Banaxi-Tech commited on
Commit
7f140a2
·
verified ·
1 Parent(s): c89d6bf

Save 95% checkpoint at 28,501,344,256 tokens

Browse files
checkpoint_metadata.json CHANGED
@@ -49,10 +49,10 @@
49
  "model_type": "bird_alpha",
50
  "output_attentions": false
51
  },
52
- "training_percent": 90,
53
- "step": 25750,
54
  "total_steps": 28611,
55
- "tokens_seen": 27000832000,
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": 8011.137375593185
72
  }
 
49
  "model_type": "bird_alpha",
50
  "output_attentions": false
51
  },
52
+ "training_percent": 95,
53
+ "step": 27181,
54
  "total_steps": 28611,
55
+ "tokens_seen": 28501344256,
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": 8466.48712849617
72
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:976d2a2f231229f3ee003600a15c33415293287286c85abb4b93b11c80ceb4cf
3
  size 13860292
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:535350a02b0abc9f365a527962456e13c600caa89c5930becec650f8b3f930ae
3
  size 13860292
training_metrics.jsonl CHANGED
@@ -2574,3 +2574,146 @@
2574
  {"step": 25730, "total_steps": 28611, "tokens": 26979860480, "loss": 2.7938910126686096, "ce_loss": 2.7930843830108643, "z_loss": 8.066987752914429, "muon_lr": 0.022698876891773304, "adamw_lr": 0.0022698876891773306, "grad_norm": 0.3298601508140564, "tokens_per_second": 4406372.351662466, "eta_seconds": 685.5860592126846, "data_wait_seconds": 1.3887882232666016e-05}
2575
  {"step": 25740, "total_steps": 28611, "tokens": 26990346240, "loss": 2.7701167464256287, "ce_loss": 2.7693235278129578, "z_loss": 7.932340621948242, "muon_lr": 0.022604397807203854, "adamw_lr": 0.002260439780720386, "grad_norm": 0.3571643829345703, "tokens_per_second": 3119894.630613135, "eta_seconds": 964.9241568803787, "data_wait_seconds": 0.18865185976028442}
2576
  {"step": 25750, "total_steps": 28611, "tokens": 27000832000, "loss": 2.769474685192108, "ce_loss": 2.768497943878174, "z_loss": 9.766818284988403, "muon_lr": 0.022509510921858156, "adamw_lr": 0.0022509510921858156, "grad_norm": 0.2806997299194336, "tokens_per_second": 4777044.827392793, "eta_seconds": 627.9982801914215, "data_wait_seconds": 0.2757408618927002}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2574
  {"step": 25730, "total_steps": 28611, "tokens": 26979860480, "loss": 2.7938910126686096, "ce_loss": 2.7930843830108643, "z_loss": 8.066987752914429, "muon_lr": 0.022698876891773304, "adamw_lr": 0.0022698876891773306, "grad_norm": 0.3298601508140564, "tokens_per_second": 4406372.351662466, "eta_seconds": 685.5860592126846, "data_wait_seconds": 1.3887882232666016e-05}
2575
  {"step": 25740, "total_steps": 28611, "tokens": 26990346240, "loss": 2.7701167464256287, "ce_loss": 2.7693235278129578, "z_loss": 7.932340621948242, "muon_lr": 0.022604397807203854, "adamw_lr": 0.002260439780720386, "grad_norm": 0.3571643829345703, "tokens_per_second": 3119894.630613135, "eta_seconds": 964.9241568803787, "data_wait_seconds": 0.18865185976028442}
2576
  {"step": 25750, "total_steps": 28611, "tokens": 27000832000, "loss": 2.769474685192108, "ce_loss": 2.768497943878174, "z_loss": 9.766818284988403, "muon_lr": 0.022509510921858156, "adamw_lr": 0.0022509510921858156, "grad_norm": 0.2806997299194336, "tokens_per_second": 4777044.827392793, "eta_seconds": 627.9982801914215, "data_wait_seconds": 0.2757408618927002}
2577
+ {"step": 25760, "total_steps": 28611, "tokens": 27011317760, "loss": 2.7429105639457703, "ce_loss": 2.7419955134391785, "z_loss": 9.150910377502441, "muon_lr": 0.0224142213242319, "adamw_lr": 0.0022414221324231903, "grad_norm": 0.2954011559486389, "tokens_per_second": 9732661.166681014, "eta_seconds": 307.1606136083603, "data_wait_seconds": 1.4841556549072266e-05}
2578
+ {"step": 25770, "total_steps": 28611, "tokens": 27021803520, "loss": 2.754826247692108, "ce_loss": 2.7538185119628906, "z_loss": 10.077370405197144, "muon_lr": 0.022318534124417023, "adamw_lr": 0.002231853412441702, "grad_norm": 0.32338324189186096, "tokens_per_second": 4400502.8718070425, "eta_seconds": 676.9690880298615, "data_wait_seconds": 1.436471939086914e-05}
2579
+ {"step": 25780, "total_steps": 28611, "tokens": 27032289280, "loss": 2.7224809527397156, "ce_loss": 2.721453368663788, "z_loss": 10.27631425857544, "muon_lr": 0.022222454453827645, "adamw_lr": 0.002222245445382765, "grad_norm": 0.3401945233345032, "tokens_per_second": 3967353.044955462, "eta_seconds": 748.2365754604339, "data_wait_seconds": 1.4185905456542969e-05}
2580
+ {"step": 25790, "total_steps": 28611, "tokens": 27042775040, "loss": 2.762700617313385, "ce_loss": 2.761747717857361, "z_loss": 9.528626918792725, "muon_lr": 0.022125987464924925, "adamw_lr": 0.002212598746492493, "grad_norm": 0.36300548911094666, "tokens_per_second": 4223881.920703074, "eta_seconds": 700.3114555597306, "data_wait_seconds": 1.9669532775878906e-05}
2581
+ {"step": 25800, "total_steps": 28611, "tokens": 27053260800, "loss": 2.75443297624588, "ce_loss": 2.7536250948905945, "z_loss": 8.078536629676819, "muon_lr": 0.022029138330940706, "adamw_lr": 0.0022029138330940706, "grad_norm": 0.29878026247024536, "tokens_per_second": 2990873.6550843734, "eta_seconds": 985.5137581586838, "data_wait_seconds": 0.2842711806297302}
2582
+ {"step": 25810, "total_steps": 28611, "tokens": 27063746560, "loss": 2.780574083328247, "ce_loss": 2.779647648334503, "z_loss": 9.26466679573059, "muon_lr": 0.02193191224560013, "adamw_lr": 0.002193191224560013, "grad_norm": 0.35022127628326416, "tokens_per_second": 4208607.00897329, "eta_seconds": 697.8701907157898, "data_wait_seconds": 1.4543533325195312e-05}
2583
+ {"step": 25820, "total_steps": 28611, "tokens": 27074232320, "loss": 2.7390986680984497, "ce_loss": 2.7383487820625305, "z_loss": 7.498966217041016, "muon_lr": 0.021834314422843077, "adamw_lr": 0.0021834314422843077, "grad_norm": 0.32352742552757263, "tokens_per_second": 2930717.2645435473, "eta_seconds": 998.5868140220642, "data_wait_seconds": 1.8477439880371094e-05}
2584
+ {"step": 25830, "total_steps": 28611, "tokens": 27084718080, "loss": 2.781489074230194, "ce_loss": 2.7805129885673523, "z_loss": 9.760838508605957, "muon_lr": 0.0217363500965446, "adamw_lr": 0.00217363500965446, "grad_norm": 0.3018725514411926, "tokens_per_second": 3111625.307606874, "eta_seconds": 937.1597052097321, "data_wait_seconds": 0.25350385904312134}
2585
+ {"step": 25840, "total_steps": 28611, "tokens": 27095203840, "loss": 2.7855513095855713, "ce_loss": 2.7846760749816895, "z_loss": 8.751397967338562, "muon_lr": 0.021638024520234212, "adamw_lr": 0.002163802452023421, "grad_norm": 0.36768388748168945, "tokens_per_second": 4277985.281964202, "eta_seconds": 679.1991801023484, "data_wait_seconds": 1.6450881958007812e-05}
2586
+ {"step": 25850, "total_steps": 28611, "tokens": 27105689600, "loss": 2.788604736328125, "ce_loss": 2.7878265380859375, "z_loss": 7.781946539878845, "muon_lr": 0.02153934296681415, "adamw_lr": 0.0021539342966814153, "grad_norm": 0.27277305722236633, "tokens_per_second": 3954840.9340807465, "eta_seconds": 732.0441919803619, "data_wait_seconds": 1.609325408935547e-05}
2587
+ {"step": 25860, "total_steps": 28611, "tokens": 27116175360, "loss": 2.7858205437660217, "ce_loss": 2.785024106502533, "z_loss": 7.964649200439453, "muon_lr": 0.021440310728276665, "adamw_lr": 0.0021440310728276664, "grad_norm": 0.2805539667606354, "tokens_per_second": 2829324.818882346, "eta_seconds": 1019.54804084301, "data_wait_seconds": 1.430511474609375e-05}
2588
+ {"step": 25870, "total_steps": 28611, "tokens": 27126661120, "loss": 2.7360675930976868, "ce_loss": 2.7352611422538757, "z_loss": 8.064285635948181, "muon_lr": 0.02134093311542014, "adamw_lr": 0.0021340933115420145, "grad_norm": 0.3200775682926178, "tokens_per_second": 2985874.6884851824, "eta_seconds": 962.5811917304992, "data_wait_seconds": 1.4007091522216797e-05}
2589
+ {"step": 25880, "total_steps": 28611, "tokens": 27137146880, "loss": 2.789152681827545, "ce_loss": 2.7883121371269226, "z_loss": 8.405495762825012, "muon_lr": 0.021241215457564376, "adamw_lr": 0.002124121545756438, "grad_norm": 0.3392302691936493, "tokens_per_second": 2770202.255444842, "eta_seconds": 1033.737175822258, "data_wait_seconds": 1.615285873413086e-05}
2590
+ {"step": 25890, "total_steps": 28611, "tokens": 27147632640, "loss": 2.781426250934601, "ce_loss": 2.7805135846138, "z_loss": 9.126629829406738, "muon_lr": 0.02114116310226473, "adamw_lr": 0.002114116310226473, "grad_norm": 0.33178555965423584, "tokens_per_second": 2980135.157053766, "eta_seconds": 957.3979519844055, "data_wait_seconds": 1.245737075805664e-05}
2591
+ {"step": 25900, "total_steps": 28611, "tokens": 27158118400, "loss": 2.7500420212745667, "ce_loss": 2.7492787837982178, "z_loss": 7.632302403450012, "muon_lr": 0.021040781415025366, "adamw_lr": 0.002104078141502537, "grad_norm": 0.26871681213378906, "tokens_per_second": 3035518.302093409, "eta_seconds": 936.4758347988128, "data_wait_seconds": 1.7523765563964844e-05}
2592
+ {"step": 25910, "total_steps": 28611, "tokens": 27168604160, "loss": 2.770920693874359, "ce_loss": 2.769979178905487, "z_loss": 9.415191888809204, "muon_lr": 0.02094007577901154, "adamw_lr": 0.0020940075779011537, "grad_norm": 0.3875736594200134, "tokens_per_second": 4254262.64396822, "eta_seconds": 665.7331747055053, "data_wait_seconds": 1.2814998626708984e-05}
2593
+ {"step": 25920, "total_steps": 28611, "tokens": 27179089920, "loss": 2.7909462451934814, "ce_loss": 2.7899730801582336, "z_loss": 9.731908082962036, "muon_lr": 0.020839051594760873, "adamw_lr": 0.0020839051594760873, "grad_norm": 0.3397677540779114, "tokens_per_second": 2891964.953742859, "eta_seconds": 975.7096165180205, "data_wait_seconds": 1.4781951904296875e-05}
2594
+ {"step": 25930, "total_steps": 28611, "tokens": 27189575680, "loss": 2.7523191571235657, "ce_loss": 2.7514378428459167, "z_loss": 8.813210487365723, "muon_lr": 0.020737714279893783, "adamw_lr": 0.002073771427989378, "grad_norm": 0.3446667194366455, "tokens_per_second": 3157978.925706854, "eta_seconds": 890.1998151779175, "data_wait_seconds": 0.24786972999572754}
2595
+ {"step": 25940, "total_steps": 28611, "tokens": 27200061440, "loss": 2.7676095962524414, "ce_loss": 2.7668434977531433, "z_loss": 7.660900235176086, "muon_lr": 0.020636069268822922, "adamw_lr": 0.0020636069268822922, "grad_norm": 0.3314830958843231, "tokens_per_second": 3821506.5223024865, "eta_seconds": 732.8906753540039, "data_wait_seconds": 1.3947486877441406e-05}
2596
+ {"step": 25950, "total_steps": 28611, "tokens": 27210547200, "loss": 2.7419410347938538, "ce_loss": 2.7409366369247437, "z_loss": 10.043427467346191, "muon_lr": 0.020534122012461756, "adamw_lr": 0.0020534122012461757, "grad_norm": 0.28972163796424866, "tokens_per_second": 2721889.5228438256, "eta_seconds": 1025.1190258026124, "data_wait_seconds": 0.2949783205986023}
2597
+ {"step": 25960, "total_steps": 28611, "tokens": 27221032960, "loss": 2.7713447213172913, "ce_loss": 2.770559251308441, "z_loss": 7.854948163032532, "muon_lr": 0.02043187797793225, "adamw_lr": 0.0020431877977932253, "grad_norm": 0.31812477111816406, "tokens_per_second": 3038366.1245683776, "eta_seconds": 914.8913797855378, "data_wait_seconds": 1.7821788787841797e-05}
2598
+ {"step": 25970, "total_steps": 28611, "tokens": 27231518720, "loss": 2.7619125843048096, "ce_loss": 2.761078953742981, "z_loss": 8.336507558822632, "muon_lr": 0.020329342648271683, "adamw_lr": 0.002032934264827168, "grad_norm": 0.2936194837093353, "tokens_per_second": 3116745.556136189, "eta_seconds": 888.5195041179658, "data_wait_seconds": 1.3887882232666016e-05}
2599
+ {"step": 25980, "total_steps": 28611, "tokens": 27242004480, "loss": 2.7600425481796265, "ce_loss": 2.759084463119507, "z_loss": 9.580877780914307, "muon_lr": 0.020226521522138604, "adamw_lr": 0.0020226521522138607, "grad_norm": 0.31179100275039673, "tokens_per_second": 2310674.2253867066, "eta_seconds": 1193.9387325525286, "data_wait_seconds": 0.25851744413375854}
2600
+ {"step": 25990, "total_steps": 28611, "tokens": 27252490240, "loss": 2.754921078681946, "ce_loss": 2.7542075514793396, "z_loss": 7.135005354881287, "muon_lr": 0.020123420113517962, "adamw_lr": 0.002012342011351796, "grad_norm": 0.33211976289749146, "tokens_per_second": 2989740.1680218927, "eta_seconds": 919.2496810913085, "data_wait_seconds": 0.25927668809890747}
2601
+ {"step": 26000, "total_steps": 28611, "tokens": 27262976000, "loss": 2.7977447509765625, "ce_loss": 2.7968913912773132, "z_loss": 8.533602952957153, "muon_lr": 0.0200200439514254, "adamw_lr": 0.00200200439514254, "grad_norm": 0.2878177762031555, "tokens_per_second": 3908946.256027228, "eta_seconds": 700.4015293836594, "data_wait_seconds": 1.5914440155029297e-05}
2602
+ {"step": 26010, "total_steps": 28611, "tokens": 27273461760, "loss": 2.7709718346595764, "ce_loss": 2.770164132118225, "z_loss": 8.07757031917572, "muon_lr": 0.01991639857961077, "adamw_lr": 0.001991639857961077, "grad_norm": 0.25894439220428467, "tokens_per_second": 3048701.775655216, "eta_seconds": 894.5926419496536, "data_wait_seconds": 1.531839370727539e-05}
2603
+ {"step": 26020, "total_steps": 28611, "tokens": 27283947520, "loss": 2.7584837675094604, "ce_loss": 2.7575544714927673, "z_loss": 9.293219566345215, "muon_lr": 0.01981248955626081, "adamw_lr": 0.0019812489556260814, "grad_norm": 0.31002160906791687, "tokens_per_second": 2512539.0029873075, "eta_seconds": 1081.3206930398942, "data_wait_seconds": 1.800060272216797e-05}
2604
+ {"step": 26030, "total_steps": 28611, "tokens": 27294433280, "loss": 2.7644932866096497, "ce_loss": 2.7637252807617188, "z_loss": 7.680354833602905, "muon_lr": 0.019708322453701098, "adamw_lr": 0.0019708322453701097, "grad_norm": 0.3131362497806549, "tokens_per_second": 4197209.210919652, "eta_seconds": 644.8033729076386, "data_wait_seconds": 1.4185905456542969e-05}
2605
+ {"step": 26040, "total_steps": 28611, "tokens": 27304919040, "loss": 2.7284770607948303, "ce_loss": 2.727694034576416, "z_loss": 7.8305593729019165, "muon_lr": 0.01960390285809721, "adamw_lr": 0.001960390285809721, "grad_norm": 0.35161903500556946, "tokens_per_second": 2868550.37018388, "eta_seconds": 939.808805179596, "data_wait_seconds": 1.329183578491211e-05}
2606
+ {"step": 26050, "total_steps": 28611, "tokens": 27315404800, "loss": 2.7571101784706116, "ce_loss": 2.7563359141349792, "z_loss": 7.742649555206299, "muon_lr": 0.019499236369155157, "adamw_lr": 0.001949923636915516, "grad_norm": 0.31332850456237793, "tokens_per_second": 3113367.653474377, "eta_seconds": 862.5396788597106, "data_wait_seconds": 1.5914440155029297e-05}
2607
+ {"step": 26060, "total_steps": 28611, "tokens": 27325890560, "loss": 2.7490808367729187, "ce_loss": 2.7481027841567993, "z_loss": 9.77999496459961, "muon_lr": 0.019394328599821105, "adamw_lr": 0.0019394328599821105, "grad_norm": 0.3464483618736267, "tokens_per_second": 2462112.1233913796, "eta_seconds": 1086.4319908857346, "data_wait_seconds": 0.26116943359375}
2608
+ {"step": 26070, "total_steps": 28611, "tokens": 27336376320, "loss": 2.770158112049103, "ce_loss": 2.7694015502929688, "z_loss": 7.564716100692749, "muon_lr": 0.019289185175980335, "adamw_lr": 0.0019289185175980336, "grad_norm": 0.32806187868118286, "tokens_per_second": 3778235.968274837, "eta_seconds": 705.2051905632019, "data_wait_seconds": 1.2874603271484375e-05}
2609
+ {"step": 26080, "total_steps": 28611, "tokens": 27346862080, "loss": 2.763249397277832, "ce_loss": 2.7625352144241333, "z_loss": 7.1423832178115845, "muon_lr": 0.019183811736155583, "adamw_lr": 0.0019183811736155586, "grad_norm": 0.3079216182231903, "tokens_per_second": 2762614.935691678, "eta_seconds": 960.6644131660461, "data_wait_seconds": 1.8358230590820312e-05}
2610
+ {"step": 26090, "total_steps": 28611, "tokens": 27357347840, "loss": 2.753617823123932, "ce_loss": 2.752638578414917, "z_loss": 9.792866230010986, "muon_lr": 0.01907821393120463, "adamw_lr": 0.0019078213931204631, "grad_norm": 0.3359750807285309, "tokens_per_second": 3275645.3288987214, "eta_seconds": 807.0043703079224, "data_wait_seconds": 1.710653305053711e-05}
2611
+ {"step": 26100, "total_steps": 28611, "tokens": 27367833600, "loss": 2.7575230598449707, "ce_loss": 2.756687104701996, "z_loss": 8.359660387039185, "muon_lr": 0.018972397424017285, "adamw_lr": 0.0018972397424017284, "grad_norm": 0.3062448799610138, "tokens_per_second": 2939311.8788807797, "eta_seconds": 895.7791634559632, "data_wait_seconds": 1.8656253814697266e-05}
2612
+ {"step": 26110, "total_steps": 28611, "tokens": 27378319360, "loss": 2.7472593188285828, "ce_loss": 2.746386170387268, "z_loss": 8.731846332550049, "muon_lr": 0.018866367889211685, "adamw_lr": 0.0018866367889211685, "grad_norm": 0.31606006622314453, "tokens_per_second": 3141500.389221925, "eta_seconds": 834.7885567665099, "data_wait_seconds": 0.26721835136413574}
2613
+ {"step": 26120, "total_steps": 28611, "tokens": 27388805120, "loss": 2.7950974106788635, "ce_loss": 2.794289290904999, "z_loss": 8.081138014793396, "muon_lr": 0.01876013101283001, "adamw_lr": 0.001876013101283001, "grad_norm": 0.3174426853656769, "tokens_per_second": 3077499.9147041924, "eta_seconds": 848.7417996406555, "data_wait_seconds": 1.823902130126953e-05}
2614
+ {"step": 26130, "total_steps": 28611, "tokens": 27399290880, "loss": 2.7722023725509644, "ce_loss": 2.771245777606964, "z_loss": 9.566459655761719, "muon_lr": 0.018653692492033514, "adamw_lr": 0.0018653692492033513, "grad_norm": 0.3023165762424469, "tokens_per_second": 3260731.4304524777, "eta_seconds": 797.8323610782624, "data_wait_seconds": 1.5676021575927734e-05}
2615
+ {"step": 26140, "total_steps": 28611, "tokens": 27409776640, "loss": 2.7803955674171448, "ce_loss": 2.7794684767723083, "z_loss": 9.270641088485718, "muon_lr": 0.01854705803479706, "adamw_lr": 0.001854705803479706, "grad_norm": 0.3644014596939087, "tokens_per_second": 2344733.182050772, "eta_seconds": 1105.0431306362152, "data_wait_seconds": 1.811981201171875e-05}
2616
+ {"step": 26150, "total_steps": 28611, "tokens": 27420262400, "loss": 2.769804835319519, "ce_loss": 2.768913447856903, "z_loss": 8.914056777954102, "muon_lr": 0.01844023335960296, "adamw_lr": 0.001844023335960296, "grad_norm": 0.29282984137535095, "tokens_per_second": 2831250.0204255986, "eta_seconds": 911.4509553670882, "data_wait_seconds": 1.8477439880371094e-05}
2617
+ {"step": 26160, "total_steps": 28611, "tokens": 27430748160, "loss": 2.749649941921234, "ce_loss": 2.7489187717437744, "z_loss": 7.3116419315338135, "muon_lr": 0.018333224195134345, "adamw_lr": 0.0018333224195134347, "grad_norm": 0.30877798795700073, "tokens_per_second": 4138782.8089917353, "eta_seconds": 620.9699553251266, "data_wait_seconds": 1.8775463104248047e-05}
2618
+ {"step": 26170, "total_steps": 28611, "tokens": 27441233920, "loss": 2.7164220213890076, "ce_loss": 2.7156007885932922, "z_loss": 8.21247124671936, "muon_lr": 0.01822603627996796, "adamw_lr": 0.0018226036279967962, "grad_norm": 0.30561503767967224, "tokens_per_second": 2963177.5593646606, "eta_seconds": 863.7936690330505, "data_wait_seconds": 0.2716623544692993}
2619
+ {"step": 26180, "total_steps": 28611, "tokens": 27451719680, "loss": 2.7330090403556824, "ce_loss": 2.732032537460327, "z_loss": 9.764594554901123, "muon_lr": 0.018118675362266388, "adamw_lr": 0.0018118675362266388, "grad_norm": 0.28972482681274414, "tokens_per_second": 4652209.264170656, "eta_seconds": 547.9306951284409, "data_wait_seconds": 1.3709068298339844e-05}
2620
+ {"step": 26190, "total_steps": 28611, "tokens": 27462205440, "loss": 2.7572240829467773, "ce_loss": 2.7564744353294373, "z_loss": 7.496246695518494, "muon_lr": 0.018011147199469835, "adamw_lr": 0.0018011147199469835, "grad_norm": 0.26354873180389404, "tokens_per_second": 2909878.682964991, "eta_seconds": 872.4083621978759, "data_wait_seconds": 1.71661376953125e-05}
2621
+ {"step": 26200, "total_steps": 28611, "tokens": 27472691200, "loss": 2.7643584609031677, "ce_loss": 2.763618588447571, "z_loss": 7.398406386375427, "muon_lr": 0.017903457557987317, "adamw_lr": 0.001790345755798732, "grad_norm": 0.3559613823890686, "tokens_per_second": 2820247.309029018, "eta_seconds": 896.4166822910308, "data_wait_seconds": 1.6808509826660156e-05}
2622
+ {"step": 26210, "total_steps": 28611, "tokens": 27483176960, "loss": 2.7306548953056335, "ce_loss": 2.729721784591675, "z_loss": 9.331246852874756, "muon_lr": 0.017795612212887515, "adamw_lr": 0.0017795612212887517, "grad_norm": 0.29300716519355774, "tokens_per_second": 4468165.431216074, "eta_seconds": 563.4596602916718, "data_wait_seconds": 1.7344951629638672e-05}
2623
+ {"step": 26220, "total_steps": 28611, "tokens": 27493662720, "loss": 2.662670135498047, "ce_loss": 2.66154682636261, "z_loss": 11.232792019844055, "muon_lr": 0.017687616947588972, "adamw_lr": 0.0017687616947588973, "grad_norm": 0.6154298782348633, "tokens_per_second": 4557071.443311221, "eta_seconds": 550.1658789396287, "data_wait_seconds": 1.4543533325195312e-05}
2624
+ {"step": 26230, "total_steps": 28611, "tokens": 27504148480, "loss": 2.7605772018432617, "ce_loss": 2.759667992591858, "z_loss": 9.092389106750488, "muon_lr": 0.017579477553550025, "adamw_lr": 0.0017579477553550027, "grad_norm": 0.2834141254425049, "tokens_per_second": 2997388.2590885567, "eta_seconds": 832.9449641466141, "data_wait_seconds": 1.4007091522216797e-05}
2625
+ {"step": 26240, "total_steps": 28611, "tokens": 27514634240, "loss": 2.7353795766830444, "ce_loss": 2.7345675826072693, "z_loss": 8.120115160942078, "muon_lr": 0.017471199829958192, "adamw_lr": 0.0017471199829958192, "grad_norm": 0.2982260286808014, "tokens_per_second": 2927229.375483748, "eta_seconds": 849.3265737295151, "data_wait_seconds": 1.704692840576172e-05}
2626
+ {"step": 26250, "total_steps": 28611, "tokens": 27525120000, "loss": 2.7542139887809753, "ce_loss": 2.7533202171325684, "z_loss": 8.937641382217407, "muon_lr": 0.017362789583419178, "adamw_lr": 0.0017362789583419177, "grad_norm": 0.33217599987983704, "tokens_per_second": 4171434.974678576, "eta_seconds": 593.4859229564667, "data_wait_seconds": 1.5914440155029297e-05}
2627
+ {"step": 26260, "total_steps": 28611, "tokens": 27535605760, "loss": 2.755538582801819, "ce_loss": 2.7545723915100098, "z_loss": 9.661813497543335, "muon_lr": 0.017254252627645496, "adamw_lr": 0.0017254252627645498, "grad_norm": 0.3055459260940552, "tokens_per_second": 2936747.4237652337, "eta_seconds": 839.4328215122223, "data_wait_seconds": 1.7583370208740234e-05}
2628
+ {"step": 26270, "total_steps": 28611, "tokens": 27546091520, "loss": 2.7378026843070984, "ce_loss": 2.7369441390037537, "z_loss": 8.58471155166626, "muon_lr": 0.017145594783144695, "adamw_lr": 0.0017145594783144697, "grad_norm": 0.30548152327537537, "tokens_per_second": 4647916.651100502, "eta_seconds": 528.1326237678528, "data_wait_seconds": 1.2636184692382812e-05}
2629
+ {"step": 26280, "total_steps": 28611, "tokens": 27556577280, "loss": 2.75322026014328, "ce_loss": 2.7523263692855835, "z_loss": 8.93905258178711, "muon_lr": 0.017036821876907218, "adamw_lr": 0.001703682187690722, "grad_norm": 0.2832395136356354, "tokens_per_second": 3004033.689563245, "eta_seconds": 813.6495487689972, "data_wait_seconds": 1.3828277587890625e-05}
2630
+ {"step": 26290, "total_steps": 28611, "tokens": 27567063040, "loss": 2.7518829703330994, "ce_loss": 2.7510328888893127, "z_loss": 8.500756978988647, "muon_lr": 0.016927939742093917, "adamw_lr": 0.001692793974209392, "grad_norm": 0.318345844745636, "tokens_per_second": 4327224.131904361, "eta_seconds": 562.4263550519943, "data_wait_seconds": 1.4007091522216797e-05}
2631
+ {"step": 26300, "total_steps": 28611, "tokens": 27577548800, "loss": 2.767248213291168, "ce_loss": 2.766485035419464, "z_loss": 7.6312642097473145, "muon_lr": 0.016818954217723243, "adamw_lr": 0.0016818954217723244, "grad_norm": 0.25301840901374817, "tokens_per_second": 3062565.9387246785, "eta_seconds": 791.2512528657912, "data_wait_seconds": 1.4781951904296875e-05}
2632
+ {"step": 26310, "total_steps": 28611, "tokens": 27588034560, "loss": 2.7903785705566406, "ce_loss": 2.7895711064338684, "z_loss": 8.074960708618164, "muon_lr": 0.016709871148358108, "adamw_lr": 0.0016709871148358109, "grad_norm": 0.3508443236351013, "tokens_per_second": 4674402.317985875, "eta_seconds": 516.1672470331192, "data_wait_seconds": 1.7940998077392578e-05}
2633
+ {"step": 26320, "total_steps": 28611, "tokens": 27598520320, "loss": 2.751608431339264, "ce_loss": 2.750706911087036, "z_loss": 9.015471935272217, "muon_lr": 0.01660069638379248, "adamw_lr": 0.0016600696383792478, "grad_norm": 0.3034832775592804, "tokens_per_second": 3092599.168582, "eta_seconds": 776.7859606266022, "data_wait_seconds": 1.4960765838623047e-05}
2634
+ {"step": 26330, "total_steps": 28611, "tokens": 27609006080, "loss": 2.7422419786453247, "ce_loss": 2.741479456424713, "z_loss": 7.6254953145980835, "muon_lr": 0.016491435778737646, "adamw_lr": 0.0016491435778737647, "grad_norm": 0.28473734855651855, "tokens_per_second": 4446041.529414024, "eta_seconds": 537.9621040821075, "data_wait_seconds": 1.3947486877441406e-05}
2635
+ {"step": 26340, "total_steps": 28611, "tokens": 27619491840, "loss": 2.7205586433410645, "ce_loss": 2.7196022868156433, "z_loss": 9.563462018966675, "muon_lr": 0.016382095192508276, "adamw_lr": 0.0016382095192508277, "grad_norm": 0.2838290333747864, "tokens_per_second": 3060205.8951995634, "eta_seconds": 778.1555155277252, "data_wait_seconds": 1.4543533325195312e-05}
2636
+ {"step": 26350, "total_steps": 28611, "tokens": 27629977600, "loss": 2.711856782436371, "ce_loss": 2.7109304666519165, "z_loss": 9.263899803161621, "muon_lr": 0.01627268048870817, "adamw_lr": 0.001627268048870817, "grad_norm": 0.27889150381088257, "tokens_per_second": 3026704.3001071517, "eta_seconds": 783.3042480945587, "data_wait_seconds": 1.8477439880371094e-05}
2637
+ {"step": 26360, "total_steps": 28611, "tokens": 27640463360, "loss": 2.757643699645996, "ce_loss": 2.756847858428955, "z_loss": 7.957672953605652, "muon_lr": 0.016163197534915838, "adamw_lr": 0.001616319753491584, "grad_norm": 0.2607639729976654, "tokens_per_second": 4490205.280360226, "eta_seconds": 525.66518201828, "data_wait_seconds": 1.2218952178955078e-05}
2638
+ {"step": 26370, "total_steps": 28611, "tokens": 27650949120, "loss": 2.76812481880188, "ce_loss": 2.7673123478889465, "z_loss": 8.124794483184814, "muon_lr": 0.01605365220236983, "adamw_lr": 0.0016053652202369833, "grad_norm": 0.2988812029361725, "tokens_per_second": 4878585.202074231, "eta_seconds": 481.66808996200564, "data_wait_seconds": 1.4603137969970703e-05}
2639
+ {"step": 26380, "total_steps": 28611, "tokens": 27661434880, "loss": 2.7427104711532593, "ce_loss": 2.7419933676719666, "z_loss": 7.171622157096863, "muon_lr": 0.01594405036565388, "adamw_lr": 0.001594405036565388, "grad_norm": 0.2932640314102173, "tokens_per_second": 3132264.1237793877, "eta_seconds": 746.86327958107, "data_wait_seconds": 1.5020370483398438e-05}
2640
+ {"step": 26390, "total_steps": 28611, "tokens": 27671920640, "loss": 2.7156805396080017, "ce_loss": 2.714719533920288, "z_loss": 9.609955787658691, "muon_lr": 0.01583439790238186, "adamw_lr": 0.0015834397902381862, "grad_norm": 0.3402598202228546, "tokens_per_second": 2994826.937947815, "eta_seconds": 777.6366862773896, "data_wait_seconds": 1.430511474609375e-05}
2641
+ {"step": 26400, "total_steps": 28611, "tokens": 27682406400, "loss": 2.706893026828766, "ce_loss": 2.7060712575912476, "z_loss": 8.217705607414246, "muon_lr": 0.015724700692882607, "adamw_lr": 0.0015724700692882605, "grad_norm": 0.2649121880531311, "tokens_per_second": 3109046.4314024225, "eta_seconds": 745.6953722476959, "data_wait_seconds": 0.2231278419494629}
2642
+ {"step": 26410, "total_steps": 28611, "tokens": 27692892160, "loss": 2.7247959971427917, "ce_loss": 2.7239701747894287, "z_loss": 8.258752346038818, "muon_lr": 0.015614964619884545, "adamw_lr": 0.0015614964619884545, "grad_norm": 0.2528570294380188, "tokens_per_second": 4782085.777516193, "eta_seconds": 482.6169758081436, "data_wait_seconds": 1.7702579498291016e-05}
2643
+ {"step": 26420, "total_steps": 28611, "tokens": 27703377920, "loss": 2.713837683200836, "ce_loss": 2.712916314601898, "z_loss": 9.213483095169067, "muon_lr": 0.015505195568200235, "adamw_lr": 0.0015505195568200235, "grad_norm": 0.2900152802467346, "tokens_per_second": 4558482.7679710165, "eta_seconds": 503.9900626897812, "data_wait_seconds": 1.2099742889404297e-05}
2644
+ {"step": 26430, "total_steps": 28611, "tokens": 27713863680, "loss": 2.755844473838806, "ce_loss": 2.75497829914093, "z_loss": 8.661197423934937, "muon_lr": 0.015395399424410799, "adamw_lr": 0.00153953994244108, "grad_norm": 0.2913181185722351, "tokens_per_second": 3199603.619357994, "eta_seconds": 714.758616399765, "data_wait_seconds": 1.4126300811767578e-05}
2645
+ {"step": 26440, "total_steps": 28611, "tokens": 27724349440, "loss": 2.7233792543411255, "ce_loss": 2.7225053906440735, "z_loss": 8.739257097244263, "muon_lr": 0.015285582076550197, "adamw_lr": 0.0015285582076550199, "grad_norm": 0.26472872495651245, "tokens_per_second": 2890515.8767719995, "eta_seconds": 787.5613188266755, "data_wait_seconds": 1.7464160919189453e-05}
2646
+ {"step": 26450, "total_steps": 28611, "tokens": 27734835200, "loss": 2.7105036973953247, "ce_loss": 2.7096492052078247, "z_loss": 8.544463396072388, "muon_lr": 0.015175749413789538, "adamw_lr": 0.001517574941378954, "grad_norm": 0.2457660734653473, "tokens_per_second": 4245922.332015837, "eta_seconds": 533.6820975065232, "data_wait_seconds": 1.7464160919189453e-05}
2647
+ {"step": 26460, "total_steps": 28611, "tokens": 27745320960, "loss": 2.717690885066986, "ce_loss": 2.716879367828369, "z_loss": 8.115259170532227, "muon_lr": 0.015065907326121179, "adamw_lr": 0.0015065907326121178, "grad_norm": 0.27877920866012573, "tokens_per_second": 3020286.18598613, "eta_seconds": 746.7792245864869, "data_wait_seconds": 0.2814367413520813}
2648
+ {"step": 26470, "total_steps": 28611, "tokens": 27755806720, "loss": 2.712958514690399, "ce_loss": 2.712203085422516, "z_loss": 7.554829955101013, "muon_lr": 0.014956061704042947, "adamw_lr": 0.0014956061704042946, "grad_norm": 0.28974154591560364, "tokens_per_second": 4604696.543031837, "eta_seconds": 487.545964217186, "data_wait_seconds": 1.4185905456542969e-05}
2649
+ {"step": 26480, "total_steps": 28611, "tokens": 27766292480, "loss": 2.6761586666107178, "ce_loss": 2.6754003763198853, "z_loss": 7.583141088485718, "muon_lr": 0.014846218438242176, "adamw_lr": 0.0014846218438242177, "grad_norm": 0.2912432849407196, "tokens_per_second": 3128266.7644333346, "eta_seconds": 714.2982438087463, "data_wait_seconds": 1.4007091522216797e-05}
2650
+ {"step": 26490, "total_steps": 28611, "tokens": 27776778240, "loss": 2.73828125, "ce_loss": 2.7374194264411926, "z_loss": 8.619094848632812, "muon_lr": 0.01473638341927986, "adamw_lr": 0.001473638341927986, "grad_norm": 0.2751820385456085, "tokens_per_second": 3209630.21269596, "eta_seconds": 692.923965883255, "data_wait_seconds": 0.5186869502067566}
2651
+ {"step": 26500, "total_steps": 28611, "tokens": 27787264000, "loss": 2.70274156332016, "ce_loss": 2.701888084411621, "z_loss": 8.534332752227783, "muon_lr": 0.014626562537274729, "adamw_lr": 0.001462656253727473, "grad_norm": 0.26670995354652405, "tokens_per_second": 4353874.279600315, "eta_seconds": 508.40786707401276, "data_wait_seconds": 1.3232231140136719e-05}
2652
+ {"step": 26510, "total_steps": 28611, "tokens": 27797749760, "loss": 2.7248154878616333, "ce_loss": 2.7239603400230408, "z_loss": 8.551278591156006, "muon_lr": 0.014516761681587408, "adamw_lr": 0.0014516761681587409, "grad_norm": 0.29612937569618225, "tokens_per_second": 2899174.278745078, "eta_seconds": 759.8915981531144, "data_wait_seconds": 0.294177770614624}
2653
+ {"step": 26520, "total_steps": 28611, "tokens": 27808235520, "loss": 2.7393670678138733, "ce_loss": 2.7384772896766663, "z_loss": 8.8975989818573, "muon_lr": 0.014406986740504548, "adamw_lr": 0.0014406986740504548, "grad_norm": 0.34156861901283264, "tokens_per_second": 4405341.757053681, "eta_seconds": 497.7076778411865, "data_wait_seconds": 1.3947486877441406e-05}
2654
+ {"step": 26530, "total_steps": 28611, "tokens": 27818721280, "loss": 2.746934950351715, "ce_loss": 2.7460761070251465, "z_loss": 8.588245630264282, "muon_lr": 0.01429724360092309, "adamw_lr": 0.001429724360092309, "grad_norm": 0.245132178068161, "tokens_per_second": 2807991.896560751, "eta_seconds": 777.0986300468444, "data_wait_seconds": 1.4901161193847656e-05}
2655
+ {"step": 26540, "total_steps": 28611, "tokens": 27829207040, "loss": 2.751780331134796, "ce_loss": 2.7509268522262573, "z_loss": 8.534478187561035, "muon_lr": 0.014187538148034569, "adamw_lr": 0.001418753814803457, "grad_norm": 0.27487945556640625, "tokens_per_second": 4525693.236263488, "eta_seconds": 479.8382883310318, "data_wait_seconds": 1.5020370483398438e-05}
2656
+ {"step": 26550, "total_steps": 28611, "tokens": 27839692800, "loss": 2.7097268104553223, "ce_loss": 2.7088281512260437, "z_loss": 8.987117528915405, "muon_lr": 0.014077876265009464, "adamw_lr": 0.0014077876265009466, "grad_norm": 0.3019894063472748, "tokens_per_second": 3064906.771235701, "eta_seconds": 705.1161086797714, "data_wait_seconds": 1.6927719116210938e-05}
2657
+ {"step": 26560, "total_steps": 28611, "tokens": 27850178560, "loss": 2.7687771320343018, "ce_loss": 2.7679494619369507, "z_loss": 8.27637505531311, "muon_lr": 0.013968263832681772, "adamw_lr": 0.0013968263832681773, "grad_norm": 0.24289247393608093, "tokens_per_second": 4549912.112625583, "eta_seconds": 472.67492706775664, "data_wait_seconds": 1.2755393981933594e-05}
2658
+ {"step": 26570, "total_steps": 28611, "tokens": 27860664320, "loss": 2.7835633754730225, "ce_loss": 2.782877564430237, "z_loss": 6.858276724815369, "muon_lr": 0.01385870672923357, "adamw_lr": 0.0013858706729233572, "grad_norm": 0.2965216338634491, "tokens_per_second": 3137117.1105274274, "eta_seconds": 682.2007405519486, "data_wait_seconds": 1.4603137969970703e-05}
2659
+ {"step": 26580, "total_steps": 28611, "tokens": 27871150080, "loss": 2.7519531846046448, "ce_loss": 2.7511415481567383, "z_loss": 8.115761756896973, "muon_lr": 0.013749210829879843, "adamw_lr": 0.0013749210829879844, "grad_norm": 0.2736496925354004, "tokens_per_second": 3043585.3077241196, "eta_seconds": 699.720113182068, "data_wait_seconds": 0.5694562792778015}
2660
+ {"step": 26590, "total_steps": 28611, "tokens": 27881635840, "loss": 2.738549828529358, "ce_loss": 2.7378260493278503, "z_loss": 7.237285852432251, "muon_lr": 0.013639782006553358, "adamw_lr": 0.001363978200655336, "grad_norm": 0.282884418964386, "tokens_per_second": 4708399.234099926, "eta_seconds": 450.08334906101226, "data_wait_seconds": 1.3887882232666016e-05}
2661
+ {"step": 26600, "total_steps": 28611, "tokens": 27892121600, "loss": 2.7340269684791565, "ce_loss": 2.733354687690735, "z_loss": 6.7226762771606445, "muon_lr": 0.013530426127589814, "adamw_lr": 0.0013530426127589814, "grad_norm": 0.3182922303676605, "tokens_per_second": 3064355.6024467763, "eta_seconds": 688.1336925506591, "data_wait_seconds": 0.2761712670326233}
2662
+ {"step": 26610, "total_steps": 28611, "tokens": 27902607360, "loss": 2.705107569694519, "ce_loss": 2.7042235732078552, "z_loss": 8.840360760688782, "muon_lr": 0.013421149057413127, "adamw_lr": 0.0013421149057413126, "grad_norm": 0.2357780933380127, "tokens_per_second": 4472668.057712062, "eta_seconds": 469.11609556674955, "data_wait_seconds": 1.3530254364013672e-05}
2663
+ {"step": 26620, "total_steps": 28611, "tokens": 27913093120, "loss": 2.7476431727409363, "ce_loss": 2.7468525767326355, "z_loss": 7.906053185462952, "muon_lr": 0.013311956656220918, "adamw_lr": 0.0013311956656220918, "grad_norm": 0.28364676237106323, "tokens_per_second": 3151182.9680175507, "eta_seconds": 662.5178027391433, "data_wait_seconds": 1.806020736694336e-05}
2664
+ {"step": 26630, "total_steps": 28611, "tokens": 27923578880, "loss": 2.7110171914100647, "ce_loss": 2.7102543115615845, "z_loss": 7.628970503807068, "muon_lr": 0.013202854779670295, "adamw_lr": 0.0013202854779670295, "grad_norm": 0.2973152697086334, "tokens_per_second": 4677543.287401791, "eta_seconds": 444.08547999858854, "data_wait_seconds": 1.5497207641601562e-05}
2665
+ {"step": 26640, "total_steps": 28611, "tokens": 27934064640, "loss": 2.740604877471924, "ce_loss": 2.7397149801254272, "z_loss": 8.898489952087402, "muon_lr": 0.01309384927856378, "adamw_lr": 0.0013093849278563781, "grad_norm": 0.23207299411296844, "tokens_per_second": 2921562.270703464, "eta_seconds": 707.410318350792, "data_wait_seconds": 1.3709068298339844e-05}
2666
+ {"step": 26650, "total_steps": 28611, "tokens": 27944550400, "loss": 2.734389066696167, "ce_loss": 2.7336992621421814, "z_loss": 6.897962689399719, "muon_lr": 0.012984945998535601, "adamw_lr": 0.0012984945998535601, "grad_norm": 0.2920228838920593, "tokens_per_second": 2892711.536570676, "eta_seconds": 710.8408529520035, "data_wait_seconds": 1.4543533325195312e-05}
2667
+ {"step": 26660, "total_steps": 28611, "tokens": 27955036160, "loss": 2.7158798575401306, "ce_loss": 2.7151190042495728, "z_loss": 7.608613610267639, "muon_lr": 0.012876150779738138, "adamw_lr": 0.001287615077973814, "grad_norm": 0.28383511304855347, "tokens_per_second": 3096412.530090081, "eta_seconds": 660.6909628868102, "data_wait_seconds": 0.2589154839515686}
2668
+ {"step": 26670, "total_steps": 28611, "tokens": 27965521920, "loss": 2.72643905878067, "ce_loss": 2.725532829761505, "z_loss": 9.062609195709229, "muon_lr": 0.012767469456528813, "adamw_lr": 0.0012767469456528813, "grad_norm": 0.3277507424354553, "tokens_per_second": 4080146.014528061, "eta_seconds": 498.82675981521606, "data_wait_seconds": 1.436471939086914e-05}
2669
+ {"step": 26680, "total_steps": 28611, "tokens": 27976007680, "loss": 2.7317329049110413, "ce_loss": 2.731046497821808, "z_loss": 6.863732576370239, "muon_lr": 0.01265890785715716, "adamw_lr": 0.001265890785715716, "grad_norm": 0.2368118017911911, "tokens_per_second": 3222094.7138600624, "eta_seconds": 628.4111535549164, "data_wait_seconds": 1.341104507446289e-05}
2670
+ {"step": 26690, "total_steps": 28611, "tokens": 27986493440, "loss": 2.7479217648506165, "ce_loss": 2.747135579586029, "z_loss": 7.8618305921554565, "muon_lr": 0.01255047180345228, "adamw_lr": 0.001255047180345228, "grad_norm": 0.25538742542266846, "tokens_per_second": 2895387.180274919, "eta_seconds": 695.697801566124, "data_wait_seconds": 1.4841556549072266e-05}
2671
+ {"step": 26700, "total_steps": 28611, "tokens": 27996979200, "loss": 2.7890418767929077, "ce_loss": 2.7882535457611084, "z_loss": 7.883543014526367, "muon_lr": 0.012442167110510643, "adamw_lr": 0.0012442167110510644, "grad_norm": 0.26896271109580994, "tokens_per_second": 2966021.6151118483, "eta_seconds": 675.5947852134705, "data_wait_seconds": 1.8894672393798828e-05}
2672
+ {"step": 26710, "total_steps": 28611, "tokens": 28007464960, "loss": 2.7366027235984802, "ce_loss": 2.7357633113861084, "z_loss": 8.394238710403442, "muon_lr": 0.012333999586384254, "adamw_lr": 0.0012333999586384254, "grad_norm": 0.2619398832321167, "tokens_per_second": 4708710.262892715, "eta_seconds": 423.3309897422791, "data_wait_seconds": 1.436471939086914e-05}
2673
+ {"step": 26720, "total_steps": 28611, "tokens": 28017950720, "loss": 2.733469247817993, "ce_loss": 2.7327135801315308, "z_loss": 7.556904554367065, "muon_lr": 0.012225975031769157, "adamw_lr": 0.001222597503176916, "grad_norm": 0.27079877257347107, "tokens_per_second": 2947054.5018421574, "eta_seconds": 672.8267884969712, "data_wait_seconds": 1.8656253814697266e-05}
2674
+ {"step": 26730, "total_steps": 28611, "tokens": 28028436480, "loss": 2.7205129265785217, "ce_loss": 2.719704508781433, "z_loss": 8.084001302719116, "muon_lr": 0.012118099239694372, "adamw_lr": 0.0012118099239694373, "grad_norm": 0.25558412075042725, "tokens_per_second": 4316961.879513474, "eta_seconds": 456.8887822151184, "data_wait_seconds": 1.4901161193847656e-05}
2675
+ {"step": 26740, "total_steps": 28611, "tokens": 28038922240, "loss": 2.726345419883728, "ce_loss": 2.7255101799964905, "z_loss": 8.352516412734985, "muon_lr": 0.012010377995211253, "adamw_lr": 0.0012010377995211254, "grad_norm": 0.2648658752441406, "tokens_per_second": 2923666.7865485894, "eta_seconds": 671.036010336876, "data_wait_seconds": 1.5795230865478516e-05}
2676
+ {"step": 26750, "total_steps": 28611, "tokens": 28049408000, "loss": 2.7507498264312744, "ce_loss": 2.7500945329666138, "z_loss": 6.553381562232971, "muon_lr": 0.01190281707508322, "adamw_lr": 0.0011902817075083222, "grad_norm": 0.2509194612503052, "tokens_per_second": 3190897.2247424982, "eta_seconds": 611.5521116971969, "data_wait_seconds": 2.008676528930664e-05}
2677
+ {"step": 26760, "total_steps": 28611, "tokens": 28059893760, "loss": 2.747353971004486, "ce_loss": 2.7465673685073853, "z_loss": 7.8659751415252686, "muon_lr": 0.011795422247475986, "adamw_lr": 0.0011795422247475986, "grad_norm": 0.23906563222408295, "tokens_per_second": 4105040.6165149505, "eta_seconds": 472.81241705417636, "data_wait_seconds": 1.895427703857422e-05}
2678
+ {"step": 26770, "total_steps": 28611, "tokens": 28070379520, "loss": 2.7265522480010986, "ce_loss": 2.7258167266845703, "z_loss": 7.354736089706421, "muon_lr": 0.011688199271648226, "adamw_lr": 0.0011688199271648228, "grad_norm": 0.260437548160553, "tokens_per_second": 2920008.747387921, "eta_seconds": 661.1036414623261, "data_wait_seconds": 2.104043960571289e-05}
2679
+ {"step": 26780, "total_steps": 28611, "tokens": 28080865280, "loss": 2.7390953302383423, "ce_loss": 2.73837673664093, "z_loss": 7.186221361160278, "muon_lr": 0.01158115389764274, "adamw_lr": 0.001158115389764274, "grad_norm": 0.25544482469558716, "tokens_per_second": 4058150.439622779, "eta_seconds": 473.1078072547912, "data_wait_seconds": 1.5735626220703125e-05}
2680
+ {"step": 26790, "total_steps": 28611, "tokens": 28091351040, "loss": 2.735229969024658, "ce_loss": 2.7344279885292053, "z_loss": 8.019548535346985, "muon_lr": 0.011474291865978066, "adamw_lr": 0.0011474291865978066, "grad_norm": 0.27557703852653503, "tokens_per_second": 2424109.5879432084, "eta_seconds": 787.6941312789917, "data_wait_seconds": 0.28820526599884033}
2681
+ {"step": 26800, "total_steps": 28611, "tokens": 28101836800, "loss": 2.7219801545143127, "ce_loss": 2.721261203289032, "z_loss": 7.189438104629517, "muon_lr": 0.011367618907340647, "adamw_lr": 0.0011367618907340649, "grad_norm": 0.2540473937988281, "tokens_per_second": 4581213.982933857, "eta_seconds": 414.51264731884, "data_wait_seconds": 1.817941665649414e-05}
2682
+ {"step": 26810, "total_steps": 28611, "tokens": 28112322560, "loss": 2.7298964858055115, "ce_loss": 2.72912859916687, "z_loss": 7.679023265838623, "muon_lr": 0.01126114074227752, "adamw_lr": 0.001126114074227752, "grad_norm": 0.30210208892822266, "tokens_per_second": 4285488.164464344, "eta_seconds": 440.66983819007874, "data_wait_seconds": 1.3768672943115234e-05}
2683
+ {"step": 26820, "total_steps": 28611, "tokens": 28122808320, "loss": 2.722040116786957, "ce_loss": 2.7212986946105957, "z_loss": 7.413971543312073, "muon_lr": 0.011154863080889544, "adamw_lr": 0.0011154863080889545, "grad_norm": 0.24859920144081116, "tokens_per_second": 2973602.5659958012, "eta_seconds": 631.5570337057113, "data_wait_seconds": 1.3947486877441406e-05}
2684
+ {"step": 26830, "total_steps": 28611, "tokens": 28133294080, "loss": 2.7414810061454773, "ce_loss": 2.7406254410743713, "z_loss": 8.554717063903809, "muon_lr": 0.01104879162252515, "adamw_lr": 0.001104879162252515, "grad_norm": 0.28923532366752625, "tokens_per_second": 2708658.7545499504, "eta_seconds": 689.4607350826263, "data_wait_seconds": 1.9788742065429688e-05}
2685
+ {"step": 26840, "total_steps": 28611, "tokens": 28143779840, "loss": 2.7122743129730225, "ce_loss": 2.711532413959503, "z_loss": 7.418817043304443, "muon_lr": 0.010942932055474744, "adamw_lr": 0.0010942932055474745, "grad_norm": 0.2280685305595398, "tokens_per_second": 4285430.5390543, "eta_seconds": 433.3352458000183, "data_wait_seconds": 1.609325408935547e-05}
2686
+ {"step": 26850, "total_steps": 28611, "tokens": 28154265600, "loss": 2.7244072556495667, "ce_loss": 2.7236446142196655, "z_loss": 7.626753091812134, "muon_lr": 0.01083729005666564, "adamw_lr": 0.001083729005666564, "grad_norm": 0.23312044143676758, "tokens_per_second": 2281446.7950000637, "eta_seconds": 809.3733941316603, "data_wait_seconds": 1.9609928131103516e-05}
2687
+ {"step": 26860, "total_steps": 28611, "tokens": 28164751360, "loss": 2.72835510969162, "ce_loss": 2.7276230454444885, "z_loss": 7.320340871810913, "muon_lr": 0.010731871291357636, "adamw_lr": 0.0010731871291357637, "grad_norm": 0.27055981755256653, "tokens_per_second": 2821575.361058386, "eta_seconds": 650.7203746318818, "data_wait_seconds": 1.9252300262451172e-05}
2688
+ {"step": 26870, "total_steps": 28611, "tokens": 28175237120, "loss": 2.7652674317359924, "ce_loss": 2.76455557346344, "z_loss": 7.11878514289856, "muon_lr": 0.010626681412839173, "adamw_lr": 0.0010626681412839174, "grad_norm": 0.20989364385604858, "tokens_per_second": 5082305.214859172, "eta_seconds": 359.20133459568024, "data_wait_seconds": 1.4483928680419922e-05}
2689
+ {"step": 26880, "total_steps": 28611, "tokens": 28185722880, "loss": 2.7332836389541626, "ce_loss": 2.732613682746887, "z_loss": 6.699488997459412, "muon_lr": 0.010521726062124204, "adamw_lr": 0.0010521726062124204, "grad_norm": 0.30569443106651306, "tokens_per_second": 2822434.7375689684, "eta_seconds": 643.091948890686, "data_wait_seconds": 1.4841556549072266e-05}
2690
+ {"step": 26890, "total_steps": 28611, "tokens": 28196208640, "loss": 2.7694104313850403, "ce_loss": 2.768674314022064, "z_loss": 7.3610183000564575, "muon_lr": 0.010417010867649677, "adamw_lr": 0.0010417010867649677, "grad_norm": 0.2834499478340149, "tokens_per_second": 4161673.841113364, "eta_seconds": 433.6234325170517, "data_wait_seconds": 1.3589859008789062e-05}
2691
+ {"step": 26900, "total_steps": 28611, "tokens": 28206694400, "loss": 2.68073970079422, "ce_loss": 2.6800726652145386, "z_loss": 6.670593738555908, "muon_lr": 0.010312541444973657, "adamw_lr": 0.0010312541444973656, "grad_norm": 0.2406187206506729, "tokens_per_second": 3031801.9502207926, "eta_seconds": 591.7647542476655, "data_wait_seconds": 1.7821788787841797e-05}
2692
+ {"step": 26910, "total_steps": 28611, "tokens": 28217180160, "loss": 2.7281158566474915, "ce_loss": 2.7274099588394165, "z_loss": 7.058866262435913, "muon_lr": 0.010208323396474248, "adamw_lr": 0.001020832339647425, "grad_norm": 0.2629479765892029, "tokens_per_second": 4667775.802990104, "eta_seconds": 382.1151339054108, "data_wait_seconds": 1.6570091247558594e-05}
2693
+ {"step": 26920, "total_steps": 28611, "tokens": 28227665920, "loss": 2.75563383102417, "ce_loss": 2.7548131942749023, "z_loss": 8.206387281417847, "muon_lr": 0.010104362311049109, "adamw_lr": 0.0010104362311049108, "grad_norm": 0.26285427808761597, "tokens_per_second": 2947163.5131774275, "eta_seconds": 601.6435830831527, "data_wait_seconds": 1.5020370483398438e-05}
2694
+ {"step": 26930, "total_steps": 28611, "tokens": 28238151680, "loss": 2.6939884424209595, "ce_loss": 2.693148612976074, "z_loss": 8.398895859718323, "muon_lr": 0.01000066376381573, "adamw_lr": 0.001000066376381573, "grad_norm": 0.22646521031856537, "tokens_per_second": 3092404.550514471, "eta_seconds": 569.9953635454177, "data_wait_seconds": 0.22855925559997559}
2695
+ {"step": 26940, "total_steps": 28611, "tokens": 28248637440, "loss": 2.743200123310089, "ce_loss": 2.742470920085907, "z_loss": 7.292309880256653, "muon_lr": 0.009897233315812503, "adamw_lr": 0.0009897233315812504, "grad_norm": 0.2474866360425949, "tokens_per_second": 4616181.3931273995, "eta_seconds": 379.57141342163084, "data_wait_seconds": 1.8715858459472656e-05}
2696
+ {"step": 26950, "total_steps": 28611, "tokens": 28259123200, "loss": 2.7113893032073975, "ce_loss": 2.7106894850730896, "z_loss": 6.997965335845947, "muon_lr": 0.009794076513700456, "adamw_lr": 0.0009794076513700456, "grad_norm": 0.21782684326171875, "tokens_per_second": 2912942.302299544, "eta_seconds": 597.912541770935, "data_wait_seconds": 1.806020736694336e-05}
2697
+ {"step": 26960, "total_steps": 28611, "tokens": 28269608960, "loss": 2.7084310054779053, "ce_loss": 2.707748830318451, "z_loss": 6.8216331005096436, "muon_lr": 0.009691198889465825, "adamw_lr": 0.0009691198889465825, "grad_norm": 0.2332177460193634, "tokens_per_second": 4070746.673397489, "eta_seconds": 425.27799317836764, "data_wait_seconds": 1.3768672943115234e-05}
2698
+ {"step": 26970, "total_steps": 28611, "tokens": 28280094720, "loss": 2.716913402080536, "ce_loss": 2.716126322746277, "z_loss": 7.870720982551575, "muon_lr": 0.009588605960123371, "adamw_lr": 0.0009588605960123372, "grad_norm": 0.2579997479915619, "tokens_per_second": 2861870.3479100857, "eta_seconds": 601.2547763586044, "data_wait_seconds": 1.4483928680419922e-05}
2699
+ {"step": 26980, "total_steps": 28611, "tokens": 28290580480, "loss": 2.6692161560058594, "ce_loss": 2.668434739112854, "z_loss": 7.813466668128967, "muon_lr": 0.009486303227420551, "adamw_lr": 0.0009486303227420551, "grad_norm": 0.27125898003578186, "tokens_per_second": 3047796.8993353406, "eta_seconds": 561.1356374740601, "data_wait_seconds": 1.8477439880371094e-05}
2700
+ {"step": 26990, "total_steps": 28611, "tokens": 28301066240, "loss": 2.7423706650733948, "ce_loss": 2.7415621876716614, "z_loss": 8.085206151008606, "muon_lr": 0.009384296177542445, "adamw_lr": 0.0009384296177542445, "grad_norm": 0.22609661519527435, "tokens_per_second": 2823687.438122703, "eta_seconds": 601.9581604719161, "data_wait_seconds": 1.4722347259521484e-05}
2701
+ {"step": 27000, "total_steps": 28611, "tokens": 28311552000, "loss": 2.739210367202759, "ce_loss": 2.7383923530578613, "z_loss": 8.179716348648071, "muon_lr": 0.009282590280817559, "adamw_lr": 0.0009282590280817559, "grad_norm": 0.21757158637046814, "tokens_per_second": 2911347.822025722, "eta_seconds": 580.2315763235092, "data_wait_seconds": 1.4424324035644531e-05}
2702
+ {"step": 27010, "total_steps": 28611, "tokens": 28322037760, "loss": 2.709804117679596, "ce_loss": 2.7089868783950806, "z_loss": 8.172728300094604, "muon_lr": 0.009181190991424473, "adamw_lr": 0.0009181190991424472, "grad_norm": 0.2372680902481079, "tokens_per_second": 3082860.1455475925, "eta_seconds": 544.5495730400085, "data_wait_seconds": 1.6033649444580078e-05}
2703
+ {"step": 27020, "total_steps": 28611, "tokens": 28332523520, "loss": 2.676618814468384, "ce_loss": 2.6758031845092773, "z_loss": 8.156497478485107, "muon_lr": 0.009080103747099348, "adamw_lr": 0.000908010374709935, "grad_norm": 0.25199466943740845, "tokens_per_second": 2816007.0128615503, "eta_seconds": 592.4290700912476, "data_wait_seconds": 1.4662742614746094e-05}
2704
+ {"step": 27030, "total_steps": 28611, "tokens": 28343009280, "loss": 2.724219262599945, "ce_loss": 2.7234233021736145, "z_loss": 7.959499001502991, "muon_lr": 0.00897933396884432, "adamw_lr": 0.000897933396884432, "grad_norm": 0.2499600648880005, "tokens_per_second": 4486101.048310554, "eta_seconds": 369.541086602211, "data_wait_seconds": 1.2576580047607422e-05}
2705
+ {"step": 27040, "total_steps": 28611, "tokens": 28353495040, "loss": 2.683620512485504, "ce_loss": 2.6828457713127136, "z_loss": 7.747640371322632, "muon_lr": 0.008878887060636763, "adamw_lr": 0.0008878887060636764, "grad_norm": 0.24437925219535828, "tokens_per_second": 2995562.4944176055, "eta_seconds": 549.9177196502686, "data_wait_seconds": 1.4543533325195312e-05}
2706
+ {"step": 27050, "total_steps": 28611, "tokens": 28363980800, "loss": 2.7031782269477844, "ce_loss": 2.702421009540558, "z_loss": 7.57221519947052, "muon_lr": 0.008778768409139532, "adamw_lr": 0.0008778768409139534, "grad_norm": 0.250838041305542, "tokens_per_second": 2984493.8327586874, "eta_seconds": 548.4437990903855, "data_wait_seconds": 0.25301945209503174}
2707
+ {"step": 27060, "total_steps": 28611, "tokens": 28374466560, "loss": 2.691399335861206, "ce_loss": 2.6905882954597473, "z_loss": 8.110005378723145, "muon_lr": 0.008678983383412065, "adamw_lr": 0.0008678983383412066, "grad_norm": 0.25178536772727966, "tokens_per_second": 4246019.072035491, "eta_seconds": 383.02733652591706, "data_wait_seconds": 1.3172626495361328e-05}
2708
+ {"step": 27070, "total_steps": 28611, "tokens": 28384952320, "loss": 2.7135215997695923, "ce_loss": 2.7126927971839905, "z_loss": 8.288297057151794, "muon_lr": 0.008579537334622455, "adamw_lr": 0.0008579537334622455, "grad_norm": 0.227172389626503, "tokens_per_second": 2904290.38733896, "eta_seconds": 556.3684757709502, "data_wait_seconds": 1.4543533325195312e-05}
2709
+ {"step": 27080, "total_steps": 28611, "tokens": 28395438080, "loss": 2.7127432227134705, "ce_loss": 2.711906909942627, "z_loss": 8.36304771900177, "muon_lr": 0.008480435595760493, "adamw_lr": 0.0008480435595760494, "grad_norm": 0.22858798503875732, "tokens_per_second": 4177266.6880219807, "eta_seconds": 384.31107609272004, "data_wait_seconds": 1.5020370483398438e-05}
2710
+ {"step": 27090, "total_steps": 28611, "tokens": 28405923840, "loss": 2.7099029421806335, "ce_loss": 2.7091866731643677, "z_loss": 7.163249135017395, "muon_lr": 0.008381683481351676, "adamw_lr": 0.0008381683481351676, "grad_norm": 0.2219535857439041, "tokens_per_second": 2350623.5479342286, "eta_seconds": 678.4940521001815, "data_wait_seconds": 0.26488953828811646}
2711
+ {"step": 27100, "total_steps": 28611, "tokens": 28416409600, "loss": 2.706772565841675, "ce_loss": 2.706094980239868, "z_loss": 6.775467395782471, "muon_lr": 0.008283286287172219, "adamw_lr": 0.0008283286287172219, "grad_norm": 0.23664437234401703, "tokens_per_second": 2807672.9936340293, "eta_seconds": 564.3101385354996, "data_wait_seconds": 1.823902130126953e-05}
2712
+ {"step": 27110, "total_steps": 28611, "tokens": 28426895360, "loss": 2.701060473918915, "ce_loss": 2.7003342509269714, "z_loss": 7.262277722358704, "muon_lr": 0.008185249289965005, "adamw_lr": 0.0008185249289965005, "grad_norm": 0.2281758189201355, "tokens_per_second": 3113639.864635266, "eta_seconds": 505.48960201740266, "data_wait_seconds": 1.6868114471435547e-05}
2713
+ {"step": 27120, "total_steps": 28611, "tokens": 28437381120, "loss": 2.7357290387153625, "ce_loss": 2.7349299788475037, "z_loss": 7.990521192550659, "muon_lr": 0.008087577747156675, "adamw_lr": 0.0008087577747156676, "grad_norm": 0.22721153497695923, "tokens_per_second": 3117744.4435753417, "eta_seconds": 501.46086194515226, "data_wait_seconds": 1.3530254364013672e-05}
2714
+ {"step": 27130, "total_steps": 28611, "tokens": 28447866880, "loss": 2.6973389387130737, "ce_loss": 2.6965657472610474, "z_loss": 7.731416463851929, "muon_lr": 0.007990276896575653, "adamw_lr": 0.0007990276896575654, "grad_norm": 0.2391595095396042, "tokens_per_second": 2839470.6987567795, "eta_seconds": 546.9121610164642, "data_wait_seconds": 1.71661376953125e-05}
2715
+ {"step": 27140, "total_steps": 28611, "tokens": 28458352640, "loss": 2.7084041833877563, "ce_loss": 2.7076372504234314, "z_loss": 7.669335246086121, "muon_lr": 0.007893351956171264, "adamw_lr": 0.0007893351956171264, "grad_norm": 0.2269950658082962, "tokens_per_second": 3037422.6899180594, "eta_seconds": 507.81713757514956, "data_wait_seconds": 0.26462966203689575}
2716
+ {"step": 27150, "total_steps": 28611, "tokens": 28468838400, "loss": 2.6900553107261658, "ce_loss": 2.6893237233161926, "z_loss": 7.3156304359436035, "muon_lr": 0.007796808123733908, "adamw_lr": 0.0007796808123733908, "grad_norm": 0.2262657880783081, "tokens_per_second": 3004106.532704149, "eta_seconds": 509.9584583044052, "data_wait_seconds": 1.800060272216797e-05}
2717
+ {"step": 27160, "total_steps": 28611, "tokens": 28479324160, "loss": 2.659835457801819, "ce_loss": 2.659124255180359, "z_loss": 7.111611604690552, "muon_lr": 0.007700650576616323, "adamw_lr": 0.0007700650576616324, "grad_norm": 0.21075376868247986, "tokens_per_second": 2997634.436896441, "eta_seconds": 507.56148157119753, "data_wait_seconds": 0.2711854577064514}
2718
+ {"step": 27170, "total_steps": 28611, "tokens": 28489809920, "loss": 2.692146420478821, "ce_loss": 2.6913750767707825, "z_loss": 7.712975263595581, "muon_lr": 0.007604884471455939, "adamw_lr": 0.0007604884471455939, "grad_norm": 0.2185799479484558, "tokens_per_second": 2926439.3621653984, "eta_seconds": 516.3264394044876, "data_wait_seconds": 1.7404556274414062e-05}
2719
+ {"step": 27180, "total_steps": 28611, "tokens": 28500295680, "loss": 2.6888388991355896, "ce_loss": 2.688002347946167, "z_loss": 8.365736246109009, "muon_lr": 0.007509514943898339, "adamw_lr": 0.000750951494389834, "grad_norm": 0.21340161561965942, "tokens_per_second": 3133006.0341457003, "eta_seconds": 478.93691861629486, "data_wait_seconds": 1.3709068298339844e-05}