Banaxi-Tech commited on
Commit
864a6ce
·
verified ·
1 Parent(s): b04bab3

Save 10% checkpoint at 3,001,024,512 tokens

Browse files
checkpoint_metadata.json CHANGED
@@ -50,10 +50,10 @@
50
  "model_type": "tiger_alpha",
51
  "output_attentions": false
52
  },
53
- "training_percent": 5,
54
- "step": 1431,
55
  "total_steps": 28611,
56
- "tokens_seen": 1500512256,
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": 493.9794752597809
73
  }
 
50
  "model_type": "tiger_alpha",
51
  "output_attentions": false
52
  },
53
+ "training_percent": 10,
54
+ "step": 2862,
55
  "total_steps": 28611,
56
+ "tokens_seen": 3001024512,
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": 932.8135499954224
73
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4405cde8b5a30e43455ef5bde772fcbecd709717908cc512aa1f33d308811c63
3
  size 12811532
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05ed82075e961b7ff7c39b71fc576f89c288a2d94212e19b7ee2290b75694b46
3
  size 12811532
training_metrics.jsonl CHANGED
@@ -142,3 +142,146 @@
142
  {"step": 1410, "total_steps": 28611, "tokens": 1478492160, "loss": 2.931549310684204, "ce_loss": 2.9234726428985596, "z_loss": 80.76691818237305, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2845343351364136, "tokens_per_second": 4096805.5325329774, "eta_seconds": 6962.086813616753, "data_wait_seconds": 1.2278556823730469e-05}
143
  {"step": 1420, "total_steps": 28611, "tokens": 1488977920, "loss": 2.9241446256637573, "ce_loss": 2.9153729677200317, "z_loss": 87.71646499633789, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24240943789482117, "tokens_per_second": 2858935.460899661, "eta_seconds": 9972.883405709266, "data_wait_seconds": 1.3530254364013672e-05}
144
  {"step": 1430, "total_steps": 28611, "tokens": 1499463680, "loss": 2.901202142238617, "ce_loss": 2.892543315887451, "z_loss": 86.58905410766602, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25746312737464905, "tokens_per_second": 2721404.630351088, "eta_seconds": 10473.027031016349, "data_wait_seconds": 1.704692840576172e-05}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
142
  {"step": 1410, "total_steps": 28611, "tokens": 1478492160, "loss": 2.931549310684204, "ce_loss": 2.9234726428985596, "z_loss": 80.76691818237305, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2845343351364136, "tokens_per_second": 4096805.5325329774, "eta_seconds": 6962.086813616753, "data_wait_seconds": 1.2278556823730469e-05}
143
  {"step": 1420, "total_steps": 28611, "tokens": 1488977920, "loss": 2.9241446256637573, "ce_loss": 2.9153729677200317, "z_loss": 87.71646499633789, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24240943789482117, "tokens_per_second": 2858935.460899661, "eta_seconds": 9972.883405709266, "data_wait_seconds": 1.3530254364013672e-05}
144
  {"step": 1430, "total_steps": 28611, "tokens": 1499463680, "loss": 2.901202142238617, "ce_loss": 2.892543315887451, "z_loss": 86.58905410766602, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25746312737464905, "tokens_per_second": 2721404.630351088, "eta_seconds": 10473.027031016349, "data_wait_seconds": 1.704692840576172e-05}
145
+ {"step": 1440, "total_steps": 28611, "tokens": 1509949440, "loss": 2.978959798812866, "ce_loss": 2.9713371992111206, "z_loss": 76.22568702697754, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2624586820602417, "tokens_per_second": 3836416.083616742, "eta_seconds": 7426.425568819046, "data_wait_seconds": 1.6033649444580078e-05}
146
+ {"step": 1450, "total_steps": 28611, "tokens": 1520435200, "loss": 2.9207948446273804, "ce_loss": 2.912679135799408, "z_loss": 81.15680694580078, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24083353579044342, "tokens_per_second": 3083865.538366198, "eta_seconds": 9235.283569169045, "data_wait_seconds": 0.22692734003067017}
147
+ {"step": 1460, "total_steps": 28611, "tokens": 1530920960, "loss": 2.9458272457122803, "ce_loss": 2.9373353123664856, "z_loss": 84.91980171203613, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26226750016212463, "tokens_per_second": 3622802.6168450634, "eta_seconds": 7858.525563502311, "data_wait_seconds": 1.621246337890625e-05}
148
+ {"step": 1470, "total_steps": 28611, "tokens": 1541406720, "loss": 2.9841070771217346, "ce_loss": 2.976336121559143, "z_loss": 77.70969772338867, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25669941306114197, "tokens_per_second": 2839008.2522758143, "eta_seconds": 10024.41651699543, "data_wait_seconds": 1.4007091522216797e-05}
149
+ {"step": 1480, "total_steps": 28611, "tokens": 1551892480, "loss": 2.916088819503784, "ce_loss": 2.907802641391754, "z_loss": 82.86191940307617, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.21100950241088867, "tokens_per_second": 3073499.4755636663, "eta_seconds": 9256.196619582177, "data_wait_seconds": 0.2970985770225525}
150
+ {"step": 1490, "total_steps": 28611, "tokens": 1562378240, "loss": 2.9446420669555664, "ce_loss": 2.936994969844818, "z_loss": 76.47081184387207, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2597370445728302, "tokens_per_second": 5061253.420428628, "eta_seconds": 5618.851168608665, "data_wait_seconds": 1.3530254364013672e-05}
151
+ {"step": 1500, "total_steps": 28611, "tokens": 1572864000, "loss": 2.9609897136688232, "ce_loss": 2.9535757303237915, "z_loss": 74.1401424407959, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.29182523488998413, "tokens_per_second": 3970103.497015774, "eta_seconds": 7160.50449500084, "data_wait_seconds": 1.3828277587890625e-05}
152
+ {"step": 1510, "total_steps": 28611, "tokens": 1583349760, "loss": 2.971490204334259, "ce_loss": 2.9633942246437073, "z_loss": 80.96033668518066, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2385442703962326, "tokens_per_second": 3874017.8073484884, "eta_seconds": 7335.396890044212, "data_wait_seconds": 1.4483928680419922e-05}
153
+ {"step": 1520, "total_steps": 28611, "tokens": 1593835520, "loss": 2.934325933456421, "ce_loss": 2.926321268081665, "z_loss": 80.04651260375977, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23568379878997803, "tokens_per_second": 4143185.8888839064, "eta_seconds": 6856.311345386505, "data_wait_seconds": 1.800060272216797e-05}
154
+ {"step": 1530, "total_steps": 28611, "tokens": 1604321280, "loss": 2.9581701159477234, "ce_loss": 2.9496359825134277, "z_loss": 85.3414077758789, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26262322068214417, "tokens_per_second": 4800627.099683454, "eta_seconds": 5915.161929130554, "data_wait_seconds": 1.3709068298339844e-05}
155
+ {"step": 1540, "total_steps": 28611, "tokens": 1614807040, "loss": 2.934606969356537, "ce_loss": 2.9270864725112915, "z_loss": 75.20508575439453, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23062701523303986, "tokens_per_second": 3063829.6037656143, "eta_seconds": 9264.875847244262, "data_wait_seconds": 1.2099742889404297e-05}
156
+ {"step": 1550, "total_steps": 28611, "tokens": 1625292800, "loss": 2.989987790584564, "ce_loss": 2.9822123646736145, "z_loss": 77.75410461425781, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24167869985103607, "tokens_per_second": 2975226.1338069113, "eta_seconds": 9537.263340616226, "data_wait_seconds": 1.4424324035644531e-05}
157
+ {"step": 1560, "total_steps": 28611, "tokens": 1635778560, "loss": 2.979055166244507, "ce_loss": 2.971457004547119, "z_loss": 75.98152923583984, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25186818838119507, "tokens_per_second": 4256679.626693726, "eta_seconds": 6663.651452207565, "data_wait_seconds": 1.4126300811767578e-05}
158
+ {"step": 1570, "total_steps": 28611, "tokens": 1646264320, "loss": 2.939562976360321, "ce_loss": 2.931536853313446, "z_loss": 80.26130867004395, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24086062610149384, "tokens_per_second": 2835438.578721464, "eta_seconds": 10000.05566291809, "data_wait_seconds": 1.633167266845703e-05}
159
+ {"step": 1580, "total_steps": 28611, "tokens": 1656750080, "loss": 2.9668769240379333, "ce_loss": 2.959113121032715, "z_loss": 77.63815689086914, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24724845588207245, "tokens_per_second": 2903476.4776357105, "eta_seconds": 9762.110378479958, "data_wait_seconds": 1.2993812561035156e-05}
160
+ {"step": 1590, "total_steps": 28611, "tokens": 1667235840, "loss": 2.9473853707313538, "ce_loss": 2.9394142627716064, "z_loss": 79.71130561828613, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2955326735973358, "tokens_per_second": 2927510.1510600564, "eta_seconds": 9678.385602092743, "data_wait_seconds": 1.4543533325195312e-05}
161
+ {"step": 1600, "total_steps": 28611, "tokens": 1677721600, "loss": 2.9670751690864563, "ce_loss": 2.9602296352386475, "z_loss": 68.45521354675293, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23241528868675232, "tokens_per_second": 2955345.9284553127, "eta_seconds": 9583.678872680663, "data_wait_seconds": 0.2897411584854126}
162
+ {"step": 1610, "total_steps": 28611, "tokens": 1688207360, "loss": 2.932196259498596, "ce_loss": 2.925165295600891, "z_loss": 70.30979919433594, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2494441121816635, "tokens_per_second": 4318983.204561301, "eta_seconds": 6555.385662555695, "data_wait_seconds": 1.621246337890625e-05}
163
+ {"step": 1620, "total_steps": 28611, "tokens": 1698693120, "loss": 2.932806134223938, "ce_loss": 2.925189197063446, "z_loss": 76.16930389404297, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2881036400794983, "tokens_per_second": 4380315.868531358, "eta_seconds": 6461.204092454911, "data_wait_seconds": 1.4066696166992188e-05}
164
+ {"step": 1630, "total_steps": 28611, "tokens": 1709178880, "loss": 2.9200092554092407, "ce_loss": 2.9122785329818726, "z_loss": 77.30753326416016, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2458750605583191, "tokens_per_second": 2873980.434597706, "eta_seconds": 9844.05764055252, "data_wait_seconds": 1.5437602996826172e-05}
165
+ {"step": 1640, "total_steps": 28611, "tokens": 1719664640, "loss": 2.9349562525749207, "ce_loss": 2.9279826879501343, "z_loss": 69.73505401611328, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2804771065711975, "tokens_per_second": 3086341.714698024, "eta_seconds": 9163.322117352485, "data_wait_seconds": 1.341104507446289e-05}
166
+ {"step": 1650, "total_steps": 28611, "tokens": 1730150400, "loss": 2.963089108467102, "ce_loss": 2.955647110939026, "z_loss": 74.41985321044922, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25806912779808044, "tokens_per_second": 2923996.255558559, "eta_seconds": 9668.499910783768, "data_wait_seconds": 0.2651965022087097}
167
+ {"step": 1660, "total_steps": 28611, "tokens": 1740636160, "loss": 2.921950101852417, "ce_loss": 2.91459321975708, "z_loss": 73.56885147094727, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2465110868215561, "tokens_per_second": 3975996.473445949, "eta_seconds": 7107.695382714272, "data_wait_seconds": 1.3649463653564453e-05}
168
+ {"step": 1670, "total_steps": 28611, "tokens": 1751121920, "loss": 2.898268401622772, "ce_loss": 2.8909459710121155, "z_loss": 73.2240982055664, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23333296179771423, "tokens_per_second": 2854754.1449183286, "eta_seconds": 9895.663367819785, "data_wait_seconds": 1.5139579772949219e-05}
169
+ {"step": 1680, "total_steps": 28611, "tokens": 1761607680, "loss": 2.9899385571479797, "ce_loss": 2.982984483242035, "z_loss": 69.54074668884277, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2407737821340561, "tokens_per_second": 4448972.115319213, "eta_seconds": 6347.353843545914, "data_wait_seconds": 1.1444091796875e-05}
170
+ {"step": 1690, "total_steps": 28611, "tokens": 1772093440, "loss": 2.9493258595466614, "ce_loss": 2.942777156829834, "z_loss": 65.4871883392334, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24821516871452332, "tokens_per_second": 2939437.016948101, "eta_seconds": 9603.44254128933, "data_wait_seconds": 1.6570091247558594e-05}
171
+ {"step": 1700, "total_steps": 28611, "tokens": 1782579200, "loss": 2.957255780696869, "ce_loss": 2.950658082962036, "z_loss": 65.976806640625, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27766597270965576, "tokens_per_second": 4372949.715392391, "eta_seconds": 6452.904920601845, "data_wait_seconds": 1.3172626495361328e-05}
172
+ {"step": 1710, "total_steps": 28611, "tokens": 1793064960, "loss": 2.932209312915802, "ce_loss": 2.9254661798477173, "z_loss": 67.43155479431152, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22859248518943787, "tokens_per_second": 2960645.846889241, "eta_seconds": 9527.56406364441, "data_wait_seconds": 1.4722347259521484e-05}
173
+ {"step": 1720, "total_steps": 28611, "tokens": 1803550720, "loss": 2.9224696159362793, "ce_loss": 2.915418565273285, "z_loss": 70.51011848449707, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2504984438419342, "tokens_per_second": 2898400.862198577, "eta_seconds": 9728.556730628014, "data_wait_seconds": 1.6868114471435547e-05}
174
+ {"step": 1730, "total_steps": 28611, "tokens": 1814036480, "loss": 2.938232660293579, "ce_loss": 2.9315191507339478, "z_loss": 67.13576316833496, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22062338888645172, "tokens_per_second": 4396486.637644963, "eta_seconds": 6411.203713130952, "data_wait_seconds": 1.436471939086914e-05}
175
+ {"step": 1740, "total_steps": 28611, "tokens": 1824522240, "loss": 2.919621169567108, "ce_loss": 2.9131917357444763, "z_loss": 64.29403495788574, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24530959129333496, "tokens_per_second": 2885853.503541334, "eta_seconds": 9763.588366985321, "data_wait_seconds": 0.5391177535057068}
176
+ {"step": 1750, "total_steps": 28611, "tokens": 1835008000, "loss": 2.979372203350067, "ce_loss": 2.9720006585121155, "z_loss": 73.71574783325195, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27712076902389526, "tokens_per_second": 4139327.374042438, "eta_seconds": 6804.438835310936, "data_wait_seconds": 1.3530254364013672e-05}
177
+ {"step": 1760, "total_steps": 28611, "tokens": 1845493760, "loss": 2.8947893381118774, "ce_loss": 2.887724816799164, "z_loss": 70.6449966430664, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2661644220352173, "tokens_per_second": 2989807.238400412, "eta_seconds": 9417.100144243239, "data_wait_seconds": 1.728534698486328e-05}
178
+ {"step": 1770, "total_steps": 28611, "tokens": 1855979520, "loss": 2.9153984785079956, "ce_loss": 2.908250331878662, "z_loss": 71.48147964477539, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2487785667181015, "tokens_per_second": 4485078.1062320005, "eta_seconds": 6275.214778733253, "data_wait_seconds": 1.531839370727539e-05}
179
+ {"step": 1780, "total_steps": 28611, "tokens": 1866465280, "loss": 2.968498706817627, "ce_loss": 2.961898624897003, "z_loss": 66.00089836120605, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2343597710132599, "tokens_per_second": 3148904.2287904606, "eta_seconds": 8934.645391488075, "data_wait_seconds": 1.5854835510253906e-05}
180
+ {"step": 1790, "total_steps": 28611, "tokens": 1876951040, "loss": 2.928125262260437, "ce_loss": 2.9212717413902283, "z_loss": 68.53483772277832, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24407486617565155, "tokens_per_second": 4735886.669754975, "eta_seconds": 5938.456482839585, "data_wait_seconds": 1.6689300537109375e-05}
181
+ {"step": 1800, "total_steps": 28611, "tokens": 1887436800, "loss": 2.9194695353507996, "ce_loss": 2.913196861743927, "z_loss": 62.72634696960449, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22601667046546936, "tokens_per_second": 3110888.6466454784, "eta_seconds": 9037.086932158469, "data_wait_seconds": 1.3709068298339844e-05}
182
+ {"step": 1810, "total_steps": 28611, "tokens": 1897922560, "loss": 2.9683371782302856, "ce_loss": 2.9615435004234314, "z_loss": 67.93668556213379, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28459930419921875, "tokens_per_second": 3976944.1956932447, "eta_seconds": 7066.4520277738575, "data_wait_seconds": 1.3649463653564453e-05}
183
+ {"step": 1820, "total_steps": 28611, "tokens": 1908408320, "loss": 2.9356417059898376, "ce_loss": 2.929006576538086, "z_loss": 66.35210227966309, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2596339285373688, "tokens_per_second": 3236476.1807688973, "eta_seconds": 8679.93399207592, "data_wait_seconds": 1.8894672393798828e-05}
184
+ {"step": 1830, "total_steps": 28611, "tokens": 1918894080, "loss": 2.913980782032013, "ce_loss": 2.9076719284057617, "z_loss": 63.08887481689453, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23820985853672028, "tokens_per_second": 4256500.420471596, "eta_seconds": 6597.418320679664, "data_wait_seconds": 1.9848346710205078e-05}
185
+ {"step": 1840, "total_steps": 28611, "tokens": 1929379840, "loss": 2.915424168109894, "ce_loss": 2.90949147939682, "z_loss": 59.32719898223877, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2486334890127182, "tokens_per_second": 3908540.8544879314, "eta_seconds": 7182.073602676392, "data_wait_seconds": 1.5854835510253906e-05}
186
+ {"step": 1850, "total_steps": 28611, "tokens": 1939865600, "loss": 2.9364556670188904, "ce_loss": 2.929768681526184, "z_loss": 66.86966705322266, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23525944352149963, "tokens_per_second": 5326538.307619957, "eta_seconds": 5268.138651299477, "data_wait_seconds": 1.33514404296875e-05}
187
+ {"step": 1860, "total_steps": 28611, "tokens": 1950351360, "loss": 2.9211318492889404, "ce_loss": 2.915030002593994, "z_loss": 61.01839351654053, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2345699965953827, "tokens_per_second": 3274819.4595071143, "eta_seconds": 8565.497097730637, "data_wait_seconds": 1.3589859008789062e-05}
188
+ {"step": 1870, "total_steps": 28611, "tokens": 1960837120, "loss": 2.9257087111473083, "ce_loss": 2.9194642901420593, "z_loss": 62.44354438781738, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2413124442100525, "tokens_per_second": 4178696.6928255255, "eta_seconds": 6710.219208812713, "data_wait_seconds": 1.4066696166992188e-05}
189
+ {"step": 1880, "total_steps": 28611, "tokens": 1971322880, "loss": 2.9313281774520874, "ce_loss": 2.9248605966567993, "z_loss": 64.67623615264893, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2618255913257599, "tokens_per_second": 4442417.37585403, "eta_seconds": 6309.511845588684, "data_wait_seconds": 1.519918441772461e-05}
190
+ {"step": 1890, "total_steps": 28611, "tokens": 1981808640, "loss": 2.91885107755661, "ce_loss": 2.913427412509918, "z_loss": 54.236695289611816, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26607033610343933, "tokens_per_second": 3055546.589404813, "eta_seconds": 9169.881222939492, "data_wait_seconds": 1.6629695892333984e-05}
191
+ {"step": 1900, "total_steps": 28611, "tokens": 1992294400, "loss": 2.8541460037231445, "ce_loss": 2.8479928970336914, "z_loss": 61.530813217163086, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2512604892253876, "tokens_per_second": 4109449.335972149, "eta_seconds": 6815.636657404899, "data_wait_seconds": 1.3053417205810547e-05}
192
+ {"step": 1910, "total_steps": 28611, "tokens": 2002780160, "loss": 2.9161726236343384, "ce_loss": 2.910618484020233, "z_loss": 55.54117774963379, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2438688427209854, "tokens_per_second": 2928469.01292661, "eta_seconds": 9560.636514306068, "data_wait_seconds": 1.4901161193847656e-05}
193
+ {"step": 1920, "total_steps": 28611, "tokens": 2013265920, "loss": 2.936424493789673, "ce_loss": 2.9308801889419556, "z_loss": 55.44287967681885, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27244657278060913, "tokens_per_second": 2396579.2767856317, "eta_seconds": 11678.12068104744, "data_wait_seconds": 1.9669532775878906e-05}
194
+ {"step": 1930, "total_steps": 28611, "tokens": 2023751680, "loss": 2.9334492683410645, "ce_loss": 2.927216351032257, "z_loss": 62.32970905303955, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25655609369277954, "tokens_per_second": 4722171.144042842, "eta_seconds": 5924.617173457146, "data_wait_seconds": 1.2516975402832031e-05}
195
+ {"step": 1940, "total_steps": 28611, "tokens": 2034237440, "loss": 2.9353241324424744, "ce_loss": 2.928929090499878, "z_loss": 63.95041275024414, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24791115522384644, "tokens_per_second": 3104270.1976089184, "eta_seconds": 9009.064519429206, "data_wait_seconds": 1.3828277587890625e-05}
196
+ {"step": 1950, "total_steps": 28611, "tokens": 2044723200, "loss": 2.922396659851074, "ce_loss": 2.9161622524261475, "z_loss": 62.34431171417236, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2766133248806, "tokens_per_second": 3232888.337676394, "eta_seconds": 8647.401894521714, "data_wait_seconds": 0.21261054277420044}
197
+ {"step": 1960, "total_steps": 28611, "tokens": 2055208960, "loss": 2.931930959224701, "ce_loss": 2.926256000995636, "z_loss": 56.749799728393555, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2455151528120041, "tokens_per_second": 4310613.898342528, "eta_seconds": 6482.974266552926, "data_wait_seconds": 1.6391277313232422e-05}
198
+ {"step": 1970, "total_steps": 28611, "tokens": 2065694720, "loss": 2.8867805004119873, "ce_loss": 2.881536900997162, "z_loss": 52.43628120422363, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.242959126830101, "tokens_per_second": 4204315.03762239, "eta_seconds": 6644.391052055359, "data_wait_seconds": 1.615285873413086e-05}
199
+ {"step": 1980, "total_steps": 28611, "tokens": 2076180480, "loss": 2.9056642651557922, "ce_loss": 2.899847984313965, "z_loss": 58.162248611450195, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2555161118507385, "tokens_per_second": 3713699.3619935345, "eta_seconds": 7519.355966663361, "data_wait_seconds": 0.2296193242073059}
200
+ {"step": 1990, "total_steps": 28611, "tokens": 2086666240, "loss": 2.9174944162368774, "ce_loss": 2.9117173552513123, "z_loss": 57.7704381942749, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24746792018413544, "tokens_per_second": 4362277.579860179, "eta_seconds": 6398.983371639251, "data_wait_seconds": 1.436471939086914e-05}
201
+ {"step": 2000, "total_steps": 28611, "tokens": 2097152000, "loss": 2.939990282058716, "ce_loss": 2.934132158756256, "z_loss": 58.581180572509766, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2709045708179474, "tokens_per_second": 3141113.1300882744, "eta_seconds": 8883.365475988388, "data_wait_seconds": 1.239776611328125e-05}
202
+ {"step": 2010, "total_steps": 28611, "tokens": 2107637760, "loss": 2.966218113899231, "ce_loss": 2.9603296518325806, "z_loss": 58.88490581512451, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2552005350589752, "tokens_per_second": 3114955.9658418023, "eta_seconds": 8954.595339989663, "data_wait_seconds": 0.036783456802368164}
203
+ {"step": 2020, "total_steps": 28611, "tokens": 2118123520, "loss": 2.93824702501297, "ce_loss": 2.932889938354492, "z_loss": 53.570746421813965, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2777978777885437, "tokens_per_second": 4797483.552434643, "eta_seconds": 5811.939553570747, "data_wait_seconds": 1.430511474609375e-05}
204
+ {"step": 2030, "total_steps": 28611, "tokens": 2128609280, "loss": 2.9320484399795532, "ce_loss": 2.9263970255851746, "z_loss": 56.514241218566895, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2563401758670807, "tokens_per_second": 3094748.5537086944, "eta_seconds": 9006.288611912729, "data_wait_seconds": 1.4126300811767578e-05}
205
+ {"step": 2040, "total_steps": 28611, "tokens": 2139095040, "loss": 2.890817642211914, "ce_loss": 2.885644018650055, "z_loss": 51.73587703704834, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2275676727294922, "tokens_per_second": 3190973.855697533, "eta_seconds": 8731.413717555999, "data_wait_seconds": 1.430511474609375e-05}
206
+ {"step": 2050, "total_steps": 28611, "tokens": 2149580800, "loss": 2.9147573709487915, "ce_loss": 2.909219980239868, "z_loss": 55.37404823303223, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26050958037376404, "tokens_per_second": 4051481.562779383, "eta_seconds": 6874.331452441215, "data_wait_seconds": 1.4424324035644531e-05}
207
+ {"step": 2060, "total_steps": 28611, "tokens": 2160066560, "loss": 2.8794848918914795, "ce_loss": 2.8744720816612244, "z_loss": 50.12806415557861, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24800731241703033, "tokens_per_second": 4343222.448459342, "eta_seconds": 6410.157827830315, "data_wait_seconds": 1.633167266845703e-05}
208
+ {"step": 2070, "total_steps": 28611, "tokens": 2170552320, "loss": 2.914750397205353, "ce_loss": 2.9090641736984253, "z_loss": 56.861878395080566, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22618132829666138, "tokens_per_second": 3227730.0256990055, "eta_seconds": 8622.237731909752, "data_wait_seconds": 0.22161084413528442}
209
+ {"step": 2080, "total_steps": 28611, "tokens": 2181038080, "loss": 2.9347882866859436, "ce_loss": 2.9291155338287354, "z_loss": 56.727806091308594, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2513333261013031, "tokens_per_second": 3807410.200742238, "eta_seconds": 7306.743531489372, "data_wait_seconds": 1.3232231140136719e-05}
210
+ {"step": 2090, "total_steps": 28611, "tokens": 2191523840, "loss": 2.915239095687866, "ce_loss": 2.909987688064575, "z_loss": 52.51447677612305, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2625187635421753, "tokens_per_second": 2825084.060202817, "eta_seconds": 9843.701462817191, "data_wait_seconds": 1.919269561767578e-05}
211
+ {"step": 2100, "total_steps": 28611, "tokens": 2202009600, "loss": 2.933219850063324, "ce_loss": 2.9285926818847656, "z_loss": 46.27230644226074, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2971017360687256, "tokens_per_second": 2902940.8308659475, "eta_seconds": 9576.081620550156, "data_wait_seconds": 1.52587890625e-05}
212
+ {"step": 2110, "total_steps": 28611, "tokens": 2212495360, "loss": 2.942483603954315, "ce_loss": 2.937111794948578, "z_loss": 53.717230796813965, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2596902847290039, "tokens_per_second": 2645606.6917766053, "eta_seconds": 10503.569053697585, "data_wait_seconds": 1.7940998077392578e-05}
213
+ {"step": 2120, "total_steps": 28611, "tokens": 2222981120, "loss": 2.9119439721107483, "ce_loss": 2.9069466590881348, "z_loss": 49.973328590393066, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2826772630214691, "tokens_per_second": 4170296.604803812, "eta_seconds": 6660.875579929352, "data_wait_seconds": 1.609325408935547e-05}
214
+ {"step": 2130, "total_steps": 28611, "tokens": 2233466880, "loss": 2.9195542335510254, "ce_loss": 2.9146794080734253, "z_loss": 48.748440742492676, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25150352716445923, "tokens_per_second": 2889171.8761235, "eta_seconds": 9610.830454730987, "data_wait_seconds": 1.817941665649414e-05}
215
+ {"step": 2140, "total_steps": 28611, "tokens": 2243952640, "loss": 2.92907178401947, "ce_loss": 2.92364239692688, "z_loss": 54.293681144714355, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24885964393615723, "tokens_per_second": 3203156.4166046144, "eta_seconds": 8665.469832229614, "data_wait_seconds": 0.25210392475128174}
216
+ {"step": 2150, "total_steps": 28611, "tokens": 2254438400, "loss": 2.9178945422172546, "ce_loss": 2.913172662258148, "z_loss": 47.2181510925293, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27642887830734253, "tokens_per_second": 4061285.902457957, "eta_seconds": 6831.917329239845, "data_wait_seconds": 1.4960765838623047e-05}
217
+ {"step": 2160, "total_steps": 28611, "tokens": 2264924160, "loss": 2.8733803033828735, "ce_loss": 2.86843341588974, "z_loss": 49.46894359588623, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27149564027786255, "tokens_per_second": 2941476.480578669, "eta_seconds": 9429.238669466973, "data_wait_seconds": 1.531839370727539e-05}
218
+ {"step": 2170, "total_steps": 28611, "tokens": 2275409920, "loss": 2.9025864601135254, "ce_loss": 2.8974390625953674, "z_loss": 51.47423458099365, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2238880693912506, "tokens_per_second": 3957317.3119500093, "eta_seconds": 7006.109399485587, "data_wait_seconds": 1.3470649719238281e-05}
219
+ {"step": 2180, "total_steps": 28611, "tokens": 2285895680, "loss": 2.8919179439544678, "ce_loss": 2.8874924182891846, "z_loss": 44.25519847869873, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2778443694114685, "tokens_per_second": 3162940.4183293264, "eta_seconds": 8762.38834452629, "data_wait_seconds": 1.7762184143066406e-05}
220
+ {"step": 2190, "total_steps": 28611, "tokens": 2296381440, "loss": 2.9114782214164734, "ce_loss": 2.9059460759162903, "z_loss": 55.32165813446045, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2775927782058716, "tokens_per_second": 2919476.8676083684, "eta_seconds": 9489.517386960983, "data_wait_seconds": 1.4722347259521484e-05}
221
+ {"step": 2200, "total_steps": 28611, "tokens": 2306867200, "loss": 2.904597759246826, "ce_loss": 2.899284243583679, "z_loss": 53.13494396209717, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2704576551914215, "tokens_per_second": 2814732.6535320263, "eta_seconds": 9838.924027562141, "data_wait_seconds": 0.24607038497924805}
222
+ {"step": 2210, "total_steps": 28611, "tokens": 2317352960, "loss": 2.9394798278808594, "ce_loss": 2.934345781803131, "z_loss": 51.340158462524414, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2526310980319977, "tokens_per_second": 4269573.341431005, "eta_seconds": 6483.892595863342, "data_wait_seconds": 1.329183578491211e-05}
223
+ {"step": 2220, "total_steps": 28611, "tokens": 2327838720, "loss": 2.8772756457328796, "ce_loss": 2.87186461687088, "z_loss": 54.1108455657959, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2721002995967865, "tokens_per_second": 4472417.446335626, "eta_seconds": 6187.474570083618, "data_wait_seconds": 1.245737075805664e-05}
224
+ {"step": 2230, "total_steps": 28611, "tokens": 2338324480, "loss": 2.921660006046295, "ce_loss": 2.9170421361923218, "z_loss": 46.178372383117676, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2451232224702835, "tokens_per_second": 4087552.355391353, "eta_seconds": 6767.493367886544, "data_wait_seconds": 1.3768672943115234e-05}
225
+ {"step": 2240, "total_steps": 28611, "tokens": 2348810240, "loss": 2.8983498215675354, "ce_loss": 2.893158197402954, "z_loss": 51.916056632995605, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27676981687545776, "tokens_per_second": 3992333.594255022, "eta_seconds": 6926.274331331253, "data_wait_seconds": 1.3589859008789062e-05}
226
+ {"step": 2250, "total_steps": 28611, "tokens": 2359296000, "loss": 2.9035162329673767, "ce_loss": 2.898995280265808, "z_loss": 45.209174156188965, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24543705582618713, "tokens_per_second": 4172207.4264355022, "eta_seconds": 6625.15285334587, "data_wait_seconds": 1.800060272216797e-05}
227
+ {"step": 2260, "total_steps": 28611, "tokens": 2369781760, "loss": 2.8748847246170044, "ce_loss": 2.869967818260193, "z_loss": 49.16922950744629, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2336546629667282, "tokens_per_second": 4642312.458422557, "eta_seconds": 5951.996213841438, "data_wait_seconds": 0.25335073471069336}
228
+ {"step": 2270, "total_steps": 28611, "tokens": 2380267520, "loss": 2.849588990211487, "ce_loss": 2.8446961641311646, "z_loss": 48.92845916748047, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25017428398132324, "tokens_per_second": 4621884.856912771, "eta_seconds": 5976.0338630437855, "data_wait_seconds": 2.0265579223632812e-05}
229
+ {"step": 2280, "total_steps": 28611, "tokens": 2390753280, "loss": 2.868052899837494, "ce_loss": 2.8629866242408752, "z_loss": 50.662275314331055, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2409016489982605, "tokens_per_second": 4181796.6085246517, "eta_seconds": 6602.43843512535, "data_wait_seconds": 1.3887882232666016e-05}
230
+ {"step": 2290, "total_steps": 28611, "tokens": 2401239040, "loss": 2.893037497997284, "ce_loss": 2.888354182243347, "z_loss": 46.83289909362793, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2598644495010376, "tokens_per_second": 2911616.499478225, "eta_seconds": 9479.122302317619, "data_wait_seconds": 0.2736086845397949}
231
+ {"step": 2300, "total_steps": 28611, "tokens": 2411724800, "loss": 2.892113745212555, "ce_loss": 2.88738614320755, "z_loss": 47.27552890777588, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23829872906208038, "tokens_per_second": 3825259.4317818987, "eta_seconds": 7212.343013072014, "data_wait_seconds": 1.9609928131103516e-05}
232
+ {"step": 2310, "total_steps": 28611, "tokens": 2422210560, "loss": 2.852467656135559, "ce_loss": 2.8480140566825867, "z_loss": 44.53614807128906, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22052299976348877, "tokens_per_second": 3461574.8874480594, "eta_seconds": 7967.0665153026575, "data_wait_seconds": 0.24314409494400024}
233
+ {"step": 2320, "total_steps": 28611, "tokens": 2432696320, "loss": 2.8921367526054382, "ce_loss": 2.8876957297325134, "z_loss": 44.41061782836914, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24811512231826782, "tokens_per_second": 3926020.304669522, "eta_seconds": 7021.897360849381, "data_wait_seconds": 1.3589859008789062e-05}
234
+ {"step": 2330, "total_steps": 28611, "tokens": 2443182080, "loss": 2.89721941947937, "ce_loss": 2.892128050327301, "z_loss": 50.913655281066895, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28163769841194153, "tokens_per_second": 2874788.9759244528, "eta_seconds": 9585.964774036407, "data_wait_seconds": 1.7583370208740234e-05}
235
+ {"step": 2340, "total_steps": 28611, "tokens": 2453667840, "loss": 2.872483253479004, "ce_loss": 2.8681777715682983, "z_loss": 43.05439853668213, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23040476441383362, "tokens_per_second": 4385021.681204319, "eta_seconds": 6282.09894926548, "data_wait_seconds": 2.4557113647460938e-05}
236
+ {"step": 2350, "total_steps": 28611, "tokens": 2464153600, "loss": 2.8663591742515564, "ce_loss": 2.8619529008865356, "z_loss": 44.062068939208984, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.21598312258720398, "tokens_per_second": 3020417.89910661, "eta_seconds": 9116.835900139808, "data_wait_seconds": 1.436471939086914e-05}
237
+ {"step": 2360, "total_steps": 28611, "tokens": 2474639360, "loss": 2.868233859539032, "ce_loss": 2.863285720348358, "z_loss": 49.4813346862793, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25207433104515076, "tokens_per_second": 2727491.6787601965, "eta_seconds": 10092.118260288238, "data_wait_seconds": 2.0325183868408203e-05}
238
+ {"step": 2370, "total_steps": 28611, "tokens": 2485125120, "loss": 2.8806788325309753, "ce_loss": 2.876452147960663, "z_loss": 42.26672077178955, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2509712278842926, "tokens_per_second": 4093435.5512737404, "eta_seconds": 6721.904491066933, "data_wait_seconds": 1.2695789337158203e-05}
239
+ {"step": 2380, "total_steps": 28611, "tokens": 2495610880, "loss": 2.864222764968872, "ce_loss": 2.8596368432044983, "z_loss": 45.85894584655762, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25795236229896545, "tokens_per_second": 2733924.1218536505, "eta_seconds": 10060.702429938316, "data_wait_seconds": 2.1219253540039062e-05}
240
+ {"step": 2390, "total_steps": 28611, "tokens": 2506096640, "loss": 2.8560163974761963, "ce_loss": 2.851844072341919, "z_loss": 41.72358417510986, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2925208508968353, "tokens_per_second": 4324774.452421297, "eta_seconds": 6357.490222549439, "data_wait_seconds": 1.3053417205810547e-05}
241
+ {"step": 2400, "total_steps": 28611, "tokens": 2516582400, "loss": 2.884136199951172, "ce_loss": 2.8800997138023376, "z_loss": 40.36493110656738, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27305546402931213, "tokens_per_second": 2984032.346557699, "eta_seconds": 9210.431504774095, "data_wait_seconds": 1.621246337890625e-05}
242
+ {"step": 2410, "total_steps": 28611, "tokens": 2527068160, "loss": 2.8297327756881714, "ce_loss": 2.8253036737442017, "z_loss": 44.29108238220215, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23517559468746185, "tokens_per_second": 2863015.3502375307, "eta_seconds": 9596.085390782357, "data_wait_seconds": 2.0802021026611328e-05}
243
+ {"step": 2420, "total_steps": 28611, "tokens": 2537553920, "loss": 2.907247304916382, "ce_loss": 2.9026878476142883, "z_loss": 45.594648361206055, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2795132100582123, "tokens_per_second": 4069172.7167944955, "eta_seconds": 6749.1001064300535, "data_wait_seconds": 1.2755393981933594e-05}
244
+ {"step": 2430, "total_steps": 28611, "tokens": 2548039680, "loss": 2.8687278032302856, "ce_loss": 2.8646257519721985, "z_loss": 41.02114295959473, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2755889892578125, "tokens_per_second": 3174424.564837766, "eta_seconds": 8648.108561182024, "data_wait_seconds": 0.2570886015892029}
245
+ {"step": 2440, "total_steps": 28611, "tokens": 2558525440, "loss": 2.861086368560791, "ce_loss": 2.8561583161354065, "z_loss": 49.27982139587402, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.257871150970459, "tokens_per_second": 4442734.646088611, "eta_seconds": 6176.889839720726, "data_wait_seconds": 1.2695789337158203e-05}
246
+ {"step": 2450, "total_steps": 28611, "tokens": 2569011200, "loss": 2.865040898323059, "ce_loss": 2.860712766647339, "z_loss": 43.281307220458984, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23377303779125214, "tokens_per_second": 3266498.4547456973, "eta_seconds": 8397.921234631538, "data_wait_seconds": 1.3828277587890625e-05}
247
+ {"step": 2460, "total_steps": 28611, "tokens": 2579496960, "loss": 2.8810497522354126, "ce_loss": 2.876833140850067, "z_loss": 42.16642379760742, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24177245795726776, "tokens_per_second": 2908941.9593534227, "eta_seconds": 9426.5583016634, "data_wait_seconds": 2.1219253540039062e-05}
248
+ {"step": 2470, "total_steps": 28611, "tokens": 2589982720, "loss": 2.909602463245392, "ce_loss": 2.905421793460846, "z_loss": 41.80679416656494, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2331513613462448, "tokens_per_second": 4558894.805292781, "eta_seconds": 6012.60313885212, "data_wait_seconds": 1.5556812286376953e-05}
249
+ {"step": 2480, "total_steps": 28611, "tokens": 2600468480, "loss": 2.9014490842819214, "ce_loss": 2.8972721099853516, "z_loss": 41.76929187774658, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2812369167804718, "tokens_per_second": 3067154.284949149, "eta_seconds": 8933.472825431823, "data_wait_seconds": 1.3649463653564453e-05}
250
+ {"step": 2490, "total_steps": 28611, "tokens": 2610954240, "loss": 2.884644865989685, "ce_loss": 2.8804200291633606, "z_loss": 42.248165130615234, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2626129984855652, "tokens_per_second": 2939040.4228751957, "eta_seconds": 9319.318469667434, "data_wait_seconds": 1.7940998077392578e-05}
251
+ {"step": 2500, "total_steps": 28611, "tokens": 2621440000, "loss": 2.8514214158058167, "ce_loss": 2.8477457761764526, "z_loss": 36.75676918029785, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24733257293701172, "tokens_per_second": 4179007.9864996094, "eta_seconds": 6551.642883777618, "data_wait_seconds": 1.537799835205078e-05}
252
+ {"step": 2510, "total_steps": 28611, "tokens": 2631925760, "loss": 2.8955317735671997, "ce_loss": 2.8911412358283997, "z_loss": 43.90542411804199, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28378432989120483, "tokens_per_second": 3453333.5828459547, "eta_seconds": 7925.351408839227, "data_wait_seconds": 0.5053586363792419}
253
+ {"step": 2520, "total_steps": 28611, "tokens": 2642411520, "loss": 2.885873794555664, "ce_loss": 2.881267547607422, "z_loss": 46.06220054626465, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25472599267959595, "tokens_per_second": 6855258.332961583, "eta_seconds": 3990.8629386663433, "data_wait_seconds": 1.2695789337158203e-05}
254
+ {"step": 2530, "total_steps": 28611, "tokens": 2652897280, "loss": 2.89236319065094, "ce_loss": 2.8880324363708496, "z_loss": 43.307541847229004, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2975749969482422, "tokens_per_second": 4169054.916246772, "eta_seconds": 6559.73864710331, "data_wait_seconds": 1.3828277587890625e-05}
255
+ {"step": 2540, "total_steps": 28611, "tokens": 2663383040, "loss": 2.8553766012191772, "ce_loss": 2.851294159889221, "z_loss": 40.82407283782959, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24416594207286835, "tokens_per_second": 4074149.204830456, "eta_seconds": 6709.9714619159695, "data_wait_seconds": 1.3709068298339844e-05}
256
+ {"step": 2550, "total_steps": 28611, "tokens": 2673868800, "loss": 2.879975199699402, "ce_loss": 2.875555396080017, "z_loss": 44.19819164276123, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2547118067741394, "tokens_per_second": 4012617.725568681, "eta_seconds": 6810.2523103237145, "data_wait_seconds": 1.609325408935547e-05}
257
+ {"step": 2560, "total_steps": 28611, "tokens": 2684354560, "loss": 2.8685178756713867, "ce_loss": 2.8647612929344177, "z_loss": 37.56584644317627, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25409382581710815, "tokens_per_second": 3272776.5770973694, "eta_seconds": 8346.568344187737, "data_wait_seconds": 0.2618451714515686}
258
+ {"step": 2570, "total_steps": 28611, "tokens": 2694840320, "loss": 2.897736370563507, "ce_loss": 2.8940454721450806, "z_loss": 36.90924549102783, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.28030967712402344, "tokens_per_second": 3701328.7453814885, "eta_seconds": 7377.341893792152, "data_wait_seconds": 1.2695789337158203e-05}
259
+ {"step": 2580, "total_steps": 28611, "tokens": 2705326080, "loss": 2.9255489706993103, "ce_loss": 2.9217370748519897, "z_loss": 38.1191520690918, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2679749131202698, "tokens_per_second": 2868768.915282849, "eta_seconds": 9514.702181339264, "data_wait_seconds": 1.4424324035644531e-05}
260
+ {"step": 2590, "total_steps": 28611, "tokens": 2715811840, "loss": 2.891103506088257, "ce_loss": 2.887182652950287, "z_loss": 39.208168029785156, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24055811762809753, "tokens_per_second": 2909235.017600752, "eta_seconds": 9378.752809906005, "data_wait_seconds": 1.6808509826660156e-05}
261
+ {"step": 2600, "total_steps": 28611, "tokens": 2726297600, "loss": 2.878664970397949, "ce_loss": 2.8747215270996094, "z_loss": 39.434828758239746, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2539691925048828, "tokens_per_second": 2653538.754019257, "eta_seconds": 10278.542303061486, "data_wait_seconds": 1.6748905181884766e-05}
262
+ {"step": 2610, "total_steps": 28611, "tokens": 2736783360, "loss": 2.89937686920166, "ce_loss": 2.8954153656959534, "z_loss": 39.614529609680176, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2539970278739929, "tokens_per_second": 4146752.4813817684, "eta_seconds": 6574.789476442337, "data_wait_seconds": 1.3887882232666016e-05}
263
+ {"step": 2620, "total_steps": 28611, "tokens": 2747269120, "loss": 2.9078510999679565, "ce_loss": 2.90363872051239, "z_loss": 42.1237268447876, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26248204708099365, "tokens_per_second": 3024609.879726608, "eta_seconds": 9010.596374320983, "data_wait_seconds": 1.633167266845703e-05}
264
+ {"step": 2630, "total_steps": 28611, "tokens": 2757754880, "loss": 2.8824328184127808, "ce_loss": 2.8784485459327698, "z_loss": 39.843048095703125, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23669388890266418, "tokens_per_second": 4028192.387569925, "eta_seconds": 6763.096305942535, "data_wait_seconds": 1.3768672943115234e-05}
265
+ {"step": 2640, "total_steps": 28611, "tokens": 2768240640, "loss": 2.85481721162796, "ce_loss": 2.8510677814483643, "z_loss": 37.494038581848145, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.266274094581604, "tokens_per_second": 2770845.214841871, "eta_seconds": 9828.252819800377, "data_wait_seconds": 1.71661376953125e-05}
266
+ {"step": 2650, "total_steps": 28611, "tokens": 2778726400, "loss": 2.845838189125061, "ce_loss": 2.8423155546188354, "z_loss": 35.2260799407959, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2837708592414856, "tokens_per_second": 3941053.035998155, "eta_seconds": 6907.311646747589, "data_wait_seconds": 1.2516975402832031e-05}
267
+ {"step": 2660, "total_steps": 28611, "tokens": 2789212160, "loss": 2.862060785293579, "ce_loss": 2.858495056629181, "z_loss": 35.6576566696167, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2872838079929352, "tokens_per_second": 4072445.5314818565, "eta_seconds": 6681.880841779709, "data_wait_seconds": 1.901388168334961e-05}
268
+ {"step": 2670, "total_steps": 28611, "tokens": 2799697920, "loss": 2.8580577969551086, "ce_loss": 2.8548039197921753, "z_loss": 32.53841733932495, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2533966600894928, "tokens_per_second": 2778762.7882890096, "eta_seconds": 9788.928414702415, "data_wait_seconds": 1.3768672943115234e-05}
269
+ {"step": 2680, "total_steps": 28611, "tokens": 2810183680, "loss": 2.8989609479904175, "ce_loss": 2.8953737020492554, "z_loss": 35.8725004196167, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23133718967437744, "tokens_per_second": 2886930.7971041463, "eta_seconds": 9418.523049902915, "data_wait_seconds": 1.4185905456542969e-05}
270
+ {"step": 2690, "total_steps": 28611, "tokens": 2820669440, "loss": 2.8747634291648865, "ce_loss": 2.870830237865448, "z_loss": 39.33172416687012, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24898946285247803, "tokens_per_second": 2766493.097650072, "eta_seconds": 9824.762808585167, "data_wait_seconds": 1.6927719116210938e-05}
271
+ {"step": 2700, "total_steps": 28611, "tokens": 2831155200, "loss": 2.85181200504303, "ce_loss": 2.8483211398124695, "z_loss": 34.90819072723389, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2425118386745453, "tokens_per_second": 4448078.498050068, "eta_seconds": 6108.17744064331, "data_wait_seconds": 1.6391277313232422e-05}
272
+ {"step": 2710, "total_steps": 28611, "tokens": 2841640960, "loss": 2.870373010635376, "ce_loss": 2.8657811880111694, "z_loss": 45.91804599761963, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26494523882865906, "tokens_per_second": 2925828.8354373067, "eta_seconds": 9282.554962563514, "data_wait_seconds": 1.537799835205078e-05}
273
+ {"step": 2720, "total_steps": 28611, "tokens": 2852126720, "loss": 2.8625608086586, "ce_loss": 2.8590834736824036, "z_loss": 34.77354431152344, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.23839662969112396, "tokens_per_second": 3938568.748480627, "eta_seconds": 6893.032202744484, "data_wait_seconds": 1.3649463653564453e-05}
274
+ {"step": 2730, "total_steps": 28611, "tokens": 2862612480, "loss": 2.8670960068702698, "ce_loss": 2.86376291513443, "z_loss": 33.33061218261719, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2752220034599304, "tokens_per_second": 2874850.048285596, "eta_seconds": 9439.864688658716, "data_wait_seconds": 1.633167266845703e-05}
275
+ {"step": 2740, "total_steps": 28611, "tokens": 2873098240, "loss": 2.862023890018463, "ce_loss": 2.8586215376853943, "z_loss": 34.02348518371582, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27385133504867554, "tokens_per_second": 3858014.388455483, "eta_seconds": 7031.52113096714, "data_wait_seconds": 1.2576580047607422e-05}
276
+ {"step": 2750, "total_steps": 28611, "tokens": 2883584000, "loss": 2.84933602809906, "ce_loss": 2.845423102378845, "z_loss": 39.129271507263184, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.249404177069664, "tokens_per_second": 2743382.1509036347, "eta_seconds": 9884.595890903473, "data_wait_seconds": 1.7821788787841797e-05}
277
+ {"step": 2760, "total_steps": 28611, "tokens": 2894069760, "loss": 2.8956176042556763, "ce_loss": 2.8916420936584473, "z_loss": 39.75528144836426, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27487966418266296, "tokens_per_second": 4260710.187488642, "eta_seconds": 6362.0234616279595, "data_wait_seconds": 1.2636184692382812e-05}
278
+ {"step": 2770, "total_steps": 28611, "tokens": 2904555520, "loss": 2.8636621236801147, "ce_loss": 2.859818935394287, "z_loss": 38.431949615478516, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24970781803131104, "tokens_per_second": 2839098.236991574, "eta_seconds": 9543.964369726182, "data_wait_seconds": 1.7583370208740234e-05}
279
+ {"step": 2780, "total_steps": 28611, "tokens": 2915041280, "loss": 2.844648241996765, "ce_loss": 2.8404226303100586, "z_loss": 42.25680351257324, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26306599378585815, "tokens_per_second": 4205145.149217491, "eta_seconds": 6441.101482796668, "data_wait_seconds": 1.3232231140136719e-05}
280
+ {"step": 2790, "total_steps": 28611, "tokens": 2925527040, "loss": 2.8556094765663147, "ce_loss": 2.8513877391815186, "z_loss": 42.21787357330322, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25884318351745605, "tokens_per_second": 2737572.4743198124, "eta_seconds": 9890.251728487015, "data_wait_seconds": 1.5079975128173828e-05}
281
+ {"step": 2800, "total_steps": 28611, "tokens": 2936012800, "loss": 2.8349839448928833, "ce_loss": 2.830768883228302, "z_loss": 42.15066146850586, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2999068796634674, "tokens_per_second": 3916034.720297195, "eta_seconds": 6911.275580811501, "data_wait_seconds": 1.4424324035644531e-05}
282
+ {"step": 2810, "total_steps": 28611, "tokens": 2946498560, "loss": 2.8636972904205322, "ce_loss": 2.860248863697052, "z_loss": 34.48423480987549, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.24497531354427338, "tokens_per_second": 2856016.973252755, "eta_seconds": 9472.7411039114, "data_wait_seconds": 1.9252300262451172e-05}
283
+ {"step": 2820, "total_steps": 28611, "tokens": 2956984320, "loss": 2.8709317445755005, "ce_loss": 2.8676512241363525, "z_loss": 32.8049521446228, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.27765321731567383, "tokens_per_second": 2917012.476963771, "eta_seconds": 9271.068886256218, "data_wait_seconds": 1.233816146850586e-05}
284
+ {"step": 2830, "total_steps": 28611, "tokens": 2967470080, "loss": 2.8567874431610107, "ce_loss": 2.853254973888397, "z_loss": 35.32486820220947, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.22734446823596954, "tokens_per_second": 2987921.8789006304, "eta_seconds": 9047.538373374939, "data_wait_seconds": 1.3947486877441406e-05}
285
+ {"step": 2840, "total_steps": 28611, "tokens": 2977955840, "loss": 2.870243728160858, "ce_loss": 2.866898000240326, "z_loss": 33.45742225646973, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.2174462527036667, "tokens_per_second": 2770542.546940766, "eta_seconds": 9753.63187468052, "data_wait_seconds": 1.3947486877441406e-05}
286
+ {"step": 2850, "total_steps": 28611, "tokens": 2988441600, "loss": 2.896151125431061, "ce_loss": 2.8922274112701416, "z_loss": 39.23710060119629, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.25698357820510864, "tokens_per_second": 2863488.8208924914, "eta_seconds": 9433.375866150856, "data_wait_seconds": 1.33514404296875e-05}
287
+ {"step": 2860, "total_steps": 28611, "tokens": 2998927360, "loss": 2.855818212032318, "ce_loss": 2.8523313999176025, "z_loss": 34.86809253692627, "muon_lr": 0.03, "adamw_lr": 0.003, "grad_norm": 0.26376277208328247, "tokens_per_second": 2760942.2278659744, "eta_seconds": 9779.951316428183, "data_wait_seconds": 1.710653305053711e-05}