Banaxi-Tech commited on
Commit
f3f9eb0
·
verified ·
1 Parent(s): fe2dc42

Save 35% checkpoint at 10,500,440,064 tokens

Browse files
checkpoint_metadata.json CHANGED
@@ -50,10 +50,10 @@
50
  "model_type": "tiger_alpha",
51
  "output_attentions": false
52
  },
53
- "training_percent": 30,
54
- "step": 8584,
55
  "total_steps": 28611,
56
- "tokens_seen": 9000976384,
57
  "target_tokens": 30000000000,
58
  "trained_token_target": 30000807936,
59
  "tokens_per_step": 1048576,
@@ -69,5 +69,5 @@
69
  "dataset_id": "HuggingFaceFW/fineweb-edu",
70
  "dataset_config": "sample-100BT",
71
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
72
- "elapsed_seconds": 2749.0057089328766
73
  }
 
50
  "model_type": "tiger_alpha",
51
  "output_attentions": false
52
  },
53
+ "training_percent": 35,
54
+ "step": 10014,
55
  "total_steps": 28611,
56
+ "tokens_seen": 10500440064,
57
  "target_tokens": 30000000000,
58
  "trained_token_target": 30000807936,
59
  "tokens_per_step": 1048576,
 
69
  "dataset_id": "HuggingFaceFW/fineweb-edu",
70
  "dataset_config": "sample-100BT",
71
  "dataset_revision": "87f09149ef4734204d70ed1d046ddc9ca3f2b8f9",
72
+ "elapsed_seconds": 3217.9480624198914
73
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0fc0c29f1762d82c3d2d03617e7dbff624dbf1e151f255e8ce2dece6f7823d91
3
  size 12811532
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2eb87d43b96f8e99a7baaf929af5e73897d0e4f464a113490f822d26b4df17a
3
  size 12811532
training_metrics.jsonl CHANGED
@@ -857,3 +857,146 @@
857
  {"step": 8560, "total_steps": 28611, "tokens": 8975810560, "loss": 2.782153606414795, "ce_loss": 2.7804641723632812, "z_loss": 16.894001960754395, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3149415850639343, "tokens_per_second": 3481001.130327209, "eta_seconds": 6039.928339242935, "data_wait_seconds": 1.4960765838623047e-05}
858
  {"step": 8570, "total_steps": 28611, "tokens": 8986296320, "loss": 2.7826881408691406, "ce_loss": 2.781241238117218, "z_loss": 14.469220876693726, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3197265565395355, "tokens_per_second": 2460431.147735356, "eta_seconds": 8540.987475037575, "data_wait_seconds": 0.24492710828781128}
859
  {"step": 8580, "total_steps": 28611, "tokens": 8996782080, "loss": 2.7980841994285583, "ce_loss": 2.79668265581131, "z_loss": 14.01570749282837, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27325671911239624, "tokens_per_second": 2777287.2285189396, "eta_seconds": 7562.784878826142, "data_wait_seconds": 1.7821788787841797e-05}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
857
  {"step": 8560, "total_steps": 28611, "tokens": 8975810560, "loss": 2.782153606414795, "ce_loss": 2.7804641723632812, "z_loss": 16.894001960754395, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3149415850639343, "tokens_per_second": 3481001.130327209, "eta_seconds": 6039.928339242935, "data_wait_seconds": 1.4960765838623047e-05}
858
  {"step": 8570, "total_steps": 28611, "tokens": 8986296320, "loss": 2.7826881408691406, "ce_loss": 2.781241238117218, "z_loss": 14.469220876693726, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3197265565395355, "tokens_per_second": 2460431.147735356, "eta_seconds": 8540.987475037575, "data_wait_seconds": 0.24492710828781128}
859
  {"step": 8580, "total_steps": 28611, "tokens": 8996782080, "loss": 2.7980841994285583, "ce_loss": 2.79668265581131, "z_loss": 14.01570749282837, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27325671911239624, "tokens_per_second": 2777287.2285189396, "eta_seconds": 7562.784878826142, "data_wait_seconds": 1.7821788787841797e-05}
860
+ {"step": 8590, "total_steps": 28611, "tokens": 9007267840, "loss": 2.7730417251586914, "ce_loss": 2.7715484499931335, "z_loss": 14.933393955230713, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31156668066978455, "tokens_per_second": 3353288.637687399, "eta_seconds": 6260.582480152448, "data_wait_seconds": 1.5735626220703125e-05}
861
+ {"step": 8600, "total_steps": 28611, "tokens": 9017753600, "loss": 2.8064348101615906, "ce_loss": 2.8051427006721497, "z_loss": 12.921017169952393, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2533218264579773, "tokens_per_second": 4186100.476488936, "eta_seconds": 5012.5539159536365, "data_wait_seconds": 1.7404556274414062e-05}
862
+ {"step": 8610, "total_steps": 28611, "tokens": 9028239360, "loss": 2.8101813793182373, "ce_loss": 2.808467388153076, "z_loss": 17.13991117477417, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29599273204803467, "tokens_per_second": 2885313.9278259277, "eta_seconds": 7268.730232000351, "data_wait_seconds": 1.4007091522216797e-05}
863
+ {"step": 8620, "total_steps": 28611, "tokens": 9038725120, "loss": 2.8033064007759094, "ce_loss": 2.802065908908844, "z_loss": 12.40504002571106, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3133027255535126, "tokens_per_second": 2823682.543305959, "eta_seconds": 7423.668381452561, "data_wait_seconds": 1.71661376953125e-05}
864
+ {"step": 8630, "total_steps": 28611, "tokens": 9049210880, "loss": 2.7885285019874573, "ce_loss": 2.7870089411735535, "z_loss": 15.195698499679565, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31841880083084106, "tokens_per_second": 4177882.5456543784, "eta_seconds": 5014.88417327404, "data_wait_seconds": 1.4126300811767578e-05}
865
+ {"step": 8640, "total_steps": 28611, "tokens": 9059696640, "loss": 2.798761546611786, "ce_loss": 2.7974089980125427, "z_loss": 13.525370597839355, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3340100646018982, "tokens_per_second": 2750816.04872408, "eta_seconds": 7612.69053440094, "data_wait_seconds": 0.2883222699165344}
866
+ {"step": 8650, "total_steps": 28611, "tokens": 9070182400, "loss": 2.847305953502655, "ce_loss": 2.8458325266838074, "z_loss": 14.734018802642822, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30729150772094727, "tokens_per_second": 4355231.105155438, "eta_seconds": 4805.858754825592, "data_wait_seconds": 1.3887882232666016e-05}
867
+ {"step": 8660, "total_steps": 28611, "tokens": 9080668160, "loss": 2.7697701454162598, "ce_loss": 2.7680587768554688, "z_loss": 17.11372661590576, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.38701844215393066, "tokens_per_second": 4114806.2890790193, "eta_seconds": 5084.112909889221, "data_wait_seconds": 1.6927719116210938e-05}
868
+ {"step": 8670, "total_steps": 28611, "tokens": 9091153920, "loss": 2.804776430130005, "ce_loss": 2.803456127643585, "z_loss": 13.202905178070068, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2522461712360382, "tokens_per_second": 3039361.1846610773, "eta_seconds": 6879.621323561668, "data_wait_seconds": 1.3589859008789062e-05}
869
+ {"step": 8680, "total_steps": 28611, "tokens": 9101639680, "loss": 2.7844672203063965, "ce_loss": 2.7830925583839417, "z_loss": 13.746761083602905, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33008506894111633, "tokens_per_second": 4229157.638312571, "eta_seconds": 4941.6858020782465, "data_wait_seconds": 1.7344951629638672e-05}
870
+ {"step": 8690, "total_steps": 28611, "tokens": 9112125440, "loss": 2.834058105945587, "ce_loss": 2.8322941660881042, "z_loss": 17.639526844024658, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26901131868362427, "tokens_per_second": 2915088.319749064, "eta_seconds": 7165.711705708503, "data_wait_seconds": 1.4483928680419922e-05}
871
+ {"step": 8700, "total_steps": 28611, "tokens": 9122611200, "loss": 2.814219653606415, "ce_loss": 2.8128411173820496, "z_loss": 13.785798788070679, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30532339215278625, "tokens_per_second": 4155476.394256244, "eta_seconds": 5024.260699653625, "data_wait_seconds": 2.378225326538086e-05}
872
+ {"step": 8710, "total_steps": 28611, "tokens": 9133096960, "loss": 2.8469565510749817, "ce_loss": 2.8455621004104614, "z_loss": 13.944255352020264, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3081153333187103, "tokens_per_second": 2953591.6356243975, "eta_seconds": 7065.1984263181685, "data_wait_seconds": 1.3649463653564453e-05}
873
+ {"step": 8720, "total_steps": 28611, "tokens": 9143582720, "loss": 2.7836962938308716, "ce_loss": 2.782175362110138, "z_loss": 15.20930027961731, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2809145152568817, "tokens_per_second": 2669385.3754212903, "eta_seconds": 7813.493476080894, "data_wait_seconds": 1.6927719116210938e-05}
874
+ {"step": 8730, "total_steps": 28611, "tokens": 9154068480, "loss": 2.8184081315994263, "ce_loss": 2.8167812824249268, "z_loss": 16.26852035522461, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.346720427274704, "tokens_per_second": 3938086.301115, "eta_seconds": 5293.621790385247, "data_wait_seconds": 1.436471939086914e-05}
875
+ {"step": 8740, "total_steps": 28611, "tokens": 9164554240, "loss": 2.804417312145233, "ce_loss": 2.8031283020973206, "z_loss": 12.890193462371826, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28202325105667114, "tokens_per_second": 4111825.234788067, "eta_seconds": 5067.397689890861, "data_wait_seconds": 1.3828277587890625e-05}
876
+ {"step": 8750, "total_steps": 28611, "tokens": 9175040000, "loss": 2.8263570070266724, "ce_loss": 2.8246225118637085, "z_loss": 17.345314502716064, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27352234721183777, "tokens_per_second": 2839171.731795747, "eta_seconds": 7335.156131196022, "data_wait_seconds": 1.4901161193847656e-05}
877
+ {"step": 8760, "total_steps": 28611, "tokens": 9185525760, "loss": 2.8307520747184753, "ce_loss": 2.8293434381484985, "z_loss": 14.086578607559204, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24098137021064758, "tokens_per_second": 4292621.44545227, "eta_seconds": 4849.084048175812, "data_wait_seconds": 1.704692840576172e-05}
878
+ {"step": 8770, "total_steps": 28611, "tokens": 9196011520, "loss": 2.8315877318382263, "ce_loss": 2.8299312591552734, "z_loss": 16.564831256866455, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29420584440231323, "tokens_per_second": 2666888.1032180293, "eta_seconds": 7801.1508585214615, "data_wait_seconds": 1.728534698486328e-05}
879
+ {"step": 8780, "total_steps": 28611, "tokens": 9206497280, "loss": 2.8445637226104736, "ce_loss": 2.8427459001541138, "z_loss": 18.178067684173584, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29526621103286743, "tokens_per_second": 4002965.067266941, "eta_seconds": 5194.726985263825, "data_wait_seconds": 1.430511474609375e-05}
880
+ {"step": 8790, "total_steps": 28611, "tokens": 9216983040, "loss": 2.8117619156837463, "ce_loss": 2.8104838728904724, "z_loss": 12.780604362487793, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28945544362068176, "tokens_per_second": 2642678.024275968, "eta_seconds": 7864.682986378669, "data_wait_seconds": 1.4662742614746094e-05}
881
+ {"step": 8800, "total_steps": 28611, "tokens": 9227468800, "loss": 2.849856913089752, "ce_loss": 2.8483439683914185, "z_loss": 15.129639387130737, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3253096640110016, "tokens_per_second": 2838332.1748680673, "eta_seconds": 7318.854121422767, "data_wait_seconds": 0.2766175866127014}
882
+ {"step": 8810, "total_steps": 28611, "tokens": 9237954560, "loss": 2.809188663959503, "ce_loss": 2.8074675798416138, "z_loss": 17.2105975151062, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2580750584602356, "tokens_per_second": 3660574.07906823, "eta_seconds": 5672.021089458465, "data_wait_seconds": 1.4603137969970703e-05}
883
+ {"step": 8820, "total_steps": 28611, "tokens": 9248440320, "loss": 2.8017481565475464, "ce_loss": 2.800202250480652, "z_loss": 15.458945274353027, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32939818501472473, "tokens_per_second": 2854568.3001816184, "eta_seconds": 7269.87951722145, "data_wait_seconds": 0.2432185411453247}
884
+ {"step": 8830, "total_steps": 28611, "tokens": 9258926080, "loss": 2.813024401664734, "ce_loss": 2.811501145362854, "z_loss": 15.232667922973633, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26751160621643066, "tokens_per_second": 3518041.780987017, "eta_seconds": 5895.8600116968155, "data_wait_seconds": 1.3530254364013672e-05}
885
+ {"step": 8840, "total_steps": 28611, "tokens": 9269411840, "loss": 2.830803632736206, "ce_loss": 2.8295669555664062, "z_loss": 12.36720061302185, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25928112864494324, "tokens_per_second": 2680170.7192824483, "eta_seconds": 7735.102822685241, "data_wait_seconds": 1.4007091522216797e-05}
886
+ {"step": 8850, "total_steps": 28611, "tokens": 9279897600, "loss": 2.8672196865081787, "ce_loss": 2.866031587123871, "z_loss": 11.880568504333496, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2715982496738434, "tokens_per_second": 2804133.5418478074, "eta_seconds": 7389.416383624078, "data_wait_seconds": 1.430511474609375e-05}
887
+ {"step": 8860, "total_steps": 28611, "tokens": 9290383360, "loss": 2.8340304493904114, "ce_loss": 2.8328150510787964, "z_loss": 12.154468297958374, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2668568193912506, "tokens_per_second": 2835683.55441115, "eta_seconds": 7303.503433513641, "data_wait_seconds": 1.3232231140136719e-05}
888
+ {"step": 8870, "total_steps": 28611, "tokens": 9300869120, "loss": 2.7806116938591003, "ce_loss": 2.7791049480438232, "z_loss": 15.0672607421875, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3436506688594818, "tokens_per_second": 2300417.49212267, "eta_seconds": 8998.340034747123, "data_wait_seconds": 0.2044951319694519}
889
+ {"step": 8880, "total_steps": 28611, "tokens": 9311354880, "loss": 2.8371034264564514, "ce_loss": 2.8357738256454468, "z_loss": 13.295831441879272, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30522802472114563, "tokens_per_second": 3113955.777413, "eta_seconds": 6644.106254196167, "data_wait_seconds": 1.6927719116210938e-05}
890
+ {"step": 8890, "total_steps": 28611, "tokens": 9321840640, "loss": 2.821288049221039, "ce_loss": 2.8196410536766052, "z_loss": 16.469595909118652, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2975148856639862, "tokens_per_second": 2776048.8917155447, "eta_seconds": 7449.064516735077, "data_wait_seconds": 1.7762184143066406e-05}
891
+ {"step": 8900, "total_steps": 28611, "tokens": 9332326400, "loss": 2.8193825483322144, "ce_loss": 2.817961812019348, "z_loss": 14.207366228103638, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26339179277420044, "tokens_per_second": 2338409.782841762, "eta_seconds": 8838.691014575958, "data_wait_seconds": 0.24005401134490967}
892
+ {"step": 8910, "total_steps": 28611, "tokens": 9342812160, "loss": 2.8320904970169067, "ce_loss": 2.8307119607925415, "z_loss": 13.785569190979004, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31206953525543213, "tokens_per_second": 2705775.336462804, "eta_seconds": 7634.778652024269, "data_wait_seconds": 1.6987323760986328e-05}
893
+ {"step": 8920, "total_steps": 28611, "tokens": 9353297920, "loss": 2.8546940684318542, "ce_loss": 2.853094458580017, "z_loss": 15.996275186538696, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27871641516685486, "tokens_per_second": 3804713.929374079, "eta_seconds": 5426.823251175881, "data_wait_seconds": 1.341104507446289e-05}
894
+ {"step": 8930, "total_steps": 28611, "tokens": 9363783680, "loss": 2.7997503876686096, "ce_loss": 2.7985607981681824, "z_loss": 11.896367311477661, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3129333257675171, "tokens_per_second": 2702605.560182819, "eta_seconds": 7635.973432469368, "data_wait_seconds": 1.8417835235595703e-05}
895
+ {"step": 8940, "total_steps": 28611, "tokens": 9374269440, "loss": 2.7885895371437073, "ce_loss": 2.787076711654663, "z_loss": 15.128572463989258, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3217886686325073, "tokens_per_second": 2884583.2664746298, "eta_seconds": 7150.612962269784, "data_wait_seconds": 1.615285873413086e-05}
896
+ {"step": 8950, "total_steps": 28611, "tokens": 9384755200, "loss": 2.8449071645736694, "ce_loss": 2.8432072401046753, "z_loss": 16.998923778533936, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2867705821990967, "tokens_per_second": 3979720.5711046746, "eta_seconds": 5180.276445960999, "data_wait_seconds": 1.4722347259521484e-05}
897
+ {"step": 8960, "total_steps": 28611, "tokens": 9395240960, "loss": 2.8063745498657227, "ce_loss": 2.804858088493347, "z_loss": 15.164295196533203, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3084969222545624, "tokens_per_second": 2110858.0955874557, "eta_seconds": 9761.701660132409, "data_wait_seconds": 0.3000602126121521}
898
+ {"step": 8970, "total_steps": 28611, "tokens": 9405726720, "loss": 2.8341885805130005, "ce_loss": 2.8330268263816833, "z_loss": 11.617666244506836, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33412036299705505, "tokens_per_second": 3899702.374726933, "eta_seconds": 5281.193085265159, "data_wait_seconds": 1.3530254364013672e-05}
899
+ {"step": 8980, "total_steps": 28611, "tokens": 9416212480, "loss": 2.822474956512451, "ce_loss": 2.8211600184440613, "z_loss": 13.149129152297974, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.290561705827713, "tokens_per_second": 3000556.926337857, "eta_seconds": 6860.258265829087, "data_wait_seconds": 0.24933761358261108}
900
+ {"step": 8990, "total_steps": 28611, "tokens": 9426698240, "loss": 2.7849462628364563, "ce_loss": 2.783479928970337, "z_loss": 14.663403272628784, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25471723079681396, "tokens_per_second": 4350613.94259486, "eta_seconds": 4729.012954831123, "data_wait_seconds": 1.901388168334961e-05}
901
+ {"step": 9000, "total_steps": 28611, "tokens": 9437184000, "loss": 2.794645071029663, "ce_loss": 2.7931392788887024, "z_loss": 15.057351112365723, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24489723145961761, "tokens_per_second": 2991530.759286418, "eta_seconds": 6873.947015976905, "data_wait_seconds": 1.895427703857422e-05}
902
+ {"step": 9010, "total_steps": 28611, "tokens": 9447669760, "loss": 2.836769163608551, "ce_loss": 2.835231602191925, "z_loss": 15.375843524932861, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27775219082832336, "tokens_per_second": 4205032.974552266, "eta_seconds": 4887.747206830978, "data_wait_seconds": 1.3172626495361328e-05}
903
+ {"step": 9020, "total_steps": 28611, "tokens": 9458155520, "loss": 2.8218526244163513, "ce_loss": 2.8203757405281067, "z_loss": 14.769135475158691, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3312542736530304, "tokens_per_second": 2935076.4427196328, "eta_seconds": 6999.017850780487, "data_wait_seconds": 1.710653305053711e-05}
904
+ {"step": 9030, "total_steps": 28611, "tokens": 9468641280, "loss": 2.7842119932174683, "ce_loss": 2.782819390296936, "z_loss": 13.925942182540894, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.292176216840744, "tokens_per_second": 1230987.2864552133, "eta_seconds": 16679.43030924797, "data_wait_seconds": 0.9248417615890503}
905
+ {"step": 9040, "total_steps": 28611, "tokens": 9479127040, "loss": 2.8612383008003235, "ce_loss": 2.859929144382477, "z_loss": 13.091376543045044, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3736031949520111, "tokens_per_second": 3537564.3950595595, "eta_seconds": 5801.076278543473, "data_wait_seconds": 1.7881393432617188e-05}
906
+ {"step": 9050, "total_steps": 28611, "tokens": 9489612800, "loss": 2.825646758079529, "ce_loss": 2.8241140842437744, "z_loss": 15.32678747177124, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29628631472587585, "tokens_per_second": 2630643.763462766, "eta_seconds": 7797.024979543687, "data_wait_seconds": 0.2801538109779358}
907
+ {"step": 9060, "total_steps": 28611, "tokens": 9500098560, "loss": 2.8335589170455933, "ce_loss": 2.8321539759635925, "z_loss": 14.04915165901184, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2541733682155609, "tokens_per_second": 2705599.8934406014, "eta_seconds": 7577.14007370472, "data_wait_seconds": 0.3081730604171753}
908
+ {"step": 9070, "total_steps": 28611, "tokens": 9510584320, "loss": 2.8481569290161133, "ce_loss": 2.846889913082123, "z_loss": 12.670429229736328, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2768104374408722, "tokens_per_second": 2730419.727194345, "eta_seconds": 7504.422639465332, "data_wait_seconds": 1.5616416931152344e-05}
909
+ {"step": 9080, "total_steps": 28611, "tokens": 9521070080, "loss": 2.8985610604286194, "ce_loss": 2.8970688581466675, "z_loss": 14.921764135360718, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.35220208764076233, "tokens_per_second": 2783164.1933095288, "eta_seconds": 7358.436812758446, "data_wait_seconds": 2.2530555725097656e-05}
910
+ {"step": 9090, "total_steps": 28611, "tokens": 9531555840, "loss": 2.8457191586494446, "ce_loss": 2.8442317247390747, "z_loss": 14.874497413635254, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.307015597820282, "tokens_per_second": 4096260.6519616996, "eta_seconds": 4997.058008551598, "data_wait_seconds": 1.8417835235595703e-05}
911
+ {"step": 9100, "total_steps": 28611, "tokens": 9542041600, "loss": 2.8039161562919617, "ce_loss": 2.8025723099708557, "z_loss": 13.43823766708374, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31875136494636536, "tokens_per_second": 2973172.7812926, "eta_seconds": 6881.1225720644, "data_wait_seconds": 1.9848346710205078e-05}
912
+ {"step": 9110, "total_steps": 28611, "tokens": 9552527360, "loss": 2.784780502319336, "ce_loss": 2.783508777618408, "z_loss": 12.717275857925415, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27786973118782043, "tokens_per_second": 4543996.967295346, "eta_seconds": 4500.064749860764, "data_wait_seconds": 1.4960765838623047e-05}
913
+ {"step": 9120, "total_steps": 28611, "tokens": 9563013120, "loss": 2.789046823978424, "ce_loss": 2.7876147031784058, "z_loss": 14.320597648620605, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27545514702796936, "tokens_per_second": 2915145.126429329, "eta_seconds": 7010.901320385932, "data_wait_seconds": 1.4185905456542969e-05}
914
+ {"step": 9130, "total_steps": 28611, "tokens": 9573498880, "loss": 2.844060242176056, "ce_loss": 2.8427330255508423, "z_loss": 13.272156953811646, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30891677737236023, "tokens_per_second": 3966408.457022101, "eta_seconds": 5150.077022409439, "data_wait_seconds": 1.5497207641601562e-05}
915
+ {"step": 9140, "total_steps": 28611, "tokens": 9583984640, "loss": 2.833456039428711, "ce_loss": 2.831874430179596, "z_loss": 15.816231489181519, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3450486958026886, "tokens_per_second": 3874447.4795954973, "eta_seconds": 5269.608996772766, "data_wait_seconds": 2.6881694793701172e-05}
916
+ {"step": 9150, "total_steps": 28611, "tokens": 9594470400, "loss": 2.79126900434494, "ce_loss": 2.789583742618561, "z_loss": 16.852538108825684, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3306998908519745, "tokens_per_second": 4768062.97929232, "eta_seconds": 4279.796140408515, "data_wait_seconds": 1.6510486602783203e-05}
917
+ {"step": 9160, "total_steps": 28611, "tokens": 9604956160, "loss": 2.7929930686950684, "ce_loss": 2.7917052507400513, "z_loss": 12.877637386322021, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2671208083629608, "tokens_per_second": 3921318.14018787, "eta_seconds": 5201.274430394173, "data_wait_seconds": 1.430511474609375e-05}
918
+ {"step": 9170, "total_steps": 28611, "tokens": 9615441920, "loss": 2.80678927898407, "ce_loss": 2.805363118648529, "z_loss": 14.261299848556519, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28710559010505676, "tokens_per_second": 2803339.949245933, "eta_seconds": 7271.813759684563, "data_wait_seconds": 0.3041841387748718}
919
+ {"step": 9180, "total_steps": 28611, "tokens": 9625927680, "loss": 2.832583785057068, "ce_loss": 2.8312990069389343, "z_loss": 12.848446369171143, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2931497097015381, "tokens_per_second": 4239853.751757549, "eta_seconds": 4805.562042689324, "data_wait_seconds": 0.5969132781028748}
920
+ {"step": 9190, "total_steps": 28611, "tokens": 9636413440, "loss": 2.8544240593910217, "ce_loss": 2.853180706501007, "z_loss": 12.433303594589233, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32845190167427063, "tokens_per_second": 4268167.0416597035, "eta_seconds": 4771.227156114578, "data_wait_seconds": 1.3589859008789062e-05}
921
+ {"step": 9200, "total_steps": 28611, "tokens": 9646899200, "loss": 2.802716076374054, "ce_loss": 2.801394999027252, "z_loss": 13.211040019989014, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2940193712711334, "tokens_per_second": 3058956.9186182786, "eta_seconds": 6653.8723092556, "data_wait_seconds": 1.6987323760986328e-05}
922
+ {"step": 9210, "total_steps": 28611, "tokens": 9657384960, "loss": 2.8435535430908203, "ce_loss": 2.8423691987991333, "z_loss": 11.843485116958618, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3206748366355896, "tokens_per_second": 4201574.581319623, "eta_seconds": 4841.85692346096, "data_wait_seconds": 1.4960765838623047e-05}
923
+ {"step": 9220, "total_steps": 28611, "tokens": 9667870720, "loss": 2.8137460350990295, "ce_loss": 2.812641441822052, "z_loss": 11.0452880859375, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28693652153015137, "tokens_per_second": 2848022.8766112644, "eta_seconds": 7139.316675782203, "data_wait_seconds": 1.7702579498291016e-05}
924
+ {"step": 9230, "total_steps": 28611, "tokens": 9678356480, "loss": 2.8343218564987183, "ce_loss": 2.833043158054352, "z_loss": 12.787087440490723, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2610447406768799, "tokens_per_second": 3094171.7979593314, "eta_seconds": 6567.977728128433, "data_wait_seconds": 2.1159648895263672e-05}
925
+ {"step": 9240, "total_steps": 28611, "tokens": 9688842240, "loss": 2.803758382797241, "ce_loss": 2.802326738834381, "z_loss": 14.316800117492676, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27948451042175293, "tokens_per_second": 2885912.3957708115, "eta_seconds": 7038.316799139977, "data_wait_seconds": 1.3113021850585938e-05}
926
+ {"step": 9250, "total_steps": 28611, "tokens": 9699328000, "loss": 2.8155024647712708, "ce_loss": 2.813901960849762, "z_loss": 16.004956483840942, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26887691020965576, "tokens_per_second": 2276854.215469202, "eta_seconds": 8916.460174775124, "data_wait_seconds": 0.2797588109970093}
927
+ {"step": 9260, "total_steps": 28611, "tokens": 9709813760, "loss": 2.799150288105011, "ce_loss": 2.797906219959259, "z_loss": 12.440809726715088, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2875068485736847, "tokens_per_second": 2826931.3535407293, "eta_seconds": 7177.745632410049, "data_wait_seconds": 1.8477439880371094e-05}
928
+ {"step": 9270, "total_steps": 28611, "tokens": 9720299520, "loss": 2.816617727279663, "ce_loss": 2.8149109482765198, "z_loss": 17.06803846359253, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25239405035972595, "tokens_per_second": 4123612.6656989376, "eta_seconds": 4918.14097495079, "data_wait_seconds": 1.4483928680419922e-05}
929
+ {"step": 9280, "total_steps": 28611, "tokens": 9730785280, "loss": 2.834295094013214, "ce_loss": 2.8329572677612305, "z_loss": 13.37742280960083, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26997414231300354, "tokens_per_second": 2690306.2714939313, "eta_seconds": 7534.466566419601, "data_wait_seconds": 0.2836703062057495}
930
+ {"step": 9290, "total_steps": 28611, "tokens": 9741271040, "loss": 2.787407636642456, "ce_loss": 2.7859299778938293, "z_loss": 14.777347564697266, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26810401678085327, "tokens_per_second": 2978732.973029206, "eta_seconds": 6801.39410932064, "data_wait_seconds": 0.280234694480896}
931
+ {"step": 9300, "total_steps": 28611, "tokens": 9751756800, "loss": 2.786916136741638, "ce_loss": 2.7857056856155396, "z_loss": 12.104745149612427, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27530935406684875, "tokens_per_second": 4121458.713976039, "eta_seconds": 4913.07872800827, "data_wait_seconds": 1.7523765563964844e-05}
932
+ {"step": 9310, "total_steps": 28611, "tokens": 9762242560, "loss": 2.813118278980255, "ce_loss": 2.8118244409561157, "z_loss": 12.938410520553589, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.310531347990036, "tokens_per_second": 4150881.445269159, "eta_seconds": 4875.727154064179, "data_wait_seconds": 1.2993812561035156e-05}
933
+ {"step": 9320, "total_steps": 28611, "tokens": 9772728320, "loss": 2.791707158088684, "ce_loss": 2.790374279022217, "z_loss": 13.32869553565979, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31803715229034424, "tokens_per_second": 2807830.016039192, "eta_seconds": 7204.168165612221, "data_wait_seconds": 1.424551010131836e-05}
934
+ {"step": 9330, "total_steps": 28611, "tokens": 9783214080, "loss": 2.7901164889335632, "ce_loss": 2.7887313961982727, "z_loss": 13.851346969604492, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2713685929775238, "tokens_per_second": 3019484.332113675, "eta_seconds": 6695.7108010053635, "data_wait_seconds": 1.8775463104248047e-05}
935
+ {"step": 9340, "total_steps": 28611, "tokens": 9793699840, "loss": 2.80062198638916, "ce_loss": 2.7992645502090454, "z_loss": 13.574120998382568, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.270359605550766, "tokens_per_second": 3804682.002802182, "eta_seconds": 5311.116167163848, "data_wait_seconds": 1.424551010131836e-05}
936
+ {"step": 9350, "total_steps": 28611, "tokens": 9804185600, "loss": 2.845884621143341, "ce_loss": 2.844442129135132, "z_loss": 14.425289869308472, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31480470299720764, "tokens_per_second": 3053558.354947896, "eta_seconds": 6614.126860642433, "data_wait_seconds": 1.627206802368164e-05}
937
+ {"step": 9360, "total_steps": 28611, "tokens": 9814671360, "loss": 2.8695350885391235, "ce_loss": 2.8681676387786865, "z_loss": 13.67435598373413, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3098427653312683, "tokens_per_second": 2329705.400851557, "eta_seconds": 8664.67346842289, "data_wait_seconds": 1.811981201171875e-05}
938
+ {"step": 9370, "total_steps": 28611, "tokens": 9825157120, "loss": 2.8215821981430054, "ce_loss": 2.82023286819458, "z_loss": 13.493014574050903, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2832052409648895, "tokens_per_second": 2963734.6699136673, "eta_seconds": 6807.509127187729, "data_wait_seconds": 1.5497207641601562e-05}
939
+ {"step": 9380, "total_steps": 28611, "tokens": 9835642880, "loss": 2.8552148938179016, "ce_loss": 2.8538413047790527, "z_loss": 13.735977411270142, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.262011855840683, "tokens_per_second": 3043676.932615103, "eta_seconds": 6625.264606738091, "data_wait_seconds": 1.4781951904296875e-05}
940
+ {"step": 9390, "total_steps": 28611, "tokens": 9846128640, "loss": 2.814690589904785, "ce_loss": 2.8136239647865295, "z_loss": 10.666279554367065, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3437926173210144, "tokens_per_second": 2758239.3131141104, "eta_seconds": 7307.081441473961, "data_wait_seconds": 1.3887882232666016e-05}
941
+ {"step": 9400, "total_steps": 28611, "tokens": 9856614400, "loss": 2.837191164493561, "ce_loss": 2.835936963558197, "z_loss": 12.54219388961792, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.31988072395324707, "tokens_per_second": 3121890.439488387, "eta_seconds": 6452.562614369393, "data_wait_seconds": 0.19665461778640747}
942
+ {"step": 9410, "total_steps": 28611, "tokens": 9867100160, "loss": 2.811601996421814, "ce_loss": 2.8101556301116943, "z_loss": 14.46324348449707, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2673908472061157, "tokens_per_second": 3643729.3192619598, "eta_seconds": 5525.577234721183, "data_wait_seconds": 1.5676021575927734e-05}
943
+ {"step": 9420, "total_steps": 28611, "tokens": 9877585920, "loss": 2.8146833181381226, "ce_loss": 2.8135628700256348, "z_loss": 11.204581499099731, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29083704948425293, "tokens_per_second": 4446926.68452766, "eta_seconds": 4525.197612547874, "data_wait_seconds": 1.3887882232666016e-05}
944
+ {"step": 9430, "total_steps": 28611, "tokens": 9888071680, "loss": 2.813759684562683, "ce_loss": 2.8121864199638367, "z_loss": 15.732870101928711, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2830774486064911, "tokens_per_second": 3092803.1633064374, "eta_seconds": 6503.076721668243, "data_wait_seconds": 1.3649463653564453e-05}
945
+ {"step": 9440, "total_steps": 28611, "tokens": 9898557440, "loss": 2.8403967022895813, "ce_loss": 2.838951826095581, "z_loss": 14.448685884475708, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25940781831741333, "tokens_per_second": 3168117.6567270607, "eta_seconds": 6345.1716994524, "data_wait_seconds": 1.4483928680419922e-05}
946
+ {"step": 9450, "total_steps": 28611, "tokens": 9909043200, "loss": 2.8405603766441345, "ce_loss": 2.839168965816498, "z_loss": 13.913745880126953, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3145098090171814, "tokens_per_second": 3062704.777813494, "eta_seconds": 6560.137588691711, "data_wait_seconds": 0.2196546196937561}
947
+ {"step": 9460, "total_steps": 28611, "tokens": 9919528960, "loss": 2.853396475315094, "ce_loss": 2.8517003655433655, "z_loss": 16.96079444885254, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3008520305156708, "tokens_per_second": 3710202.7481226684, "eta_seconds": 5412.447873950005, "data_wait_seconds": 1.531839370727539e-05}
948
+ {"step": 9470, "total_steps": 28611, "tokens": 9930014720, "loss": 2.849819004535675, "ce_loss": 2.8484115600585938, "z_loss": 14.074568748474121, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27697014808654785, "tokens_per_second": 4165639.2489464064, "eta_seconds": 4818.178439497948, "data_wait_seconds": 1.4483928680419922e-05}
949
+ {"step": 9480, "total_steps": 28611, "tokens": 9940500480, "loss": 2.8422343730926514, "ce_loss": 2.841057240962982, "z_loss": 11.771397590637207, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26881781220436096, "tokens_per_second": 2821388.3811117806, "eta_seconds": 7110.083670258522, "data_wait_seconds": 1.4662742614746094e-05}
950
+ {"step": 9490, "total_steps": 28611, "tokens": 9950986240, "loss": 2.8221386671066284, "ce_loss": 2.820536196231842, "z_loss": 16.024744749069214, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2708776593208313, "tokens_per_second": 4090844.159836651, "eta_seconds": 4901.14531686306, "data_wait_seconds": 1.2099742889404297e-05}
951
+ {"step": 9500, "total_steps": 28611, "tokens": 9961472000, "loss": 2.792564034461975, "ce_loss": 2.7910647988319397, "z_loss": 14.992708444595337, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2644948363304138, "tokens_per_second": 3013132.274836686, "eta_seconds": 6650.665854716301, "data_wait_seconds": 1.424551010131836e-05}
952
+ {"step": 9510, "total_steps": 28611, "tokens": 9971957760, "loss": 2.831292688846588, "ce_loss": 2.8299097418785095, "z_loss": 13.829715967178345, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2843356728553772, "tokens_per_second": 3004645.4764912208, "eta_seconds": 6665.961203312873, "data_wait_seconds": 1.7464160919189453e-05}
953
+ {"step": 9520, "total_steps": 28611, "tokens": 9982443520, "loss": 2.8212174773216248, "ce_loss": 2.8200559616088867, "z_loss": 11.615208625793457, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32027408480644226, "tokens_per_second": 4237306.8192980215, "eta_seconds": 4724.3131710052485, "data_wait_seconds": 1.3887882232666016e-05}
954
+ {"step": 9530, "total_steps": 28611, "tokens": 9992929280, "loss": 2.818031668663025, "ce_loss": 2.8166032433509827, "z_loss": 14.283878326416016, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29670238494873047, "tokens_per_second": 4718303.736361398, "eta_seconds": 4240.48127758503, "data_wait_seconds": 1.3470649719238281e-05}
955
+ {"step": 9540, "total_steps": 28611, "tokens": 10003415040, "loss": 2.8241374492645264, "ce_loss": 2.822677254676819, "z_loss": 14.601726293563843, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2733669579029083, "tokens_per_second": 3220219.625079022, "eta_seconds": 6209.946905565263, "data_wait_seconds": 1.4424324035644531e-05}
956
+ {"step": 9550, "total_steps": 28611, "tokens": 10013900800, "loss": 2.8247159123420715, "ce_loss": 2.823478102684021, "z_loss": 12.378229141235352, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25881123542785645, "tokens_per_second": 3099699.104792685, "eta_seconds": 6448.0152622222895, "data_wait_seconds": 1.4543533325195312e-05}
957
+ {"step": 9560, "total_steps": 28611, "tokens": 10024386560, "loss": 2.8077958822250366, "ce_loss": 2.8064409494400024, "z_loss": 13.549615621566772, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34960731863975525, "tokens_per_second": 2968538.088818962, "eta_seconds": 6729.380179166794, "data_wait_seconds": 0.26208406686782837}
958
+ {"step": 9570, "total_steps": 28611, "tokens": 10034872320, "loss": 2.618810713291168, "ce_loss": 2.617134392261505, "z_loss": 16.763168334960938, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.34527650475502014, "tokens_per_second": 3987612.701095775, "eta_seconds": 5006.989673423767, "data_wait_seconds": 1.8715858459472656e-05}
959
+ {"step": 9580, "total_steps": 28611, "tokens": 10045358080, "loss": 2.881094515323639, "ce_loss": 2.8796700835227966, "z_loss": 14.243419408798218, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2881266474723816, "tokens_per_second": 2961645.088976359, "eta_seconds": 6737.961253452301, "data_wait_seconds": 1.806020736694336e-05}
960
+ {"step": 9590, "total_steps": 28611, "tokens": 10055843840, "loss": 2.8347487449645996, "ce_loss": 2.833560347557068, "z_loss": 11.884002685546875, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2787410616874695, "tokens_per_second": 4179231.5589119554, "eta_seconds": 4772.399857449532, "data_wait_seconds": 1.424551010131836e-05}
961
+ {"step": 9600, "total_steps": 28611, "tokens": 10066329600, "loss": 2.847086191177368, "ce_loss": 2.845674514770508, "z_loss": 14.116903066635132, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25186648964881897, "tokens_per_second": 3088857.8820168497, "eta_seconds": 6453.6728776216505, "data_wait_seconds": 1.5914440155029297e-05}
962
+ {"step": 9610, "total_steps": 28611, "tokens": 10076815360, "loss": 2.8635266423225403, "ce_loss": 2.8619073629379272, "z_loss": 16.192959308624268, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2619333565235138, "tokens_per_second": 3063074.8636023286, "eta_seconds": 6504.572517228127, "data_wait_seconds": 0.523722767829895}
963
+ {"step": 9620, "total_steps": 28611, "tokens": 10087301120, "loss": 2.8435025215148926, "ce_loss": 2.8421283960342407, "z_loss": 13.741036653518677, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3120863139629364, "tokens_per_second": 4454421.220630174, "eta_seconds": 4470.5037601232525, "data_wait_seconds": 1.4722347259521484e-05}
964
+ {"step": 9630, "total_steps": 28611, "tokens": 10097786880, "loss": 2.8602271676063538, "ce_loss": 2.858709990978241, "z_loss": 15.17167615890503, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3449172079563141, "tokens_per_second": 3178874.1910627745, "eta_seconds": 6261.0282319307325, "data_wait_seconds": 1.4066696166992188e-05}
965
+ {"step": 9640, "total_steps": 28611, "tokens": 10108272640, "loss": 2.81310111284256, "ce_loss": 2.811613440513611, "z_loss": 14.876000881195068, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3034234642982483, "tokens_per_second": 2890760.3921533315, "eta_seconds": 6881.419625782966, "data_wait_seconds": 1.7821788787841797e-05}
966
+ {"step": 9650, "total_steps": 28611, "tokens": 10118758400, "loss": 2.8242205381393433, "ce_loss": 2.8230010867118835, "z_loss": 12.194533348083496, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30924734473228455, "tokens_per_second": 4198585.16536981, "eta_seconds": 4735.416516017914, "data_wait_seconds": 1.7940998077392578e-05}
967
+ {"step": 9660, "total_steps": 28611, "tokens": 10129244160, "loss": 2.862133800983429, "ce_loss": 2.8608235716819763, "z_loss": 13.101827383041382, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26940983533859253, "tokens_per_second": 3288828.8095246106, "eta_seconds": 6042.1399005174635, "data_wait_seconds": 1.8656253814697266e-05}
968
+ {"step": 9670, "total_steps": 28611, "tokens": 10139729920, "loss": 2.809272885322571, "ce_loss": 2.8075400590896606, "z_loss": 17.32775115966797, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26794084906578064, "tokens_per_second": 3981881.7343872143, "eta_seconds": 4987.862357759475, "data_wait_seconds": 3.2961368560791016e-05}
969
+ {"step": 9680, "total_steps": 28611, "tokens": 10150215680, "loss": 2.8388407230377197, "ce_loss": 2.837779402732849, "z_loss": 10.612962007522583, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2973119914531708, "tokens_per_second": 3285204.018202736, "eta_seconds": 6042.422980737686, "data_wait_seconds": 1.3887882232666016e-05}
970
+ {"step": 9690, "total_steps": 28611, "tokens": 10160701440, "loss": 2.831398069858551, "ce_loss": 2.8300453424453735, "z_loss": 13.527144432067871, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28917884826660156, "tokens_per_second": 3155501.5214866716, "eta_seconds": 6287.465355634689, "data_wait_seconds": 2.7954578399658203e-05}
971
+ {"step": 9700, "total_steps": 28611, "tokens": 10171187200, "loss": 2.8336825370788574, "ce_loss": 2.8322412967681885, "z_loss": 14.412925004959106, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3268662691116333, "tokens_per_second": 3095849.3188386983, "eta_seconds": 6405.227998447419, "data_wait_seconds": 0.23084819316864014}
972
+ {"step": 9710, "total_steps": 28611, "tokens": 10181672960, "loss": 2.8136470913887024, "ce_loss": 2.8122565746307373, "z_loss": 13.905677080154419, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2656189501285553, "tokens_per_second": 4342049.701335695, "eta_seconds": 4564.465250110626, "data_wait_seconds": 1.728534698486328e-05}
973
+ {"step": 9720, "total_steps": 28611, "tokens": 10192158720, "loss": 2.841222047805786, "ce_loss": 2.839919090270996, "z_loss": 13.029723405838013, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30006977915763855, "tokens_per_second": 3308722.4397198977, "eta_seconds": 5986.796891212463, "data_wait_seconds": 1.7940998077392578e-05}
974
+ {"step": 9730, "total_steps": 28611, "tokens": 10202644480, "loss": 2.8114838004112244, "ce_loss": 2.81005322933197, "z_loss": 14.305909872055054, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2882629632949829, "tokens_per_second": 3872651.9531845534, "eta_seconds": 5112.30125901699, "data_wait_seconds": 2.1398067474365234e-05}
975
+ {"step": 9740, "total_steps": 28611, "tokens": 10213130240, "loss": 2.812370479106903, "ce_loss": 2.81087589263916, "z_loss": 14.945815563201904, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29327142238616943, "tokens_per_second": 2998287.974609037, "eta_seconds": 6599.6588264942175, "data_wait_seconds": 1.4126300811767578e-05}
976
+ {"step": 9750, "total_steps": 28611, "tokens": 10223616000, "loss": 2.7934183478355408, "ce_loss": 2.7922974228858948, "z_loss": 11.209529399871826, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3049093186855316, "tokens_per_second": 3143454.7439533807, "eta_seconds": 6291.546577548981, "data_wait_seconds": 1.6391277313232422e-05}
977
+ {"step": 9760, "total_steps": 28611, "tokens": 10234101760, "loss": 2.795899987220764, "ce_loss": 2.7945163249969482, "z_loss": 13.83653974533081, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2970407009124756, "tokens_per_second": 3122685.9750735895, "eta_seconds": 6330.03328986168, "data_wait_seconds": 0.22252893447875977}
978
+ {"step": 9770, "total_steps": 28611, "tokens": 10244587520, "loss": 2.863850712776184, "ce_loss": 2.8627561926841736, "z_loss": 10.94560170173645, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2949050962924957, "tokens_per_second": 3760304.5071483934, "eta_seconds": 5253.888449311256, "data_wait_seconds": 1.5676021575927734e-05}
979
+ {"step": 9780, "total_steps": 28611, "tokens": 10255073280, "loss": 2.8369345664978027, "ce_loss": 2.835663855075836, "z_loss": 12.707458257675171, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2725819945335388, "tokens_per_second": 4569264.149612624, "eta_seconds": 4321.425509548188, "data_wait_seconds": 1.329183578491211e-05}
980
+ {"step": 9790, "total_steps": 28611, "tokens": 10265559040, "loss": 2.8909738659858704, "ce_loss": 2.889711022377014, "z_loss": 12.627803087234497, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30095162987709045, "tokens_per_second": 2881232.3834149945, "eta_seconds": 6849.5859652280815, "data_wait_seconds": 1.4543533325195312e-05}
981
+ {"step": 9800, "total_steps": 28611, "tokens": 10276044800, "loss": 2.848194897174835, "ce_loss": 2.846541464328766, "z_loss": 16.535229206085205, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3243780732154846, "tokens_per_second": 2971974.5433975663, "eta_seconds": 6636.921968197822, "data_wait_seconds": 1.6927719116210938e-05}
982
+ {"step": 9810, "total_steps": 28611, "tokens": 10286530560, "loss": 2.852797567844391, "ce_loss": 2.8516459465026855, "z_loss": 11.515749454498291, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2960173487663269, "tokens_per_second": 2794897.498379669, "eta_seconds": 7053.667401909828, "data_wait_seconds": 1.7464160919189453e-05}
983
+ {"step": 9820, "total_steps": 28611, "tokens": 10297016320, "loss": 2.804863691329956, "ce_loss": 2.8037951588630676, "z_loss": 10.684940814971924, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.284910649061203, "tokens_per_second": 4284190.298159152, "eta_seconds": 4599.186834549904, "data_wait_seconds": 1.722574234008789e-05}
984
+ {"step": 9830, "total_steps": 28611, "tokens": 10307502080, "loss": 2.8143792748451233, "ce_loss": 2.8129988312721252, "z_loss": 13.80492377281189, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27463313937187195, "tokens_per_second": 3128546.2608107273, "eta_seconds": 6294.714610004426, "data_wait_seconds": 1.430511474609375e-05}
985
+ {"step": 9840, "total_steps": 28611, "tokens": 10317987840, "loss": 2.8302306532859802, "ce_loss": 2.8288604021072388, "z_loss": 13.702406167984009, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29104986786842346, "tokens_per_second": 3990567.652793436, "eta_seconds": 4932.335900187492, "data_wait_seconds": 1.3053417205810547e-05}
986
+ {"step": 9850, "total_steps": 28611, "tokens": 10328473600, "loss": 2.808159291744232, "ce_loss": 2.806900441646576, "z_loss": 12.589034795761108, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3308359384536743, "tokens_per_second": 2849435.9278609236, "eta_seconds": 6903.939879345894, "data_wait_seconds": 1.4483928680419922e-05}
987
+ {"step": 9860, "total_steps": 28611, "tokens": 10338959360, "loss": 2.7593142986297607, "ce_loss": 2.7576913237571716, "z_loss": 16.229899644851685, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2904652953147888, "tokens_per_second": 2819076.9826170295, "eta_seconds": 6974.569583320618, "data_wait_seconds": 1.8894672393798828e-05}
988
+ {"step": 9870, "total_steps": 28611, "tokens": 10349445120, "loss": 2.8145105838775635, "ce_loss": 2.812845230102539, "z_loss": 16.653666734695435, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2970823645591736, "tokens_per_second": 4482857.6958518205, "eta_seconds": 4383.668666124344, "data_wait_seconds": 1.919269561767578e-05}
989
+ {"step": 9880, "total_steps": 28611, "tokens": 10359930880, "loss": 2.80223685503006, "ce_loss": 2.8008116483688354, "z_loss": 14.252162456512451, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29168108105659485, "tokens_per_second": 3254215.478657996, "eta_seconds": 6035.518294596672, "data_wait_seconds": 0.23858964443206787}
990
+ {"step": 9890, "total_steps": 28611, "tokens": 10370416640, "loss": 2.807668924331665, "ce_loss": 2.80642169713974, "z_loss": 12.472139596939087, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3210304081439972, "tokens_per_second": 4488470.79489079, "eta_seconds": 4373.51431992054, "data_wait_seconds": 1.329183578491211e-05}
991
+ {"step": 9900, "total_steps": 28611, "tokens": 10380902400, "loss": 2.8031919598579407, "ce_loss": 2.801698386669159, "z_loss": 14.935458421707153, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30850645899772644, "tokens_per_second": 3182538.8222241616, "eta_seconds": 6164.859765100479, "data_wait_seconds": 1.4066696166992188e-05}
992
+ {"step": 9910, "total_steps": 28611, "tokens": 10391388160, "loss": 2.8245853185653687, "ce_loss": 2.823222041130066, "z_loss": 13.632985591888428, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23927199840545654, "tokens_per_second": 2826487.6956566893, "eta_seconds": 6937.733996200562, "data_wait_seconds": 1.8596649169921875e-05}
993
+ {"step": 9920, "total_steps": 28611, "tokens": 10401873920, "loss": 2.7980403304100037, "ce_loss": 2.796851634979248, "z_loss": 11.88725209236145, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.33535245060920715, "tokens_per_second": 4565963.944531703, "eta_seconds": 4292.397893214226, "data_wait_seconds": 1.7642974853515625e-05}
994
+ {"step": 9930, "total_steps": 28611, "tokens": 10412359680, "loss": 2.822555124759674, "ce_loss": 2.8213923573493958, "z_loss": 11.62772798538208, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2881563901901245, "tokens_per_second": 2953111.101944641, "eta_seconds": 6633.15655245781, "data_wait_seconds": 1.2814998626708984e-05}
995
+ {"step": 9940, "total_steps": 28611, "tokens": 10422845440, "loss": 2.823573648929596, "ce_loss": 2.8221479654312134, "z_loss": 14.256353855133057, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30406996607780457, "tokens_per_second": 2365800.0967955957, "eta_seconds": 8275.408612298967, "data_wait_seconds": 0.2447565793991089}
996
+ {"step": 9950, "total_steps": 28611, "tokens": 10433331200, "loss": 2.8492811918258667, "ce_loss": 2.8478938937187195, "z_loss": 13.873215198516846, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.30570054054260254, "tokens_per_second": 2770257.74343036, "eta_seconds": 7063.413786101341, "data_wait_seconds": 1.704692840576172e-05}
997
+ {"step": 9960, "total_steps": 28611, "tokens": 10443816960, "loss": 2.8195201754570007, "ce_loss": 2.8179991245269775, "z_loss": 15.210329532623291, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2843860387802124, "tokens_per_second": 4052443.585757295, "eta_seconds": 4825.974887037277, "data_wait_seconds": 1.4066696166992188e-05}
998
+ {"step": 9970, "total_steps": 28611, "tokens": 10454302720, "loss": 2.825184166431427, "ce_loss": 2.823630630970001, "z_loss": 15.534719944000244, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2541646957397461, "tokens_per_second": 2933467.0372996596, "eta_seconds": 6663.277605462074, "data_wait_seconds": 0.2717287540435791}
999
+ {"step": 9980, "total_steps": 28611, "tokens": 10464788480, "loss": 2.808980643749237, "ce_loss": 2.8078280687332153, "z_loss": 11.526022911071777, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2763822078704834, "tokens_per_second": 4283261.527070793, "eta_seconds": 4561.0148557424545, "data_wait_seconds": 1.3530254364013672e-05}
1000
+ {"step": 9990, "total_steps": 28611, "tokens": 10475274240, "loss": 2.799479305744171, "ce_loss": 2.797928512096405, "z_loss": 15.508025884628296, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29077285528182983, "tokens_per_second": 2976363.7469898304, "eta_seconds": 6560.197393798828, "data_wait_seconds": 1.5616416931152344e-05}
1001
+ {"step": 10000, "total_steps": 28611, "tokens": 10485760000, "loss": 2.8447009325027466, "ce_loss": 2.8436545729637146, "z_loss": 10.463164806365967, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.32812371850013733, "tokens_per_second": 2952908.8604886844, "eta_seconds": 6608.753895902634, "data_wait_seconds": 2.777576446533203e-05}
1002
+ {"step": 10010, "total_steps": 28611, "tokens": 10496245760, "loss": 2.8154792189598083, "ce_loss": 2.8140763640403748, "z_loss": 14.02950644493103, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.3401806652545929, "tokens_per_second": 4103180.473381481, "eta_seconds": 4753.522859287262, "data_wait_seconds": 1.6450881958007812e-05}