Hugging Face's logo

dotlabs
/
void.1

Text Generation
Transformers
TensorBoard
Safetensors
English
qwen3
byte-level
pretraining
symbolic
text-generation-inference
Model card Files
xet
 Metrics Community
3
appvoid commited on
Commit
847f7ff
·
verified ·
1 Parent(s): faeadf9

Update TensorBoard training metrics

Browse files
Files changed (1) hide show
  1. benchmark_cache/shard_scheduler.json +29 -23
benchmark_cache/shard_scheduler.json CHANGED
@@ -22,8 +22,8 @@
22
  "tokens_per_second": 1062967.6168812835,
23
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
24
  "estimated_pages_for_buffer": 1,
25
- "required_token_positions_per_second": 75740.6742637023,
26
- "required_pages_per_second_upper_estimate": 0.006565009682193805
27
  },
28
  {
29
  "source": "rewrite6",
@@ -36,8 +36,8 @@
36
  "tokens_per_second": 1206672.263377472,
37
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
38
  "estimated_pages_for_buffer": 1,
39
- "required_token_positions_per_second": 9467.584282962787,
40
- "required_pages_per_second_upper_estimate": 0.0037767609234732676
41
  },
42
  {
43
  "source": "openmath",
@@ -50,8 +50,8 @@
50
  "tokens_per_second": 410042.7581134458,
51
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
52
  "estimated_pages_for_buffer": 1,
53
- "required_token_positions_per_second": 18935.168565925575,
54
- "required_pages_per_second_upper_estimate": 0.00438031800014194
55
  },
56
  {
57
  "source": "ultra_qa",
@@ -64,8 +64,8 @@
64
  "tokens_per_second": 2163486.5624068812,
65
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
66
  "estimated_pages_for_buffer": 1,
67
- "required_token_positions_per_second": 28402.75284888836,
68
- "required_pages_per_second_upper_estimate": 0.0017295786307698
69
  },
70
  {
71
  "source": "ultra_style",
@@ -78,8 +78,8 @@
78
  "tokens_per_second": 1546694.2737633225,
79
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
80
  "estimated_pages_for_buffer": 1,
81
- "required_token_positions_per_second": 28402.75284888836,
82
- "required_pages_per_second_upper_estimate": 0.002186522538159462
83
  },
84
  {
85
  "source": "cortex",
@@ -92,8 +92,8 @@
92
  "tokens_per_second": 1015177.8395898147,
93
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
94
  "estimated_pages_for_buffer": 22,
95
- "required_token_positions_per_second": 28402.75284888836,
96
- "required_pages_per_second_upper_estimate": 0.35266713248430365
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
@@ -158,24 +158,24 @@
158
  "prefetch_batches": 62,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 690670,
161
- "producer_batch_seconds": 0.19338228726567763,
162
- "producer_p95_seconds": 0.18032291599956807,
163
- "gpu_update_seconds": 0.9734267712392568,
164
- "producer_batches_per_second": 5.171104417780276,
165
- "gpu_batches_per_second": 1.0272986418145384,
166
- "producer_headroom": 5.033691477140911,
167
  "sustainable": true,
168
- "buffer_seconds": 60.35245981683392,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
- "upload_pages_per_commit": 75,
172
  "max_pending_pages": 4096,
173
- "generation_bytes_per_second": 382597.2306414942,
174
- "upload_bytes_per_second": 3426178.087227453,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
- "estimated_full_shards_per_hour": 82.09645928796405,
179
  "selection_policy": "Smallest shard within 90% of measured local peak, enlarged to amortize 30 seconds of measured page production; bounded to 256 MiB."
180
  },
181
  "upload_measurements": [
@@ -202,6 +202,12 @@
202
  "bytes": 7732243,
203
  "pages": 17,
204
  "shards": 3
 
 
 
 
 
 
205
  }
206
  ],
207
  "cache_key": {
 
22
  "tokens_per_second": 1062967.6168812835,
23
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
24
  "estimated_pages_for_buffer": 1,
25
+ "required_token_positions_per_second": 75609.08177184862,
26
+ "required_pages_per_second_upper_estimate": 0.006553603578518049
27
  },
28
  {
29
  "source": "rewrite6",
 
36
  "tokens_per_second": 1206672.263377472,
37
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
38
  "estimated_pages_for_buffer": 1,
39
+ "required_token_positions_per_second": 9451.135221481078,
40
+ "required_pages_per_second_upper_estimate": 0.0037701991469128284
41
  },
42
  {
43
  "source": "openmath",
 
50
  "tokens_per_second": 410042.7581134458,
51
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
52
  "estimated_pages_for_buffer": 1,
53
+ "required_token_positions_per_second": 18902.270442962155,
54
+ "required_pages_per_second_upper_estimate": 0.004372707598381542
55
  },
56
  {
57
  "source": "ultra_qa",
 
64
  "tokens_per_second": 2163486.5624068812,
65
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
66
  "estimated_pages_for_buffer": 1,
67
+ "required_token_positions_per_second": 28353.405664443235,
68
+ "required_pages_per_second_upper_estimate": 0.0017265736461417594
69
  },
70
  {
71
  "source": "ultra_style",
 
78
  "tokens_per_second": 1546694.2737633225,
79
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
80
  "estimated_pages_for_buffer": 1,
81
+ "required_token_positions_per_second": 28353.405664443235,
82
+ "required_pages_per_second_upper_estimate": 0.0021827236552991267
83
  },
84
  {
85
  "source": "cortex",
 
92
  "tokens_per_second": 1015177.8395898147,
93
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
94
  "estimated_pages_for_buffer": 22,
95
+ "required_token_positions_per_second": 28353.405664443235,
96
+ "required_pages_per_second_upper_estimate": 0.3520544056078974
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
 
158
  "prefetch_batches": 62,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 690670,
161
+ "producer_batch_seconds": 0.07287284228125657,
162
+ "producer_p95_seconds": 0.1761082670000178,
163
+ "gpu_update_seconds": 0.9751209546820736,
164
+ "producer_batches_per_second": 13.722533233168638,
165
+ "gpu_batches_per_second": 1.0255138044141794,
166
+ "producer_headroom": 13.381129706983886,
167
  "sustainable": true,
168
+ "buffer_seconds": 60.457499190288566,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
+ "upload_pages_per_commit": 79,
172
  "max_pending_pages": 4096,
173
+ "generation_bytes_per_second": 382834.16564839584,
174
+ "upload_bytes_per_second": 4501863.534696111,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
+ "estimated_full_shards_per_hour": 82.14730002488047,
179
  "selection_policy": "Smallest shard within 90% of measured local peak, enlarged to amortize 30 seconds of measured page production; bounded to 256 MiB."
180
  },
181
  "upload_measurements": [
 
202
  "bytes": 7732243,
203
  "pages": 17,
204
  "shards": 3
205
+ },
206
+ {
207
+ "seconds": 1.3709215139997468,
208
+ "bytes": 10777643,
209
+ "pages": 39,
210
+ "shards": 3
211
  }
212
  ],
213
  "cache_key": {