373760, "loss": 2.8323051929473877, "ce_loss": 2.8304229974746704, "z_loss": 18.822336196899414, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30291274189949036, "tokens_per_second": 4736983.864702487, "eta_seconds": 4836.924682545661, "data_wait_seconds": 1.52587890625e-05}
678
+ {"step": 6770, "total_steps": 28611, "tokens": 7098859520, "loss": 2.8513898253440857, "ce_loss": 2.849698781967163, "z_loss": 16.910069465637207, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3099808692932129, "tokens_per_second": 2960330.5836501285, "eta_seconds": 7736.280718946457, "data_wait_seconds": 1.4960765838623047e-05}
679
+ {"step": 6780, "total_steps": 28611, "tokens": 7109345280, "loss": 2.842540919780731, "ce_loss": 2.840916156768799, "z_loss": 16.247813940048218, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2521904706954956, "tokens_per_second": 4313638.949593926, "eta_seconds": 5306.763714694976, "data_wait_seconds": 1.6689300537109375e-05}
680
+ {"step": 6790, "total_steps": 28611, "tokens": 7119831040, "loss": 2.8408830761909485, "ce_loss": 2.8393941521644592, "z_loss": 14.889392852783203, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27821946144104004, "tokens_per_second": 2938533.9831072832, "eta_seconds": 7786.527917504311, "data_wait_seconds": 1.3589859008789062e-05}
681
+ {"step": 6800, "total_steps": 28611, "tokens": 7130316800, "loss": 2.8455729484558105, "ce_loss": 2.8435815572738647, "z_loss": 19.91386842727661, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2851604223251343, "tokens_per_second": 2768883.9356970964, "eta_seconds": 8259.822970962525, "data_wait_seconds": 1.8537044525146484e-05}
682
+ {"step": 6810, "total_steps": 28611, "tokens": 7140802560, "loss": 2.868474304676056, "ce_loss": 2.866682708263397, "z_loss": 17.915527820587158, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2947057783603668, "tokens_per_second": 4684145.20004699, "eta_seconds": 4880.293927645684, "data_wait_seconds": 1.633167266845703e-05}
683
+ {"step": 6820, "total_steps": 28611, "tokens": 7151288320, "loss": 2.8430622816085815, "ce_loss": 2.8413763642311096, "z_loss": 16.85901927947998, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2643522322177887, "tokens_per_second": 3041140.6460149726, "eta_seconds": 7513.470199394225, "data_wait_seconds": 0.25697821378707886}
684
+ {"step": 6830, "total_steps": 28611, "tokens": 7161774080, "loss": 2.8579647541046143, "ce_loss": 2.856239378452301, "z_loss": 17.254097938537598, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2647244334220886, "tokens_per_second": 4271304.934056722, "eta_seconds": 5347.085775566101, "data_wait_seconds": 1.2993812561035156e-05}
685
+ {"step": 6840, "total_steps": 28611, "tokens": 7172259840, "loss": 2.844758152961731, "ce_loss": 2.842888057231903, "z_loss": 18.70147705078125, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32529571652412415, "tokens_per_second": 2990615.5714524486, "eta_seconds": 7633.394380044937, "data_wait_seconds": 1.4901161193847656e-05}
686
+ {"step": 6850, "total_steps": 28611, "tokens": 7182745600, "loss": 2.8966110944747925, "ce_loss": 2.8950164914131165, "z_loss": 15.945627450942993, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27306005358695984, "tokens_per_second": 4227786.773284976, "eta_seconds": 5397.16489019394, "data_wait_seconds": 1.430511474609375e-05}
687
+ {"step": 6860, "total_steps": 28611, "tokens": 7193231360, "loss": 2.8264166116714478, "ce_loss": 2.8246849179267883, "z_loss": 17.31718921661377, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32860109210014343, "tokens_per_second": 2283247.696583575, "eta_seconds": 9989.094310760498, "data_wait_seconds": 0.30073755979537964}
688
+ {"step": 6870, "total_steps": 28611, "tokens": 7203717120, "loss": 2.8039490580558777, "ce_loss": 2.8022422790527344, "z_loss": 17.067567825317383, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3012674152851105, "tokens_per_second": 2926326.426666523, "eta_seconds": 7790.344442868233, "data_wait_seconds": 1.7762184143066406e-05}
689
+ {"step": 6880, "total_steps": 28611, "tokens": 7214202880, "loss": 2.8609618544578552, "ce_loss": 2.859303891658783, "z_loss": 16.579545974731445, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26956692337989807, "tokens_per_second": 4155241.223094318, "eta_seconds": 5483.822438359261, "data_wait_seconds": 1.7344951629638672e-05}
690
+ {"step": 6890, "total_steps": 28611, "tokens": 7224688640, "loss": 2.824951648712158, "ce_loss": 2.8227455615997314, "z_loss": 22.06086826324463, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3055373430252075, "tokens_per_second": 2785126.537386572, "eta_seconds": 8177.768223547936, "data_wait_seconds": 2.008676528930664e-05}
691
+ {"step": 6900, "total_steps": 28611, "tokens": 7235174400, "loss": 2.8447118401527405, "ce_loss": 2.8431421518325806, "z_loss": 15.697229146957397, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2851952016353607, "tokens_per_second": 2745854.048414926, "eta_seconds": 8290.911728954316, "data_wait_seconds": 1.7344951629638672e-05}
692
+ {"step": 6910, "total_steps": 28611, "tokens": 7245660160, "loss": 2.8461649417877197, "ce_loss": 2.8442388772964478, "z_loss": 19.26102066040039, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31995338201522827, "tokens_per_second": 3057629.4564262517, "eta_seconds": 7442.088094806671, "data_wait_seconds": 1.6987323760986328e-05}
693
+ {"step": 6920, "total_steps": 28611, "tokens": 7256145920, "loss": 2.8407110571861267, "ce_loss": 2.8387425541877747, "z_loss": 19.685617446899414, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28914114832878113, "tokens_per_second": 2818431.676058349, "eta_seconds": 8069.971044254303, "data_wait_seconds": 1.800060272216797e-05}
694
+ {"step": 6930, "total_steps": 28611, "tokens": 7266631680, "loss": 2.8474809527397156, "ce_loss": 2.845392167568207, "z_loss": 20.887803077697754, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2969744801521301, "tokens_per_second": 4072784.190933956, "eta_seconds": 5581.974195098877, "data_wait_seconds": 1.2934207916259766e-05}
695
+ {"step": 6940, "total_steps": 28611, "tokens": 7277117440, "loss": 2.8411552906036377, "ce_loss": 2.8393421173095703, "z_loss": 18.131633758544922, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3321095407009125, "tokens_per_second": 2936689.5759666753, "eta_seconds": 7737.859214663506, "data_wait_seconds": 1.5020370483398438e-05}
696
+ {"step": 6950, "total_steps": 28611, "tokens": 7287603200, "loss": 2.8282394409179688, "ce_loss": 2.826275050640106, "z_loss": 19.64355754852295, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27348703145980835, "tokens_per_second": 4085859.0925884154, "eta_seconds": 5558.978961658478, "data_wait_seconds": 1.615285873413086e-05}
697
+ {"step": 6960, "total_steps": 28611, "tokens": 7298088960, "loss": 2.7983423471450806, "ce_loss": 2.7964512705802917, "z_loss": 18.910154819488525, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29815250635147095, "tokens_per_second": 2815942.4651695676, "eta_seconds": 8062.209813165664, "data_wait_seconds": 1.3828277587890625e-05}
698
+ {"step": 6970, "total_steps": 28611, "tokens": 7308574720, "loss": 2.8382362127304077, "ce_loss": 2.8361958265304565, "z_loss": 20.403419494628906, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2778300344944, "tokens_per_second": 3008435.028446101, "eta_seconds": 7542.869632029533, "data_wait_seconds": 1.7762184143066406e-05}
699
+ {"step": 6980, "total_steps": 28611, "tokens": 7319060480, "loss": 2.824591100215912, "ce_loss": 2.822999358177185, "z_loss": 15.91706657409668, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2662107050418854, "tokens_per_second": 4240862.3411202505, "eta_seconds": 5348.380973386765, "data_wait_seconds": 1.4066696166992188e-05}
700
+ {"step": 6990, "total_steps": 28611, "tokens": 7329546240, "loss": 2.8326406478881836, "ce_loss": 2.83073228597641, "z_loss": 19.08359146118164, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25459378957748413, "tokens_per_second": 4970768.518238391, "eta_seconds": 4560.916810512543, "data_wait_seconds": 1.2874603271484375e-05}
701
+ {"step": 7000, "total_steps": 28611, "tokens": 7340032000, "loss": 2.8061246275901794, "ce_loss": 2.8046345114707947, "z_loss": 14.901373386383057, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30112504959106445, "tokens_per_second": 4324758.717403017, "eta_seconds": 5239.778081679344, "data_wait_seconds": 1.3768672943115234e-05}
702
+ {"step": 7010, "total_steps": 28611, "tokens": 7350517760, "loss": 2.832112729549408, "ce_loss": 2.8304176330566406, "z_loss": 16.95162057876587, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2682516872882843, "tokens_per_second": 3114861.764140459, "eta_seconds": 7271.68391122818, "data_wait_seconds": 1.4603137969970703e-05}
703
+ {"step": 7020, "total_steps": 28611, "tokens": 7361003520, "loss": 2.8035317063331604, "ce_loss": 2.8017784357070923, "z_loss": 17.532676219940186, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24535682797431946, "tokens_per_second": 4368413.378947907, "eta_seconds": 5182.6149340868, "data_wait_seconds": 1.424551010131836e-05}
704
+ {"step": 7030, "total_steps": 28611, "tokens": 7371489280, "loss": 2.7947686314582825, "ce_loss": 2.7931588292121887, "z_loss": 16.097844123840332, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2882784605026245, "tokens_per_second": 3383496.0769816455, "eta_seconds": 6688.146857905388, "data_wait_seconds": 0.2739301919937134}
705
+ {"step": 7040, "total_steps": 28611, "tokens": 7381975040, "loss": 2.816434323787689, "ce_loss": 2.8140628337860107, "z_loss": 23.714616298675537, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3199586570262909, "tokens_per_second": 7951052.031842922, "eta_seconds": 2844.7597632884977, "data_wait_seconds": 1.3768672943115234e-05}
706
+ {"step": 7050, "total_steps": 28611, "tokens": 7392460800, "loss": 2.815377175807953, "ce_loss": 2.8138092756271362, "z_loss": 15.679394245147705, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2829550802707672, "tokens_per_second": 4288307.8554221755, "eta_seconds": 5272.090507078171, "data_wait_seconds": 1.3768672943115234e-05}
707
+ {"step": 7060, "total_steps": 28611, "tokens": 7402946560, "loss": 2.8385114669799805, "ce_loss": 2.836876690387726, "z_loss": 16.347633600234985, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3039693832397461, "tokens_per_second": 4017136.7809684356, "eta_seconds": 5625.365181255341, "data_wait_seconds": 1.424551010131836e-05}
708
+ {"step": 7070, "total_steps": 28611, "tokens": 7413432320, "loss": 2.8240159153938293, "ce_loss": 2.822198748588562, "z_loss": 18.17235231399536, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3011053204536438, "tokens_per_second": 4226379.43136098, "eta_seconds": 5344.379505634308, "data_wait_seconds": 1.5079975128173828e-05}
709
+ {"step": 7080, "total_steps": 28611, "tokens": 7423918080, "loss": 2.781406044960022, "ce_loss": 2.7799500823020935, "z_loss": 14.559504985809326, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2790869176387787, "tokens_per_second": 3136985.5393514554, "eta_seconds": 7197.0015713453295, "data_wait_seconds": 1.7523765563964844e-05}
710
+ {"step": 7090, "total_steps": 28611, "tokens": 7434403840, "loss": 2.798447906970978, "ce_loss": 2.796723425388336, "z_loss": 17.244690418243408, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3190578818321228, "tokens_per_second": 4000132.164045886, "eta_seconds": 5641.414625954628, "data_wait_seconds": 1.3470649719238281e-05}
711
+ {"step": 7100, "total_steps": 28611, "tokens": 7444889600, "loss": 2.8172391653060913, "ce_loss": 2.815069854259491, "z_loss": 21.692296504974365, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3091468811035156, "tokens_per_second": 2983367.4014079743, "eta_seconds": 7560.556680130959, "data_wait_seconds": 1.537799835205078e-05}
712
+ {"step": 7110, "total_steps": 28611, "tokens": 7455375360, "loss": 2.797574460506439, "ce_loss": 2.7961711287498474, "z_loss": 14.032901287078857, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29758724570274353, "tokens_per_second": 2986517.835192678, "eta_seconds": 7549.070127868652, "data_wait_seconds": 1.8298625946044922e-05}
713
+ {"step": 7120, "total_steps": 28611, "tokens": 7465861120, "loss": 2.814575731754303, "ce_loss": 2.812764346599579, "z_loss": 18.11399555206299, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26533743739128113, "tokens_per_second": 4334333.114619034, "eta_seconds": 5199.1727954626085, "data_wait_seconds": 1.3768672943115234e-05}
714
+ {"step": 7130, "total_steps": 28611, "tokens": 7476346880, "loss": 2.813421070575714, "ce_loss": 2.8116204738616943, "z_loss": 18.00556230545044, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3430495262145996, "tokens_per_second": 2959401.7247813763, "eta_seconds": 7611.153588032723, "data_wait_seconds": 1.531839370727539e-05}
715
+ {"step": 7140, "total_steps": 28611, "tokens": 7486832640, "loss": 2.8491832613945007, "ce_loss": 2.847568154335022, "z_loss": 16.15071153640747, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25743231177330017, "tokens_per_second": 2858148.999037156, "eta_seconds": 7877.117429351806, "data_wait_seconds": 0.28049129247665405}
716
+ {"step": 7150, "total_steps": 28611, "tokens": 7497318400, "loss": 2.8146925568580627, "ce_loss": 2.8132684230804443, "z_loss": 14.242090702056885, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3015616834163666, "tokens_per_second": 4141925.5729876226, "eta_seconds": 5433.098480272293, "data_wait_seconds": 1.817941665649414e-05}