appvoid commited on
Commit
70eb55c
·
verified ·
1 Parent(s): dfd57f5

Qwen90M step 557164; complete resumable checkpoint

Browse files
Files changed (6) hide show
  1. complete.json +5 -5
  2. coverage.json +47 -47
  3. latest.json +1 -1
  4. model.safetensors +1 -1
  5. shard_scheduler.json +125 -125
  6. trainer.pt +2 -2
complete.json CHANGED
@@ -1,18 +1,18 @@
1
  {
2
- "step": 555000,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
- "model.safetensors": "bcae9326cfbbc3209e19608c2763a0085cdbfb1dc180d042030dade3f1ff469b",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
- "trainer.pt": "4c3e5c3949b8f4716b37f94e501fd3bf9bfc15652bfd259b8ceeb75ba134374a",
12
  "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6",
13
  "performance.json": "3070ebbdf9b2b0fd6a0a237819043aa2c2a35c74a97edfa223e5c324471af4af",
14
- "shard_scheduler.json": "80a4f3be32aeb96381dd50c3cd54fbb71414db41f241a6d17b6f8416b766ca04",
15
- "coverage.json": "16acad3ee0fe0340416767bf5ff3a94ee49c9a3a058f03397d274c129f8e1bcf",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063"
 
1
  {
2
+ "step": 557164,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
+ "model.safetensors": "9aeb84d48b1a3e7876ab13cf2a9db637ae3eea5a8e14c759a56dfd289a32ce86",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
+ "trainer.pt": "625680f99c72ef12793fea5968b6c72147fca5bf0986649c268a6a0699e0b367",
12
  "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6",
13
  "performance.json": "3070ebbdf9b2b0fd6a0a237819043aa2c2a35c74a97edfa223e5c324471af4af",
14
+ "shard_scheduler.json": "21879a49824847fe89e795ae96e1fd2dc476cf1541cf6501892fb7519e8c9036",
15
+ "coverage.json": "73df3c9242608e741c41ad8c77453ef20d1273e9c95e5fe1751e37fda2234e71",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063"
coverage.json CHANGED
@@ -1,107 +1,107 @@
1
  {
2
- "step": 555000,
3
  "stats": {
4
- "climbmix/blocks": 19980000,
5
- "climbmix/targets": 33398363172,
6
- "climbmix/input_tokens": 33400479609,
7
- "ultra_qa/blocks": 7492500,
8
- "ultra_qa/targets": 12223124541,
9
- "ultra_qa/input_tokens": 12223148113,
10
- "cortex/blocks": 7492500,
11
- "cortex/targets": 284926292,
12
- "cortex/input_tokens": 11909982208,
13
- "ultra_style/blocks": 7492500,
14
- "ultra_style/targets": 11850383023,
15
- "ultra_style/input_tokens": 11851004388,
16
- "openmath/blocks": 4995000,
17
- "openmath/targets": 7718088789,
18
- "openmath/input_tokens": 7720659344,
19
- "rewrite6/blocks": 2497500,
20
- "rewrite6/targets": 4094322151,
21
- "rewrite6/input_tokens": 4098811046
22
  },
23
  "sources": {
24
  "climbmix": {
25
- "page": 2736,
26
- "row": 2897,
27
  "start": {
28
  "file": 132,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
- "row_group": 33,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
- "offset": 10240,
37
- "documents": 11209553
38
  },
39
  "rewrite6": {
40
- "page": 6,
41
- "row": 2921,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
- "row": 24576
47
  }
48
  },
49
  "epoch": 31,
50
  "offset": 0,
51
- "documents": 6395021
52
  },
53
  "openmath": {
54
- "page": 1709,
55
- "row": 3426,
56
  "start": {
57
  "file": 16,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
- "row_group": 13,
61
- "row": 664
62
  }
63
  },
64
  "epoch": 0,
65
- "offset": 2048,
66
- "documents": 7003490
67
  },
68
  "ultra_qa": {
69
- "page": 739,
70
- "row": 1676,
71
  "start": {
72
  "file": 5,
73
  "reader": {
74
  "format": "parquet-row-v1",
75
  "row_group": 6,
76
- "row": 41706
77
  }
78
  },
79
  "epoch": 0,
80
  "offset": 2048,
81
- "documents": 3028620
82
  },
83
  "ultra_style": {
84
- "page": 1128,
85
- "row": 2292,
86
  "start": {
87
  "file": 6,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
- "row_group": 5,
91
- "row": 84382
92
  }
93
  },
94
  "epoch": 0,
95
- "offset": 2048,
96
- "documents": 4622580
97
  },
98
  "cortex": {
99
- "page": 145310,
100
- "row": 26,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
- "documents": 18599706
105
  }
106
  }
107
  }
 
1
  {
2
+ "step": 557164,
3
  "stats": {
4
+ "climbmix/blocks": 20057904,
5
+ "climbmix/targets": 33528559976,
6
+ "climbmix/input_tokens": 33530684873,
7
+ "ultra_qa/blocks": 7521714,
8
+ "ultra_qa/targets": 12270782181,
9
+ "ultra_qa/input_tokens": 12270805843,
10
+ "cortex/blocks": 7521714,
11
+ "cortex/targets": 286034884,
12
+ "cortex/input_tokens": 11956373662,
13
+ "ultra_style/blocks": 7521714,
14
+ "ultra_style/targets": 11896892314,
15
+ "ultra_style/input_tokens": 11897516222,
16
+ "openmath/blocks": 5014476,
17
+ "openmath/targets": 7748188395,
18
+ "openmath/input_tokens": 7750768917,
19
+ "rewrite6/blocks": 2507238,
20
+ "rewrite6/targets": 4110276401,
21
+ "rewrite6/input_tokens": 4114782547
22
  },
23
  "sources": {
24
  "climbmix": {
25
+ "page": 2747,
26
+ "row": 2075,
27
  "start": {
28
  "file": 132,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
+ "row_group": 77,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
+ "offset": 2048,
37
+ "documents": 11253787
38
  },
39
  "rewrite6": {
40
+ "page": 12,
41
+ "row": 3016,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
+ "row": 49152
47
  }
48
  },
49
  "epoch": 31,
50
  "offset": 0,
51
+ "documents": 6419692
52
  },
53
  "openmath": {
54
+ "page": 1716,
55
+ "row": 2073,
56
  "start": {
57
  "file": 16,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
+ "row_group": 42,
61
+ "row": 336
62
  }
63
  },
64
  "epoch": 0,
65
+ "offset": 0,
66
+ "documents": 7030809
67
  },
68
  "ultra_qa": {
69
+ "page": 742,
70
+ "row": 1162,
71
  "start": {
72
  "file": 5,
73
  "reader": {
74
  "format": "parquet-row-v1",
75
  "row_group": 6,
76
+ "row": 53994
77
  }
78
  },
79
  "epoch": 0,
80
  "offset": 2048,
81
+ "documents": 3040394
82
  },
83
  "ultra_style": {
84
+ "page": 1132,
85
+ "row": 4042,
86
  "start": {
87
  "file": 6,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
+ "row_group": 6,
91
+ "row": 15848
92
  }
93
  },
94
  "epoch": 0,
95
+ "offset": 0,
96
+ "documents": 4640714
97
  },
98
  "cortex": {
99
+ "page": 145876,
100
+ "row": 97,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
+ "documents": 18672225
105
  }
106
  }
107
  }
latest.json CHANGED
@@ -1 +1 @@
1
- {"step": 555000, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "bcae9326cfbbc3209e19608c2763a0085cdbfb1dc180d042030dade3f1ff469b", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "4c3e5c3949b8f4716b37f94e501fd3bf9bfc15652bfd259b8ceeb75ba134374a", "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6", "performance.json": "3070ebbdf9b2b0fd6a0a237819043aa2c2a35c74a97edfa223e5c324471af4af", "shard_scheduler.json": "80a4f3be32aeb96381dd50c3cd54fbb71414db41f241a6d17b6f8416b766ca04", "coverage.json": "16acad3ee0fe0340416767bf5ff3a94ee49c9a3a058f03397d274c129f8e1bcf", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063", "folder": "step-000000555000"}
 
1
+ {"step": 557164, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "9aeb84d48b1a3e7876ab13cf2a9db637ae3eea5a8e14c759a56dfd289a32ce86", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "625680f99c72ef12793fea5968b6c72147fca5bf0986649c268a6a0699e0b367", "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6", "performance.json": "3070ebbdf9b2b0fd6a0a237819043aa2c2a35c74a97edfa223e5c324471af4af", "shard_scheduler.json": "21879a49824847fe89e795ae96e1fd2dc476cf1541cf6501892fb7519e8c9036", "coverage.json": "73df3c9242608e741c41ad8c77453ef20d1273e9c95e5fe1751e37fda2234e71", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063", "folder": "step-000000557164"}
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bcae9326cfbbc3209e19608c2763a0085cdbfb1dc180d042030dade3f1ff469b
3
  size 360622744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9aeb84d48b1a3e7876ab13cf2a9db637ae3eea5a8e14c759a56dfd289a32ce86
3
  size 360622744
shard_scheduler.json CHANGED
@@ -22,8 +22,8 @@
22
  "tokens_per_second": 598315.063954628,
23
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
24
  "estimated_pages_for_buffer": 1,
25
- "required_token_positions_per_second": 77005.1087248329,
26
- "required_pages_per_second_upper_estimate": 0.005327392098052995
27
  },
28
  {
29
  "source": "rewrite6",
@@ -36,8 +36,8 @@
36
  "tokens_per_second": 8533.957213384612,
37
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
38
  "estimated_pages_for_buffer": 1,
39
- "required_token_positions_per_second": 9625.638590604112,
40
- "required_pages_per_second_upper_estimate": 0.004030053871694994
41
  },
42
  {
43
  "source": "openmath",
@@ -50,8 +50,8 @@
50
  "tokens_per_second": 380144.35761751566,
51
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
52
  "estimated_pages_for_buffer": 1,
53
- "required_token_positions_per_second": 19251.277181208225,
54
- "required_pages_per_second_upper_estimate": 0.004009533346525709
55
  },
56
  {
57
  "source": "ultra_qa",
@@ -64,8 +64,8 @@
64
  "tokens_per_second": 43742.16556775435,
65
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
66
  "estimated_pages_for_buffer": 1,
67
- "required_token_positions_per_second": 28876.915771812335,
68
- "required_pages_per_second_upper_estimate": 0.0017311270080228043
69
  },
70
  {
71
  "source": "ultra_style",
@@ -78,8 +78,8 @@
78
  "tokens_per_second": 22018.86276523721,
79
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
80
  "estimated_pages_for_buffer": 1,
81
- "required_token_positions_per_second": 28876.915771812335,
82
- "required_pages_per_second_upper_estimate": 0.0022450531851294074
83
  },
84
  {
85
  "source": "cortex",
@@ -92,8 +92,8 @@
92
  "tokens_per_second": 910315.184299418,
93
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
94
  "estimated_pages_for_buffer": 23,
95
- "required_token_positions_per_second": 28876.915771812335,
96
- "required_pages_per_second_upper_estimate": 0.3704020699043411
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
@@ -158,129 +158,27 @@
158
  "prefetch_batches": 63,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 704566,
161
- "producer_batch_seconds": 0.09249245312517473,
162
- "producer_p95_seconds": 0.272625303001405,
163
- "gpu_update_seconds": 0.9574429699652371,
164
- "producer_batches_per_second": 10.811692913438561,
165
- "gpu_batches_per_second": 1.0444486317929809,
166
- "producer_headroom": 10.351579373394724,
167
  "sustainable": true,
168
- "buffer_seconds": 60.31890710780994,
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": 264013.5722249477,
174
- "upload_bytes_per_second": 3976907.631126424,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
- "estimated_full_shards_per_hour": 56.651166678059795,
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": [
182
- {
183
- "seconds": 1.2181368159981503,
184
- "bytes": 12003212,
185
- "pages": 41,
186
- "shards": 4
187
- },
188
- {
189
- "seconds": 0.9120353730031638,
190
- "bytes": 952497,
191
- "pages": 31,
192
- "shards": 1
193
- },
194
- {
195
- "seconds": 1.6672031760026584,
196
- "bytes": 13248105,
197
- "pages": 37,
198
- "shards": 4
199
- },
200
- {
201
- "seconds": 1.4338108570009354,
202
- "bytes": 5962058,
203
- "pages": 28,
204
- "shards": 2
205
- },
206
- {
207
- "seconds": 1.2595222179988923,
208
- "bytes": 5231936,
209
- "pages": 41,
210
- "shards": 2
211
- },
212
- {
213
- "seconds": 1.146318651000911,
214
- "bytes": 7213324,
215
- "pages": 30,
216
- "shards": 3
217
- },
218
- {
219
- "seconds": 1.1155308010020235,
220
- "bytes": 6292434,
221
- "pages": 37,
222
- "shards": 2
223
- },
224
- {
225
- "seconds": 1.3688695919991005,
226
- "bytes": 12940523,
227
- "pages": 39,
228
- "shards": 4
229
- },
230
- {
231
- "seconds": 1.308492041000136,
232
- "bytes": 6108009,
233
- "pages": 32,
234
- "shards": 2
235
- },
236
- {
237
- "seconds": 0.922122173000389,
238
- "bytes": 1095644,
239
- "pages": 31,
240
- "shards": 1
241
- },
242
- {
243
- "seconds": 1.0743960049985617,
244
- "bytes": 7509044,
245
- "pages": 39,
246
- "shards": 3
247
- },
248
- {
249
- "seconds": 1.3776889159998973,
250
- "bytes": 10102958,
251
- "pages": 36,
252
- "shards": 3
253
- },
254
- {
255
- "seconds": 4.997759090001637,
256
- "bytes": 838514,
257
- "pages": 26,
258
- "shards": 1
259
- },
260
- {
261
- "seconds": 34.2643265609986,
262
- "bytes": 13380891,
263
- "pages": 44,
264
- "shards": 4
265
- },
266
- {
267
- "seconds": 1.1768819699973392,
268
- "bytes": 10254247,
269
- "pages": 45,
270
- "shards": 3
271
- },
272
- {
273
- "seconds": 2.0107650110003306,
274
- "bytes": 2402302,
275
- "pages": 28,
276
- "shards": 2
277
- },
278
- {
279
- "seconds": 1.039769940001861,
280
- "bytes": 6252309,
281
- "pages": 38,
282
- "shards": 2
283
- },
284
  {
285
  "seconds": 1.272125319999759,
286
  "bytes": 7866146,
@@ -562,6 +460,108 @@
562
  "bytes": 5775284,
563
  "pages": 29,
564
  "shards": 2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
565
  }
566
  ],
567
  "cache_key": {
 
22
  "tokens_per_second": 598315.063954628,
23
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
24
  "estimated_pages_for_buffer": 1,
25
+ "required_token_positions_per_second": 76613.12993076043,
26
+ "required_pages_per_second_upper_estimate": 0.005300274095562953
27
  },
28
  {
29
  "source": "rewrite6",
 
36
  "tokens_per_second": 8533.957213384612,
37
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
38
  "estimated_pages_for_buffer": 1,
39
+ "required_token_positions_per_second": 9576.641241345054,
40
+ "required_pages_per_second_upper_estimate": 0.004009539704741229
41
  },
42
  {
43
  "source": "openmath",
 
50
  "tokens_per_second": 380144.35761751566,
51
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
52
  "estimated_pages_for_buffer": 1,
53
+ "required_token_positions_per_second": 19153.28248269011,
54
+ "required_pages_per_second_upper_estimate": 0.003989123635118372
55
  },
56
  {
57
  "source": "ultra_qa",
 
64
  "tokens_per_second": 43742.16556775435,
65
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
66
  "estimated_pages_for_buffer": 1,
67
+ "required_token_positions_per_second": 28729.923724035165,
68
+ "required_pages_per_second_upper_estimate": 0.0017223150592024243
69
  },
70
  {
71
  "source": "ultra_style",
 
78
  "tokens_per_second": 22018.86276523721,
79
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
80
  "estimated_pages_for_buffer": 1,
81
+ "required_token_positions_per_second": 28729.923724035165,
82
+ "required_pages_per_second_upper_estimate": 0.002233625199964421
83
  },
84
  {
85
  "source": "cortex",
 
92
  "tokens_per_second": 910315.184299418,
93
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
94
  "estimated_pages_for_buffer": 23,
95
+ "required_token_positions_per_second": 28729.923724035165,
96
+ "required_pages_per_second_upper_estimate": 0.36851661374321987
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
 
158
  "prefetch_batches": 63,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 704566,
161
+ "producer_batch_seconds": 0.09595589186756115,
162
+ "producer_p95_seconds": 0.29021746900252765,
163
+ "gpu_update_seconds": 0.9623415733912987,
164
+ "producer_batches_per_second": 10.421454905345525,
165
+ "gpu_batches_per_second": 1.0391320791390033,
166
+ "producer_headroom": 10.028999310636681,
167
  "sustainable": true,
168
+ "buffer_seconds": 60.627519123651815,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
+ "upload_pages_per_commit": 82,
172
  "max_pending_pages": 4096,
173
+ "generation_bytes_per_second": 370915.98441075865,
174
+ "upload_bytes_per_second": 5961453.864962328,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
+ "estimated_full_shards_per_hour": 79.58993577234335,
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": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
182
  {
183
  "seconds": 1.272125319999759,
184
  "bytes": 7866146,
 
460
  "bytes": 5775284,
461
  "pages": 29,
462
  "shards": 2
463
+ },
464
+ {
465
+ "seconds": 1.2868636950006476,
466
+ "bytes": 7532566,
467
+ "pages": 40,
468
+ "shards": 3
469
+ },
470
+ {
471
+ "seconds": 0.9102224750022287,
472
+ "bytes": 989956,
473
+ "pages": 26,
474
+ "shards": 1
475
+ },
476
+ {
477
+ "seconds": 1.172385066998686,
478
+ "bytes": 10249964,
479
+ "pages": 40,
480
+ "shards": 3
481
+ },
482
+ {
483
+ "seconds": 1.0767950209992705,
484
+ "bytes": 2405647,
485
+ "pages": 29,
486
+ "shards": 2
487
+ },
488
+ {
489
+ "seconds": 1.3306630199986103,
490
+ "bytes": 11933752,
491
+ "pages": 33,
492
+ "shards": 3
493
+ },
494
+ {
495
+ "seconds": 1.1605643660004716,
496
+ "bytes": 6187685,
497
+ "pages": 38,
498
+ "shards": 2
499
+ },
500
+ {
501
+ "seconds": 1.1438976530007494,
502
+ "bytes": 6974681,
503
+ "pages": 33,
504
+ "shards": 3
505
+ },
506
+ {
507
+ "seconds": 0.9301952720015834,
508
+ "bytes": 6018270,
509
+ "pages": 37,
510
+ "shards": 2
511
+ },
512
+ {
513
+ "seconds": 1.0365287800013903,
514
+ "bytes": 2455252,
515
+ "pages": 29,
516
+ "shards": 2
517
+ },
518
+ {
519
+ "seconds": 1.3521412089976366,
520
+ "bytes": 6109192,
521
+ "pages": 37,
522
+ "shards": 2
523
+ },
524
+ {
525
+ "seconds": 1.5889534400012053,
526
+ "bytes": 16077810,
527
+ "pages": 39,
528
+ "shards": 4
529
+ },
530
+ {
531
+ "seconds": 0.8900750899993,
532
+ "bytes": 2373140,
533
+ "pages": 29,
534
+ "shards": 2
535
+ },
536
+ {
537
+ "seconds": 1.2026867530003074,
538
+ "bytes": 6250314,
539
+ "pages": 37,
540
+ "shards": 2
541
+ },
542
+ {
543
+ "seconds": 1.0975493779988028,
544
+ "bytes": 5901759,
545
+ "pages": 32,
546
+ "shards": 2
547
+ },
548
+ {
549
+ "seconds": 1.0364178640011232,
550
+ "bytes": 6898240,
551
+ "pages": 40,
552
+ "shards": 3
553
+ },
554
+ {
555
+ "seconds": 1.1201825170028314,
556
+ "bytes": 6019295,
557
+ "pages": 28,
558
+ "shards": 2
559
+ },
560
+ {
561
+ "seconds": 1.523898775998532,
562
+ "bytes": 8240555,
563
+ "pages": 36,
564
+ "shards": 3
565
  }
566
  ],
567
  "cache_key": {
trainer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4c3e5c3949b8f4716b37f94e501fd3bf9bfc15652bfd259b8ceeb75ba134374a
3
- size 721768511
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:625680f99c72ef12793fea5968b6c72147fca5bf0986649c268a6a0699e0b367
3
+ size 721775807