Banaxi-Tech commited on
Commit
a60975c
·
verified ·
1 Parent(s): e51a111

Save 15% checkpoint at 4,500,488,192 tokens

Browse files
checkpoint_metadata.json CHANGED
@@ -49,10 +49,10 @@
49
  "model_type": "bird_alpha",
50
  "output_attentions": false
51
  },
52
- "training_percent": 10,
53
- "step": 2862,
54
  "total_steps": 28611,
55
- "tokens_seen": 3001024512,
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": 904.7069578170776
72
  }
 
49
  "model_type": "bird_alpha",
50
  "output_attentions": false
51
  },
52
+ "training_percent": 15,
53
+ "step": 4292,
54
  "total_steps": 28611,
55
+ "tokens_seen": 4500488192,
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": 1350.2359328269958
72
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:61fbd3dff01152e0b26d6924c1c5590d0a8a2f88d1eff18e4a0d7d734486673a
3
  size 13860292
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5dbf1afb9cbf6681d54d00b54cd1b2815128b88aaf0aab371b7819eaf9cc33a
3
  size 13860292
training_metrics.jsonl CHANGED
@@ -285,3 +285,146 @@
285
  {"step": 2840, "total_steps": 28611, "tokens": 2977955840, "loss": 2.8871544003486633, "ce_loss": 2.8839539885520935, "z_loss": 32.00378465652466, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26920804381370544, "tokens_per_second": 2800610.3146087537, "eta_seconds": 9648.915436410904, "data_wait_seconds": 1.424551010131836e-05}
286
  {"step": 2850, "total_steps": 28611, "tokens": 2988441600, "loss": 2.9108477234840393, "ce_loss": 2.9073694944381714, "z_loss": 34.78234577178955, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2519266903400421, "tokens_per_second": 2913887.39614185, "eta_seconds": 9270.216265654564, "data_wait_seconds": 1.7762184143066406e-05}
287
  {"step": 2860, "total_steps": 28611, "tokens": 2998927360, "loss": 2.8743998408317566, "ce_loss": 2.8707590103149414, "z_loss": 36.40823459625244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2784413993358612, "tokens_per_second": 2859233.4001113516, "eta_seconds": 9443.748305034638, "data_wait_seconds": 2.0205974578857422e-05}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
285
  {"step": 2840, "total_steps": 28611, "tokens": 2977955840, "loss": 2.8871544003486633, "ce_loss": 2.8839539885520935, "z_loss": 32.00378465652466, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26920804381370544, "tokens_per_second": 2800610.3146087537, "eta_seconds": 9648.915436410904, "data_wait_seconds": 1.424551010131836e-05}
286
  {"step": 2850, "total_steps": 28611, "tokens": 2988441600, "loss": 2.9108477234840393, "ce_loss": 2.9073694944381714, "z_loss": 34.78234577178955, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2519266903400421, "tokens_per_second": 2913887.39614185, "eta_seconds": 9270.216265654564, "data_wait_seconds": 1.7762184143066406e-05}
287
  {"step": 2860, "total_steps": 28611, "tokens": 2998927360, "loss": 2.8743998408317566, "ce_loss": 2.8707590103149414, "z_loss": 36.40823459625244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2784413993358612, "tokens_per_second": 2859233.4001113516, "eta_seconds": 9443.748305034638, "data_wait_seconds": 2.0205974578857422e-05}
288
+ {"step": 2870, "total_steps": 28611, "tokens": 3009413120, "loss": 2.916546046733856, "ce_loss": 2.9131298065185547, "z_loss": 34.162925720214844, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24387343227863312, "tokens_per_second": 9419859.643014012, "eta_seconds": 2865.371230453253, "data_wait_seconds": 1.430511474609375e-05}
289
+ {"step": 2880, "total_steps": 28611, "tokens": 3019898880, "loss": 2.8398979902267456, "ce_loss": 2.836251199245453, "z_loss": 36.46790027618408, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22986264526844025, "tokens_per_second": 2702835.2620917414, "eta_seconds": 9982.44674191475, "data_wait_seconds": 1.4960765838623047e-05}
290
+ {"step": 2890, "total_steps": 28611, "tokens": 3030384640, "loss": 2.8537875413894653, "ce_loss": 2.8501358032226562, "z_loss": 36.517751693725586, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.267050176858902, "tokens_per_second": 2845111.522666495, "eta_seconds": 9479.566295075416, "data_wait_seconds": 1.7583370208740234e-05}
291
+ {"step": 2900, "total_steps": 28611, "tokens": 3040870400, "loss": 2.865070879459381, "ce_loss": 2.861818790435791, "z_loss": 32.520859718322754, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24041295051574707, "tokens_per_second": 4487684.875104806, "eta_seconds": 6007.53802602291, "data_wait_seconds": 1.9490718841552734e-05}
292
+ {"step": 2910, "total_steps": 28611, "tokens": 3051356160, "loss": 2.8788344264030457, "ce_loss": 2.8755608201026917, "z_loss": 32.73593521118164, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24248911440372467, "tokens_per_second": 2938092.095522924, "eta_seconds": 9172.43261947632, "data_wait_seconds": 0.2791767716407776}
293
+ {"step": 2920, "total_steps": 28611, "tokens": 3061841920, "loss": 2.856222093105316, "ce_loss": 2.852574110031128, "z_loss": 36.4799165725708, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25646868348121643, "tokens_per_second": 4512039.581498905, "eta_seconds": 5970.4631418704985, "data_wait_seconds": 1.2636184692382812e-05}
294
+ {"step": 2930, "total_steps": 28611, "tokens": 3072327680, "loss": 2.908151090145111, "ce_loss": 2.9049218893051147, "z_loss": 32.29164266586304, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2635818421840668, "tokens_per_second": 2798764.4782532635, "eta_seconds": 9621.559965205193, "data_wait_seconds": 1.3887882232666016e-05}
295
+ {"step": 2940, "total_steps": 28611, "tokens": 3082813440, "loss": 2.898173451423645, "ce_loss": 2.894744038581848, "z_loss": 34.293938636779785, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2598417103290558, "tokens_per_second": 2936304.308940278, "eta_seconds": 9167.30408835411, "data_wait_seconds": 1.7523765563964844e-05}
296
+ {"step": 2950, "total_steps": 28611, "tokens": 3093299200, "loss": 2.861096441745758, "ce_loss": 2.857015311717987, "z_loss": 40.81107807159424, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26132166385650635, "tokens_per_second": 3259787.4201582884, "eta_seconds": 8254.375291347504, "data_wait_seconds": 1.4662742614746094e-05}
297
+ {"step": 2960, "total_steps": 28611, "tokens": 3103784960, "loss": 2.8082245588302612, "ce_loss": 2.8046393394470215, "z_loss": 35.85195064544678, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28148338198661804, "tokens_per_second": 2682314.1741630496, "eta_seconds": 10027.54384071827, "data_wait_seconds": 1.6868114471435547e-05}
298
+ {"step": 2970, "total_steps": 28611, "tokens": 3114270720, "loss": 2.9017773270606995, "ce_loss": 2.8981144428253174, "z_loss": 36.62891483306885, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2728983461856842, "tokens_per_second": 2699484.2929202877, "eta_seconds": 9959.879109692572, "data_wait_seconds": 1.8894672393798828e-05}
299
+ {"step": 2980, "total_steps": 28611, "tokens": 3124756480, "loss": 2.8799219727516174, "ce_loss": 2.87665331363678, "z_loss": 32.68609142303467, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29763904213905334, "tokens_per_second": 4160261.758010784, "eta_seconds": 6460.18280081749, "data_wait_seconds": 1.3709068298339844e-05}
300
+ {"step": 2990, "total_steps": 28611, "tokens": 3135242240, "loss": 2.9234662652015686, "ce_loss": 2.919987440109253, "z_loss": 34.78783559799194, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2761668264865875, "tokens_per_second": 2876284.021020849, "eta_seconds": 9340.37302980423, "data_wait_seconds": 0.29478317499160767}
301
+ {"step": 3000, "total_steps": 28611, "tokens": 3145728000, "loss": 2.865002930164337, "ce_loss": 2.8620213866233826, "z_loss": 29.815858364105225, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2695026099681854, "tokens_per_second": 4216931.010955851, "eta_seconds": 6368.394423866272, "data_wait_seconds": 1.3887882232666016e-05}
302
+ {"step": 3010, "total_steps": 28611, "tokens": 3156213760, "loss": 2.854705333709717, "ce_loss": 2.851305902004242, "z_loss": 33.99454212188721, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25757357478141785, "tokens_per_second": 2990113.1581359487, "eta_seconds": 8977.785373425484, "data_wait_seconds": 1.823902130126953e-05}
303
+ {"step": 3020, "total_steps": 28611, "tokens": 3166699520, "loss": 2.9177645444869995, "ce_loss": 2.9142134189605713, "z_loss": 35.51128005981445, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24591702222824097, "tokens_per_second": 4363434.873004355, "eta_seconds": 6149.767143774033, "data_wait_seconds": 1.1920928955078125e-05}
304
+ {"step": 3030, "total_steps": 28611, "tokens": 3177185280, "loss": 2.884656608104706, "ce_loss": 2.8811251521110535, "z_loss": 35.31455326080322, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24901461601257324, "tokens_per_second": 3099044.507392073, "eta_seconds": 8655.449314141273, "data_wait_seconds": 0.5606999397277832}
305
+ {"step": 3040, "total_steps": 28611, "tokens": 3187671040, "loss": 2.8555720448493958, "ce_loss": 2.852483332157135, "z_loss": 30.886861324310303, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2906537353992462, "tokens_per_second": 4628939.396903358, "eta_seconds": 5792.501175093651, "data_wait_seconds": 1.4007091522216797e-05}
306
+ {"step": 3050, "total_steps": 28611, "tokens": 3198156800, "loss": 2.893101751804352, "ce_loss": 2.8895614743232727, "z_loss": 35.4028959274292, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24810445308685303, "tokens_per_second": 2881816.1298247715, "eta_seconds": 9300.611117625236, "data_wait_seconds": 1.6033649444580078e-05}
307
+ {"step": 3060, "total_steps": 28611, "tokens": 3208642560, "loss": 2.8887308835983276, "ce_loss": 2.885348081588745, "z_loss": 33.82797145843506, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2656812369823456, "tokens_per_second": 4173950.054445841, "eta_seconds": 6418.899370265007, "data_wait_seconds": 1.6748905181884766e-05}
308
+ {"step": 3070, "total_steps": 28611, "tokens": 3219128320, "loss": 2.8657562136650085, "ce_loss": 2.8622581362724304, "z_loss": 34.98049736022949, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23533643782138824, "tokens_per_second": 2894519.767922255, "eta_seconds": 9252.546799921989, "data_wait_seconds": 1.7642974853515625e-05}
309
+ {"step": 3080, "total_steps": 28611, "tokens": 3229614080, "loss": 2.8901990056037903, "ce_loss": 2.886555850505829, "z_loss": 36.43114471435547, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23857377469539642, "tokens_per_second": 4751632.598646653, "eta_seconds": 5634.104342079162, "data_wait_seconds": 1.7940998077392578e-05}
310
+ {"step": 3090, "total_steps": 28611, "tokens": 3240099840, "loss": 2.8778046369552612, "ce_loss": 2.8747059106826782, "z_loss": 30.98683214187622, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24789373576641083, "tokens_per_second": 2849927.078216063, "eta_seconds": 9389.962396073342, "data_wait_seconds": 1.4960765838623047e-05}
311
+ {"step": 3100, "total_steps": 28611, "tokens": 3250585600, "loss": 2.93124121427536, "ce_loss": 2.9278863668441772, "z_loss": 33.548420906066895, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26118797063827515, "tokens_per_second": 4183164.066091391, "eta_seconds": 6394.734204387665, "data_wait_seconds": 1.609325408935547e-05}
312
+ {"step": 3110, "total_steps": 28611, "tokens": 3261071360, "loss": 2.883722245693207, "ce_loss": 2.880606532096863, "z_loss": 31.157007694244385, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.288857638835907, "tokens_per_second": 2958164.616894156, "eta_seconds": 9039.299714183808, "data_wait_seconds": 1.519918441772461e-05}
313
+ {"step": 3120, "total_steps": 28611, "tokens": 3271557120, "loss": 2.894721567630768, "ce_loss": 2.891453742980957, "z_loss": 32.67856216430664, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2713092565536499, "tokens_per_second": 4430357.551764776, "eta_seconds": 6033.203980422019, "data_wait_seconds": 1.6391277313232422e-05}
314
+ {"step": 3130, "total_steps": 28611, "tokens": 3282042880, "loss": 2.8789072036743164, "ce_loss": 2.875434100627899, "z_loss": 34.73122310638428, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2822877764701843, "tokens_per_second": 3030516.3227919154, "eta_seconds": 8816.571900653838, "data_wait_seconds": 1.5437602996826172e-05}
315
+ {"step": 3140, "total_steps": 28611, "tokens": 3292528640, "loss": 2.8831822276115417, "ce_loss": 2.880283236503601, "z_loss": 28.990469455718994, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2334170639514923, "tokens_per_second": 2744298.487603022, "eta_seconds": 9732.279275250434, "data_wait_seconds": 0.30412864685058594}
316
+ {"step": 3150, "total_steps": 28611, "tokens": 3303014400, "loss": 2.8726117610931396, "ce_loss": 2.868997573852539, "z_loss": 36.141730308532715, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2562882602214813, "tokens_per_second": 2840005.733615187, "eta_seconds": 9400.612548065186, "data_wait_seconds": 1.8596649169921875e-05}
317
+ {"step": 3160, "total_steps": 28611, "tokens": 3313500160, "loss": 2.8902586102485657, "ce_loss": 2.8870092630386353, "z_loss": 32.493109703063965, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25964176654815674, "tokens_per_second": 2781615.519089947, "eta_seconds": 9594.175612282754, "data_wait_seconds": 1.6391277313232422e-05}
318
+ {"step": 3170, "total_steps": 28611, "tokens": 3323985920, "loss": 2.8871325850486755, "ce_loss": 2.8836175203323364, "z_loss": 35.15096855163574, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29834485054016113, "tokens_per_second": 3002006.7848895127, "eta_seconds": 8886.329687952995, "data_wait_seconds": 1.823902130126953e-05}
319
+ {"step": 3180, "total_steps": 28611, "tokens": 3334471680, "loss": 2.8799163103103638, "ce_loss": 2.876733660697937, "z_loss": 31.82649850845337, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24526891112327576, "tokens_per_second": 2362413.4387695557, "eta_seconds": 11287.751677322387, "data_wait_seconds": 0.27362287044525146}
320
+ {"step": 3190, "total_steps": 28611, "tokens": 3344957440, "loss": 2.872092843055725, "ce_loss": 2.8688647747039795, "z_loss": 32.28059768676758, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2565166652202606, "tokens_per_second": 4079123.8816302335, "eta_seconds": 6534.6999182939535, "data_wait_seconds": 1.5616416931152344e-05}
321
+ {"step": 3200, "total_steps": 28611, "tokens": 3355443200, "loss": 2.861978769302368, "ce_loss": 2.85849392414093, "z_loss": 34.84890937805176, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2559097409248352, "tokens_per_second": 2784428.2787192096, "eta_seconds": 9569.42038681507, "data_wait_seconds": 1.9609928131103516e-05}
322
+ {"step": 3210, "total_steps": 28611, "tokens": 3365928960, "loss": 2.8794209361076355, "ce_loss": 2.876593232154846, "z_loss": 28.277143955230713, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2801010310649872, "tokens_per_second": 4046097.526942419, "eta_seconds": 6582.8563940048225, "data_wait_seconds": 1.4960765838623047e-05}
323
+ {"step": 3220, "total_steps": 28611, "tokens": 3376414720, "loss": 2.869552254676819, "ce_loss": 2.866482436656952, "z_loss": 30.69802761077881, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2648608088493347, "tokens_per_second": 3011139.4626220423, "eta_seconds": 8841.966154837608, "data_wait_seconds": 1.33514404296875e-05}
324
+ {"step": 3230, "total_steps": 28611, "tokens": 3386900480, "loss": 2.9063108563423157, "ce_loss": 2.903097152709961, "z_loss": 32.13708209991455, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29611116647720337, "tokens_per_second": 2955985.9231927646, "eta_seconds": 9003.394517946243, "data_wait_seconds": 0.27626103162765503}
325
+ {"step": 3240, "total_steps": 28611, "tokens": 3397386240, "loss": 2.9245113134384155, "ce_loss": 2.921626031398773, "z_loss": 28.85341787338257, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2883613705635071, "tokens_per_second": 3828569.067947416, "eta_seconds": 6948.659205007553, "data_wait_seconds": 1.2576580047607422e-05}
326
+ {"step": 3250, "total_steps": 28611, "tokens": 3407872000, "loss": 2.9011733531951904, "ce_loss": 2.898045063018799, "z_loss": 31.283095836639404, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28363341093063354, "tokens_per_second": 2305827.4538434376, "eta_seconds": 11532.925367712975, "data_wait_seconds": 0.211927592754364}
327
+ {"step": 3260, "total_steps": 28611, "tokens": 3418357760, "loss": 2.850818932056427, "ce_loss": 2.8471381664276123, "z_loss": 36.807634353637695, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22865699231624603, "tokens_per_second": 3926068.6695034993, "eta_seconds": 6770.755280590058, "data_wait_seconds": 1.2695789337158203e-05}
328
+ {"step": 3270, "total_steps": 28611, "tokens": 3428843520, "loss": 2.9011740684509277, "ce_loss": 2.8983503580093384, "z_loss": 28.23693084716797, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2546396851539612, "tokens_per_second": 2982268.3111941693, "eta_seconds": 8909.984496116638, "data_wait_seconds": 0.3289283514022827}
329
+ {"step": 3280, "total_steps": 28611, "tokens": 3439329280, "loss": 2.9071662425994873, "ce_loss": 2.9040514826774597, "z_loss": 31.14702796936035, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2931673526763916, "tokens_per_second": 4690934.38105386, "eta_seconds": 5662.300194025039, "data_wait_seconds": 1.3649463653564453e-05}
330
+ {"step": 3290, "total_steps": 28611, "tokens": 3449815040, "loss": 2.886256694793701, "ce_loss": 2.883497714996338, "z_loss": 27.589672565460205, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25123724341392517, "tokens_per_second": 4186073.7813914143, "eta_seconds": 6342.695872688293, "data_wait_seconds": 1.4185905456542969e-05}
331
+ {"step": 3300, "total_steps": 28611, "tokens": 3460300800, "loss": 2.883204698562622, "ce_loss": 2.880733907222748, "z_loss": 24.70747709274292, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2756970524787903, "tokens_per_second": 4143809.697226202, "eta_seconds": 6404.8566597461695, "data_wait_seconds": 1.3947486877441406e-05}
332
+ {"step": 3310, "total_steps": 28611, "tokens": 3470786560, "loss": 2.877068340778351, "ce_loss": 2.874182879924774, "z_loss": 28.85512351989746, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26119348406791687, "tokens_per_second": 4115367.667535707, "eta_seconds": 6446.573798322677, "data_wait_seconds": 1.9490718841552734e-05}
333
+ {"step": 3320, "total_steps": 28611, "tokens": 3481272320, "loss": 2.915669023990631, "ce_loss": 2.9124456644058228, "z_loss": 32.23306941986084, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29705339670181274, "tokens_per_second": 9448023.661278829, "eta_seconds": 2806.8870873689652, "data_wait_seconds": 1.2874603271484375e-05}
334
+ {"step": 3330, "total_steps": 28611, "tokens": 3491758080, "loss": 2.8439339995384216, "ce_loss": 2.8409007787704468, "z_loss": 30.332047939300537, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24415256083011627, "tokens_per_second": 4148361.6041821986, "eta_seconds": 6390.24568862915, "data_wait_seconds": 1.519918441772461e-05}
335
+ {"step": 3340, "total_steps": 28611, "tokens": 3502243840, "loss": 2.9553523659706116, "ce_loss": 2.9523842334747314, "z_loss": 29.681249618530273, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27161428332328796, "tokens_per_second": 2899718.669015515, "eta_seconds": 9138.322410082817, "data_wait_seconds": 1.901388168334961e-05}
336
+ {"step": 3350, "total_steps": 28611, "tokens": 3512729600, "loss": 2.892702043056488, "ce_loss": 2.889846622943878, "z_loss": 28.554262161254883, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28953710198402405, "tokens_per_second": 4298875.326313736, "eta_seconds": 6161.629804396629, "data_wait_seconds": 1.722574234008789e-05}
337
+ {"step": 3360, "total_steps": 28611, "tokens": 3523215360, "loss": 2.8805675506591797, "ce_loss": 2.877698004245758, "z_loss": 28.69543981552124, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2645251154899597, "tokens_per_second": 9184607.427140499, "eta_seconds": 2882.8224598646166, "data_wait_seconds": 1.6033649444580078e-05}
338
+ {"step": 3370, "total_steps": 28611, "tokens": 3533701120, "loss": 2.87846040725708, "ce_loss": 2.8754794001579285, "z_loss": 29.810107707977295, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2628192603588104, "tokens_per_second": 3246887.7758876765, "eta_seconds": 8151.531140851974, "data_wait_seconds": 1.3828277587890625e-05}
339
+ {"step": 3380, "total_steps": 28611, "tokens": 3544186880, "loss": 2.871984124183655, "ce_loss": 2.869065225124359, "z_loss": 29.188774585723877, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2594360411167145, "tokens_per_second": 2905644.4626183677, "eta_seconds": 9105.250623869895, "data_wait_seconds": 2.014636993408203e-05}
340
+ {"step": 3390, "total_steps": 28611, "tokens": 3554672640, "loss": 2.8475882411003113, "ce_loss": 2.844340205192566, "z_loss": 32.48035430908203, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2597535252571106, "tokens_per_second": 3112655.9415670494, "eta_seconds": 8496.324615526199, "data_wait_seconds": 1.9431114196777344e-05}
341
+ {"step": 3400, "total_steps": 28611, "tokens": 3565158400, "loss": 2.8825706839561462, "ce_loss": 2.879528820514679, "z_loss": 30.41857147216797, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24934855103492737, "tokens_per_second": 3893914.6240393473, "eta_seconds": 6788.964856290818, "data_wait_seconds": 1.823902130126953e-05}
342
+ {"step": 3410, "total_steps": 28611, "tokens": 3575644160, "loss": 2.8980414271354675, "ce_loss": 2.8953505754470825, "z_loss": 26.908515453338623, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2555544078350067, "tokens_per_second": 3037962.9527886347, "eta_seconds": 8698.316663718224, "data_wait_seconds": 0.2557406425476074}
343
+ {"step": 3420, "total_steps": 28611, "tokens": 3586129920, "loss": 2.8384000062942505, "ce_loss": 2.8356472849845886, "z_loss": 27.526902198791504, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27823206782341003, "tokens_per_second": 2855494.9849545965, "eta_seconds": 9250.47256436348, "data_wait_seconds": 1.3530254364013672e-05}
344
+ {"step": 3430, "total_steps": 28611, "tokens": 3596615680, "loss": 2.8973003029823303, "ce_loss": 2.8947436213493347, "z_loss": 25.566946029663086, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2304992377758026, "tokens_per_second": 3084451.0044585904, "eta_seconds": 8560.41876425743, "data_wait_seconds": 1.2516975402832031e-05}
345
+ {"step": 3440, "total_steps": 28611, "tokens": 3607101440, "loss": 2.8683072328567505, "ce_loss": 2.865647315979004, "z_loss": 26.59895372390747, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2680673599243164, "tokens_per_second": 2902669.729048084, "eta_seconds": 9092.90720603466, "data_wait_seconds": 1.519918441772461e-05}
346
+ {"step": 3450, "total_steps": 28611, "tokens": 3617587200, "loss": 2.922609269618988, "ce_loss": 2.919357419013977, "z_loss": 32.51847219467163, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27738863229751587, "tokens_per_second": 4538321.490039635, "eta_seconds": 5813.431418180465, "data_wait_seconds": 1.2874603271484375e-05}
347
+ {"step": 3460, "total_steps": 28611, "tokens": 3628072960, "loss": 2.8956320881843567, "ce_loss": 2.8924534916877747, "z_loss": 31.786152839660645, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24719467759132385, "tokens_per_second": 3091902.136007948, "eta_seconds": 8529.615044689179, "data_wait_seconds": 1.704692840576172e-05}
348
+ {"step": 3470, "total_steps": 28611, "tokens": 3638558720, "loss": 2.8810558319091797, "ce_loss": 2.8785287737846375, "z_loss": 25.270629405975342, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2930685877799988, "tokens_per_second": 3081055.7156614964, "eta_seconds": 8556.239045596123, "data_wait_seconds": 0.24744123220443726}
349
+ {"step": 3480, "total_steps": 28611, "tokens": 3649044480, "loss": 2.894436478614807, "ce_loss": 2.891871929168701, "z_loss": 25.645724296569824, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26069340109825134, "tokens_per_second": 3665716.305817192, "eta_seconds": 7188.707815217972, "data_wait_seconds": 1.3232231140136719e-05}
350
+ {"step": 3490, "total_steps": 28611, "tokens": 3659530240, "loss": 2.8853651881217957, "ce_loss": 2.8830569982528687, "z_loss": 23.08219289779663, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2482461929321289, "tokens_per_second": 2695492.385868586, "eta_seconds": 9772.34357407093, "data_wait_seconds": 1.919269561767578e-05}
351
+ {"step": 3500, "total_steps": 28611, "tokens": 3670016000, "loss": 2.9004395604133606, "ce_loss": 2.8975459933280945, "z_loss": 28.936074256896973, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2902616560459137, "tokens_per_second": 4121383.014968765, "eta_seconds": 6388.82429523468, "data_wait_seconds": 1.3828277587890625e-05}
352
+ {"step": 3510, "total_steps": 28611, "tokens": 3680501760, "loss": 2.9191659688949585, "ce_loss": 2.9166293144226074, "z_loss": 25.366440296173096, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26079291105270386, "tokens_per_second": 2762368.8884830032, "eta_seconds": 9528.164860868454, "data_wait_seconds": 1.5795230865478516e-05}
353
+ {"step": 3520, "total_steps": 28611, "tokens": 3690987520, "loss": 2.8883764147758484, "ce_loss": 2.885204493999481, "z_loss": 31.719143867492676, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28800150752067566, "tokens_per_second": 2867590.51338586, "eta_seconds": 9174.8875207901, "data_wait_seconds": 0.2555384635925293}
354
+ {"step": 3530, "total_steps": 28611, "tokens": 3701473280, "loss": 2.913647711277008, "ce_loss": 2.910608410835266, "z_loss": 30.39303493499756, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2526278793811798, "tokens_per_second": 2889032.572613601, "eta_seconds": 9103.163081407547, "data_wait_seconds": 0.2777436375617981}
355
+ {"step": 3540, "total_steps": 28611, "tokens": 3711959040, "loss": 2.898298680782318, "ce_loss": 2.895829200744629, "z_loss": 24.695244789123535, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2791030704975128, "tokens_per_second": 4208963.86043078, "eta_seconds": 6245.919368219375, "data_wait_seconds": 1.4007091522216797e-05}
356
+ {"step": 3550, "total_steps": 28611, "tokens": 3722444800, "loss": 2.8777138590812683, "ce_loss": 2.8743021488189697, "z_loss": 34.11749839782715, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2540091872215271, "tokens_per_second": 3013603.4258402046, "eta_seconds": 8719.914143538474, "data_wait_seconds": 1.3768672943115234e-05}
357
+ {"step": 3560, "total_steps": 28611, "tokens": 3732930560, "loss": 2.9061741828918457, "ce_loss": 2.903666377067566, "z_loss": 25.077598571777344, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2712169587612152, "tokens_per_second": 2445128.087166171, "eta_seconds": 10742.945334386824, "data_wait_seconds": 0.22987627983093262}
358
+ {"step": 3570, "total_steps": 28611, "tokens": 3743416320, "loss": 2.8616889119148254, "ce_loss": 2.858570873737335, "z_loss": 31.180416107177734, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.20598703622817993, "tokens_per_second": 2982296.0161978053, "eta_seconds": 8804.4216514349, "data_wait_seconds": 2.396106719970703e-05}
359
+ {"step": 3580, "total_steps": 28611, "tokens": 3753902080, "loss": 2.9141327142715454, "ce_loss": 2.9110668897628784, "z_loss": 30.657902240753174, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2732083797454834, "tokens_per_second": 4734804.765225929, "eta_seconds": 5543.397702217102, "data_wait_seconds": 1.728534698486328e-05}
360
+ {"step": 3590, "total_steps": 28611, "tokens": 3764387840, "loss": 2.889146327972412, "ce_loss": 2.88612300157547, "z_loss": 30.2337703704834, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2712371051311493, "tokens_per_second": 3111247.799568151, "eta_seconds": 8432.764532494544, "data_wait_seconds": 1.5020370483398438e-05}
361
+ {"step": 3600, "total_steps": 28611, "tokens": 3774873600, "loss": 2.9319531321525574, "ce_loss": 2.9292629957199097, "z_loss": 26.901082038879395, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29004156589508057, "tokens_per_second": 3046726.23521794, "eta_seconds": 8607.906425213814, "data_wait_seconds": 0.26003140211105347}
362
+ {"step": 3610, "total_steps": 28611, "tokens": 3785359360, "loss": 2.847970426082611, "ce_loss": 2.8454629778862, "z_loss": 25.07443141937256, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28253883123397827, "tokens_per_second": 4516696.051346424, "eta_seconds": 5804.120595669747, "data_wait_seconds": 1.3053417205810547e-05}
363
+ {"step": 3620, "total_steps": 28611, "tokens": 3795845120, "loss": 2.881714701652527, "ce_loss": 2.8785241842269897, "z_loss": 31.9050235748291, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.254898339509964, "tokens_per_second": 2946310.991093566, "eta_seconds": 8894.160492634774, "data_wait_seconds": 1.996755599975586e-05}
364
+ {"step": 3630, "total_steps": 28611, "tokens": 3806330880, "loss": 2.874914765357971, "ce_loss": 2.8721293807029724, "z_loss": 27.854199409484863, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25415289402008057, "tokens_per_second": 4549713.484982639, "eta_seconds": 5757.390469193459, "data_wait_seconds": 1.615285873413086e-05}
365
+ {"step": 3640, "total_steps": 28611, "tokens": 3816816640, "loss": 2.8962759375572205, "ce_loss": 2.893887519836426, "z_loss": 23.88450813293457, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30204063653945923, "tokens_per_second": 2869902.590809466, "eta_seconds": 9123.651576137543, "data_wait_seconds": 1.2695789337158203e-05}
366
+ {"step": 3650, "total_steps": 28611, "tokens": 3827302400, "loss": 2.8983744382858276, "ce_loss": 2.895540714263916, "z_loss": 28.33759880065918, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27405640482902527, "tokens_per_second": 2222847.8499501585, "eta_seconds": 11774.762513136862, "data_wait_seconds": 0.3016403317451477}
367
+ {"step": 3660, "total_steps": 28611, "tokens": 3837788160, "loss": 2.8638484477996826, "ce_loss": 2.861266613006592, "z_loss": 25.817912101745605, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2648298144340515, "tokens_per_second": 4236519.87069022, "eta_seconds": 6175.5923669815065, "data_wait_seconds": 1.4066696166992188e-05}
368
+ {"step": 3670, "total_steps": 28611, "tokens": 3848273920, "loss": 2.878785490989685, "ce_loss": 2.8762466311454773, "z_loss": 25.388705253601074, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.265455037355423, "tokens_per_second": 2867904.846234519, "eta_seconds": 9119.038259005547, "data_wait_seconds": 1.7821788787841797e-05}
369
+ {"step": 3680, "total_steps": 28611, "tokens": 3858759680, "loss": 2.8772818446159363, "ce_loss": 2.8751190304756165, "z_loss": 21.627578258514404, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2599073052406311, "tokens_per_second": 4612383.511652123, "eta_seconds": 5667.795878195762, "data_wait_seconds": 1.2516975402832031e-05}
370
+ {"step": 3690, "total_steps": 28611, "tokens": 3869245440, "loss": 2.8885242342948914, "ce_loss": 2.8855212330818176, "z_loss": 30.03023624420166, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2723615765571594, "tokens_per_second": 2987739.1973813605, "eta_seconds": 8746.266246700287, "data_wait_seconds": 1.5795230865478516e-05}
371
+ {"step": 3700, "total_steps": 28611, "tokens": 3879731200, "loss": 2.9066967368125916, "ce_loss": 2.904225468635559, "z_loss": 24.712857723236084, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.293924480676651, "tokens_per_second": 2871872.5528313178, "eta_seconds": 9095.486048030853, "data_wait_seconds": 0.26993829011917114}
372
+ {"step": 3710, "total_steps": 28611, "tokens": 3890216960, "loss": 2.9104215502738953, "ce_loss": 2.907707452774048, "z_loss": 27.14108657836914, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30345869064331055, "tokens_per_second": 4339918.938813217, "eta_seconds": 6016.377573895455, "data_wait_seconds": 1.329183578491211e-05}
373
+ {"step": 3720, "total_steps": 28611, "tokens": 3900702720, "loss": 2.9028640389442444, "ce_loss": 2.900670349597931, "z_loss": 21.936374187469482, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2753293812274933, "tokens_per_second": 2995468.6429267502, "eta_seconds": 8713.19593935013, "data_wait_seconds": 1.728534698486328e-05}
374
+ {"step": 3730, "total_steps": 28611, "tokens": 3911188480, "loss": 2.8872520327568054, "ce_loss": 2.88452410697937, "z_loss": 27.278852939605713, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2514028251171112, "tokens_per_second": 4714737.806966868, "eta_seconds": 5533.631036162376, "data_wait_seconds": 1.6570091247558594e-05}
375
+ {"step": 3740, "total_steps": 28611, "tokens": 3921674240, "loss": 2.88142192363739, "ce_loss": 2.8785064220428467, "z_loss": 29.154428482055664, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25635334849357605, "tokens_per_second": 2833891.456986261, "eta_seconds": 9202.587357997894, "data_wait_seconds": 1.5139579772949219e-05}
376
+ {"step": 3750, "total_steps": 28611, "tokens": 3932160000, "loss": 2.863167643547058, "ce_loss": 2.86070054769516, "z_loss": 24.670913696289062, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2782605290412903, "tokens_per_second": 2731449.7280714666, "eta_seconds": 9543.887140989304, "data_wait_seconds": 1.8656253814697266e-05}
377
+ {"step": 3760, "total_steps": 28611, "tokens": 3942645760, "loss": 2.8608888387680054, "ce_loss": 2.858385384082794, "z_loss": 25.035053730010986, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2786448895931244, "tokens_per_second": 3004681.604543129, "eta_seconds": 8672.520288538934, "data_wait_seconds": 0.2462749481201172}
378
+ {"step": 3770, "total_steps": 28611, "tokens": 3953131520, "loss": 2.9080300331115723, "ce_loss": 2.9050240516662598, "z_loss": 30.059865474700928, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2723010778427124, "tokens_per_second": 3016321.9224370415, "eta_seconds": 8635.575739526748, "data_wait_seconds": 1.8417835235595703e-05}
379
+ {"step": 3780, "total_steps": 28611, "tokens": 3963617280, "loss": 2.8906461596488953, "ce_loss": 2.8879833221435547, "z_loss": 26.628015518188477, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.289775550365448, "tokens_per_second": 3158407.7793848924, "eta_seconds": 8243.771062731743, "data_wait_seconds": 1.3649463653564453e-05}
380
+ {"step": 3790, "total_steps": 28611, "tokens": 3974103040, "loss": 2.9458969235420227, "ce_loss": 2.9435579776763916, "z_loss": 23.38923215866089, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2894861698150635, "tokens_per_second": 3096751.558237501, "eta_seconds": 8404.51822063923, "data_wait_seconds": 1.7702579498291016e-05}
381
+ {"step": 3800, "total_steps": 28611, "tokens": 3984588800, "loss": 2.8919557332992554, "ce_loss": 2.889167547225952, "z_loss": 27.88196325302124, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2874719202518463, "tokens_per_second": 4671154.430050309, "eta_seconds": 5569.548069024086, "data_wait_seconds": 1.7404556274414062e-05}
382
+ {"step": 3810, "total_steps": 28611, "tokens": 3995074560, "loss": 2.8708443641662598, "ce_loss": 2.868781626224518, "z_loss": 20.627039909362793, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26126742362976074, "tokens_per_second": 2907957.9620733033, "eta_seconds": 8942.953686118126, "data_wait_seconds": 1.531839370727539e-05}
383
+ {"step": 3820, "total_steps": 28611, "tokens": 4005560320, "loss": 2.8870365619659424, "ce_loss": 2.8844481706619263, "z_loss": 25.883880138397217, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24966609477996826, "tokens_per_second": 3031026.1397423353, "eta_seconds": 8576.385163807869, "data_wait_seconds": 1.5616416931152344e-05}
384
+ {"step": 3830, "total_steps": 28611, "tokens": 4016046080, "loss": 2.894494891166687, "ce_loss": 2.892111361026764, "z_loss": 23.835350513458252, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27806156873703003, "tokens_per_second": 2997695.7322642985, "eta_seconds": 8668.245271301268, "data_wait_seconds": 0.24686533212661743}
385
+ {"step": 3840, "total_steps": 28611, "tokens": 4026531840, "loss": 2.886387884616852, "ce_loss": 2.883584201335907, "z_loss": 28.037161350250244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29115572571754456, "tokens_per_second": 4494343.666812219, "eta_seconds": 5779.325752902031, "data_wait_seconds": 1.5020370483398438e-05}
386
+ {"step": 3850, "total_steps": 28611, "tokens": 4037017600, "loss": 2.8848310708999634, "ce_loss": 2.8820821046829224, "z_loss": 27.489654541015625, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29101279377937317, "tokens_per_second": 4583548.684999209, "eta_seconds": 5664.560828375816, "data_wait_seconds": 1.5914440155029297e-05}
387
+ {"step": 3860, "total_steps": 28611, "tokens": 4047503360, "loss": 2.869674026966095, "ce_loss": 2.8674304485321045, "z_loss": 22.43585205078125, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2526594400405884, "tokens_per_second": 3183629.189664911, "eta_seconds": 8152.112896895409, "data_wait_seconds": 1.4901161193847656e-05}
388
+ {"step": 3870, "total_steps": 28611, "tokens": 4057989120, "loss": 2.872779905796051, "ce_loss": 2.8699289560317993, "z_loss": 28.509647846221924, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27965477108955383, "tokens_per_second": 3186164.362399688, "eta_seconds": 8142.335380482674, "data_wait_seconds": 1.52587890625e-05}
389
+ {"step": 3880, "total_steps": 28611, "tokens": 4068474880, "loss": 2.865875005722046, "ce_loss": 2.863316535949707, "z_loss": 25.585248947143555, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2457529902458191, "tokens_per_second": 3212753.2471650564, "eta_seconds": 8071.685268354416, "data_wait_seconds": 0.22879523038864136}
390
+ {"step": 3890, "total_steps": 28611, "tokens": 4078960640, "loss": 2.8930917978286743, "ce_loss": 2.890199661254883, "z_loss": 28.921570777893066, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26920709013938904, "tokens_per_second": 4419445.466050617, "eta_seconds": 5865.407208919525, "data_wait_seconds": 1.4901161193847656e-05}
391
+ {"step": 3900, "total_steps": 28611, "tokens": 4089446400, "loss": 2.885398745536804, "ce_loss": 2.882913589477539, "z_loss": 24.851596355438232, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24747046828269958, "tokens_per_second": 4853435.6411267705, "eta_seconds": 5338.766896677018, "data_wait_seconds": 1.3470649719238281e-05}
392
+ {"step": 3910, "total_steps": 28611, "tokens": 4099932160, "loss": 2.8451263904571533, "ce_loss": 2.842821419239044, "z_loss": 23.049731254577637, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2502731680870056, "tokens_per_second": 4360713.570087764, "eta_seconds": 5939.595747280121, "data_wait_seconds": 1.4483928680419922e-05}
393
+ {"step": 3920, "total_steps": 28611, "tokens": 4110417920, "loss": 2.8715677857398987, "ce_loss": 2.8689383268356323, "z_loss": 26.294832706451416, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29730093479156494, "tokens_per_second": 4455105.27187352, "eta_seconds": 5811.398033499719, "data_wait_seconds": 1.5020370483398438e-05}
394
+ {"step": 3930, "total_steps": 28611, "tokens": 4120903680, "loss": 2.898460805416107, "ce_loss": 2.896242082118988, "z_loss": 22.187402725219727, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30773016810417175, "tokens_per_second": 4448885.257608267, "eta_seconds": 5817.166044402123, "data_wait_seconds": 1.4603137969970703e-05}
395
+ {"step": 3940, "total_steps": 28611, "tokens": 4131389440, "loss": 2.8709369897842407, "ce_loss": 2.8682071566581726, "z_loss": 27.29817295074463, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25586068630218506, "tokens_per_second": 4547432.393696362, "eta_seconds": 5688.796722269058, "data_wait_seconds": 1.895427703857422e-05}
396
+ {"step": 3950, "total_steps": 28611, "tokens": 4141875200, "loss": 2.8609384894371033, "ce_loss": 2.8583003878593445, "z_loss": 26.38067388534546, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24489066004753113, "tokens_per_second": 9531720.72636232, "eta_seconds": 2712.9343670845033, "data_wait_seconds": 1.6450881958007812e-05}
397
+ {"step": 3960, "total_steps": 28611, "tokens": 4152360960, "loss": 2.843279540538788, "ce_loss": 2.841322124004364, "z_loss": 19.573914051055908, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25654515624046326, "tokens_per_second": 4534019.02113647, "eta_seconds": 5701.001000547409, "data_wait_seconds": 1.4483928680419922e-05}
398
+ {"step": 3970, "total_steps": 28611, "tokens": 4162846720, "loss": 2.8423238396644592, "ce_loss": 2.839631736278534, "z_loss": 26.92048692703247, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2634884715080261, "tokens_per_second": 4139736.476372266, "eta_seconds": 6241.45072119236, "data_wait_seconds": 1.436471939086914e-05}
399
+ {"step": 3980, "total_steps": 28611, "tokens": 4173332480, "loss": 2.8560901880264282, "ce_loss": 2.853487193584442, "z_loss": 26.029609203338623, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2978611886501312, "tokens_per_second": 3146260.740206169, "eta_seconds": 8208.94311966896, "data_wait_seconds": 0.24235492944717407}
400
+ {"step": 3990, "total_steps": 28611, "tokens": 4183818240, "loss": 2.8676905632019043, "ce_loss": 2.8648945093154907, "z_loss": 27.96070623397827, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27539995312690735, "tokens_per_second": 4027867.7429122915, "eta_seconds": 6409.592207050324, "data_wait_seconds": 1.9788742065429688e-05}
401
+ {"step": 4000, "total_steps": 28611, "tokens": 4194304000, "loss": 2.8317272663116455, "ce_loss": 2.828980267047882, "z_loss": 27.470091342926025, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2498764544725418, "tokens_per_second": 3080479.5208791336, "eta_seconds": 8377.430773711205, "data_wait_seconds": 0.26825952529907227}
402
+ {"step": 4010, "total_steps": 28611, "tokens": 4204789760, "loss": 2.9265817999839783, "ce_loss": 2.923917829990387, "z_loss": 26.639820098876953, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2871566116809845, "tokens_per_second": 3760565.907969545, "eta_seconds": 6859.610709476471, "data_wait_seconds": 1.424551010131836e-05}
403
+ {"step": 4020, "total_steps": 28611, "tokens": 4215275520, "loss": 2.8929157853126526, "ce_loss": 2.8905717730522156, "z_loss": 23.439539432525635, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24263080954551697, "tokens_per_second": 2692054.288761331, "eta_seconds": 9578.384999012947, "data_wait_seconds": 1.6510486602783203e-05}
404
+ {"step": 4030, "total_steps": 28611, "tokens": 4225761280, "loss": 2.814459502696991, "ce_loss": 2.8120715618133545, "z_loss": 23.87947654724121, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28262457251548767, "tokens_per_second": 3476994.494266696, "eta_seconds": 7413.024869179726, "data_wait_seconds": 1.704692840576172e-05}
405
+ {"step": 4040, "total_steps": 28611, "tokens": 4236247040, "loss": 2.8808262944221497, "ce_loss": 2.878300189971924, "z_loss": 25.26109027862549, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2822110652923584, "tokens_per_second": 3072174.1717853895, "eta_seconds": 8386.42585196495, "data_wait_seconds": 1.3947486877441406e-05}
406
+ {"step": 4050, "total_steps": 28611, "tokens": 4246732800, "loss": 2.853496015071869, "ce_loss": 2.850788950920105, "z_loss": 27.070913791656494, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2624323070049286, "tokens_per_second": 2685643.7804793376, "eta_seconds": 9589.535039305687, "data_wait_seconds": 1.4722347259521484e-05}
407
+ {"step": 4060, "total_steps": 28611, "tokens": 4257218560, "loss": 2.771559774875641, "ce_loss": 2.769033968448639, "z_loss": 25.25845241546631, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2451464831829071, "tokens_per_second": 2799637.4584882385, "eta_seconds": 9195.329665970803, "data_wait_seconds": 2.008676528930664e-05}
408
+ {"step": 4070, "total_steps": 28611, "tokens": 4267704320, "loss": 2.836099088191986, "ce_loss": 2.8335224986076355, "z_loss": 25.76589870452881, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30325695872306824, "tokens_per_second": 4527478.985847197, "eta_seconds": 5683.7599238872535, "data_wait_seconds": 1.3828277587890625e-05}
409
+ {"step": 4080, "total_steps": 28611, "tokens": 4278190080, "loss": 2.798755168914795, "ce_loss": 2.7965914607048035, "z_loss": 21.637099266052246, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23870669305324554, "tokens_per_second": 2892307.4780377997, "eta_seconds": 8893.458960127831, "data_wait_seconds": 0.29030346870422363}
410
+ {"step": 4090, "total_steps": 28611, "tokens": 4288675840, "loss": 2.8402563333511353, "ce_loss": 2.837853193283081, "z_loss": 24.030916213989258, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2851092219352722, "tokens_per_second": 4389009.97845935, "eta_seconds": 5858.298846936226, "data_wait_seconds": 1.436471939086914e-05}
411
+ {"step": 4100, "total_steps": 28611, "tokens": 4299161600, "loss": 2.8005067110061646, "ce_loss": 2.7981993556022644, "z_loss": 23.074020385742188, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26443108916282654, "tokens_per_second": 3032916.527409366, "eta_seconds": 8474.234652924537, "data_wait_seconds": 1.710653305053711e-05}
412
+ {"step": 4110, "total_steps": 28611, "tokens": 4309647360, "loss": 2.829935073852539, "ce_loss": 2.827256143093109, "z_loss": 26.78932237625122, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24240489304065704, "tokens_per_second": 4077734.3604862634, "eta_seconds": 6300.351691603661, "data_wait_seconds": 1.33514404296875e-05}
413
+ {"step": 4120, "total_steps": 28611, "tokens": 4320133120, "loss": 2.8378302454948425, "ce_loss": 2.8351752758026123, "z_loss": 26.549213886260986, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27311551570892334, "tokens_per_second": 3137036.107919463, "eta_seconds": 8186.28601410389, "data_wait_seconds": 0.28760552406311035}
414
+ {"step": 4130, "total_steps": 28611, "tokens": 4330618880, "loss": 2.7907434701919556, "ce_loss": 2.788271903991699, "z_loss": 24.715639114379883, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27529406547546387, "tokens_per_second": 3951184.8857402722, "eta_seconds": 6496.833177471161, "data_wait_seconds": 1.3828277587890625e-05}
415
+ {"step": 4140, "total_steps": 28611, "tokens": 4341104640, "loss": 2.8120338320732117, "ce_loss": 2.809478521347046, "z_loss": 25.553118228912354, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.258984237909317, "tokens_per_second": 2878475.8135021343, "eta_seconds": 8914.336947226524, "data_wait_seconds": 1.52587890625e-05}
416
+ {"step": 4150, "total_steps": 28611, "tokens": 4351590400, "loss": 2.812987804412842, "ce_loss": 2.8108667135238647, "z_loss": 21.210986137390137, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2817680835723877, "tokens_per_second": 4356492.542635327, "eta_seconds": 5887.584400749207, "data_wait_seconds": 1.728534698486328e-05}
417
+ {"step": 4160, "total_steps": 28611, "tokens": 4362076160, "loss": 2.837258040904999, "ce_loss": 2.8348063826560974, "z_loss": 24.51671838760376, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2690242826938629, "tokens_per_second": 3033788.7309971154, "eta_seconds": 8451.060389947892, "data_wait_seconds": 0.2885240912437439}
418
+ {"step": 4170, "total_steps": 28611, "tokens": 4372561920, "loss": 2.8639367818832397, "ce_loss": 2.8620396852493286, "z_loss": 18.971240520477295, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2736714482307434, "tokens_per_second": 4545613.489171053, "eta_seconds": 5638.0169754982, "data_wait_seconds": 1.4424324035644531e-05}
419
+ {"step": 4180, "total_steps": 28611, "tokens": 4383047680, "loss": 2.828619122505188, "ce_loss": 2.826331377029419, "z_loss": 22.877708435058594, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2355462908744812, "tokens_per_second": 2775696.0347752282, "eta_seconds": 9229.310390996934, "data_wait_seconds": 1.4662742614746094e-05}
420
+ {"step": 4190, "total_steps": 28611, "tokens": 4393533440, "loss": 2.8075677156448364, "ce_loss": 2.805174946784973, "z_loss": 23.927815437316895, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2847018539905548, "tokens_per_second": 4555500.082661047, "eta_seconds": 5621.177484655381, "data_wait_seconds": 1.3887882232666016e-05}
421
+ {"step": 4200, "total_steps": 28611, "tokens": 4404019200, "loss": 2.8398322463035583, "ce_loss": 2.8372313380241394, "z_loss": 26.008440017700195, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3061939775943756, "tokens_per_second": 2866995.5079023093, "eta_seconds": 8928.088190388678, "data_wait_seconds": 1.5616416931152344e-05}
422
+ {"step": 4210, "total_steps": 28611, "tokens": 4414504960, "loss": 2.862689256668091, "ce_loss": 2.860727310180664, "z_loss": 19.619568824768066, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3027901351451874, "tokens_per_second": 3833908.3650706923, "eta_seconds": 6673.686624622344, "data_wait_seconds": 1.4662742614746094e-05}
423
+ {"step": 4220, "total_steps": 28611, "tokens": 4424990720, "loss": 2.8524940609931946, "ce_loss": 2.849484384059906, "z_loss": 30.096381664276123, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3021683096885681, "tokens_per_second": 2515380.526267175, "eta_seconds": 10167.772608923913, "data_wait_seconds": 1.5795230865478516e-05}
424
+ {"step": 4230, "total_steps": 28611, "tokens": 4435476480, "loss": 2.8206724524497986, "ce_loss": 2.818001389503479, "z_loss": 26.710964679718018, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25538572669029236, "tokens_per_second": 4019374.0472402144, "eta_seconds": 6360.525583219528, "data_wait_seconds": 1.2993812561035156e-05}
425
+ {"step": 4240, "total_steps": 28611, "tokens": 4445962240, "loss": 2.813224673271179, "ce_loss": 2.811019718647003, "z_loss": 22.04939889907837, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2826862335205078, "tokens_per_second": 3124130.23486349, "eta_seconds": 8179.827271866799, "data_wait_seconds": 1.4960765838623047e-05}
426
+ {"step": 4250, "total_steps": 28611, "tokens": 4456448000, "loss": 2.8534772396087646, "ce_loss": 2.85111927986145, "z_loss": 23.579538822174072, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2799414396286011, "tokens_per_second": 2238142.2898436897, "eta_seconds": 11413.197477173804, "data_wait_seconds": 1.7821788787841797e-05}
427
+ {"step": 4260, "total_steps": 28611, "tokens": 4466933760, "loss": 2.8407851457595825, "ce_loss": 2.8383723497390747, "z_loss": 24.1282320022583, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25657927989959717, "tokens_per_second": 2868578.060699983, "eta_seconds": 8901.23037815094, "data_wait_seconds": 1.9252300262451172e-05}
428
+ {"step": 4270, "total_steps": 28611, "tokens": 4477419520, "loss": 2.8449864983558655, "ce_loss": 2.8426221013069153, "z_loss": 23.643829345703125, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2728963792324066, "tokens_per_second": 3990832.353878371, "eta_seconds": 6395.505035734177, "data_wait_seconds": 1.5616416931152344e-05}
429
+ {"step": 4280, "total_steps": 28611, "tokens": 4487905280, "loss": 2.8036940693855286, "ce_loss": 2.8014734983444214, "z_loss": 22.2059006690979, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27203822135925293, "tokens_per_second": 4173912.026579889, "eta_seconds": 6112.467750525475, "data_wait_seconds": 1.3053417205810547e-05}
430
+ {"step": 4290, "total_steps": 28611, "tokens": 4498391040, "loss": 2.818882703781128, "ce_loss": 2.8169109225273132, "z_loss": 19.718111515045166, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2529017925262451, "tokens_per_second": 2806021.200423336, "eta_seconds": 9088.461944675444, "data_wait_seconds": 1.4543533325195312e-05}