appvoid commited on
Commit
aa80474
·
verified ·
1 Parent(s): e5ff0b2

Qwen90M step 450000; complete resumable checkpoint

Browse files
Files changed (6) hide show
  1. complete.json +5 -5
  2. coverage.json +49 -49
  3. latest.json +1 -1
  4. model.safetensors +1 -1
  5. shard_scheduler.json +263 -95
  6. trainer.pt +2 -2
complete.json CHANGED
@@ -1,18 +1,18 @@
1
  {
2
- "step": 445000,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
- "model.safetensors": "e31fd28a60f36256443cfb50640d20094f71da6c3cbb0aa80afc0a4cedec13b2",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
- "trainer.pt": "8b5075127ef2811f3ef44f7a6a9a1feed435ce847072f56baa6aec7dfad54d50",
12
  "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6",
13
  "performance.json": "58f8b6f79b191aa11b2bedb4594ca2897b6fdad2419ea7c9210688ac8022e7cb",
14
- "shard_scheduler.json": "efbac20cd8a66e3c031c9dfa928f6451599605e8f1347de2378f5b7118ef7df7",
15
- "coverage.json": "5537feb0d0fc33dea31497140125b57b887abc964d22bb91f1d5ee833a726e2b",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063"
 
1
  {
2
+ "step": 450000,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
+ "model.safetensors": "58dd2d60d7e62c078b341a3c996cc27526c94608917b12a0f0e4f255a30b38da",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
+ "trainer.pt": "8c82c09115063d52f8a9ddefbf6416afedc0c31c39ed9119e29edc7b2e51cf40",
12
  "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6",
13
  "performance.json": "58f8b6f79b191aa11b2bedb4594ca2897b6fdad2419ea7c9210688ac8022e7cb",
14
+ "shard_scheduler.json": "126657fe7a26409dd6081531cc72f78f47edafb341bf2475098d0f234db9b473",
15
+ "coverage.json": "3b743dc6b08c6b934daa36f47ff27ebedb7ae5d69624abd90badff03367e3348",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063"
coverage.json CHANGED
@@ -1,107 +1,107 @@
1
  {
2
- "step": 445000,
3
  "stats": {
4
- "climbmix/blocks": 16020000,
5
- "climbmix/targets": 26779783410,
6
- "climbmix/input_tokens": 26781479575,
7
- "ultra_qa/blocks": 6007500,
8
- "ultra_qa/targets": 9800065187,
9
- "ultra_qa/input_tokens": 9800083982,
10
- "cortex/blocks": 6007500,
11
- "cortex/targets": 228478974,
12
- "cortex/input_tokens": 9549691996,
13
- "ultra_style/blocks": 6007500,
14
- "ultra_style/targets": 9501732854,
15
- "ultra_style/input_tokens": 9502231083,
16
- "openmath/blocks": 4005000,
17
- "openmath/targets": 6188564834,
18
- "openmath/input_tokens": 6190627055,
19
- "rewrite6/blocks": 2002500,
20
- "rewrite6/targets": 3282863241,
21
- "rewrite6/input_tokens": 3286462391
22
  },
23
  "sources": {
24
  "climbmix": {
25
- "page": 2193,
26
- "row": 3524,
27
  "start": {
28
- "file": 106,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
- "row_group": 13,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
- "offset": 2048,
37
- "documents": 8986052
38
  },
39
  "rewrite6": {
40
- "page": 48,
41
- "row": 1213,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
- "row": 196608
47
  }
48
  },
49
- "epoch": 24,
50
- "offset": 0,
51
- "documents": 5127517
52
  },
53
  "openmath": {
54
- "page": 1371,
55
- "row": 1084,
56
  "start": {
57
- "file": 12,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
- "row_group": 375,
61
- "row": 816
62
  }
63
  },
64
  "epoch": 0,
65
  "offset": 0,
66
- "documents": 5616700
67
  },
68
  "ultra_qa": {
69
- "page": 592,
70
- "row": 3518,
71
  "start": {
72
  "file": 4,
73
  "reader": {
74
  "format": "parquet-row-v1",
75
  "row_group": 5,
76
- "row": 23370
77
  }
78
  },
79
  "epoch": 0,
80
  "offset": 2048,
81
- "documents": 2428350
82
  },
83
  "ultra_style": {
84
- "page": 905,
85
- "row": 84,
86
  "start": {
87
  "file": 5,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
  "row_group": 3,
91
- "row": 25655
92
  }
93
  },
94
  "epoch": 0,
95
- "offset": 0,
96
- "documents": 3706964
97
  },
98
  "cortex": {
99
- "page": 116515,
100
- "row": 115,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
- "documents": 14914035
105
  }
106
  }
107
  }
 
1
  {
2
+ "step": 450000,
3
  "stats": {
4
+ "climbmix/blocks": 16200000,
5
+ "climbmix/targets": 27080092741,
6
+ "climbmix/input_tokens": 27081807861,
7
+ "ultra_qa/blocks": 6075000,
8
+ "ultra_qa/targets": 9910123737,
9
+ "ultra_qa/input_tokens": 9910142766,
10
+ "cortex/blocks": 6075000,
11
+ "cortex/targets": 231042998,
12
+ "cortex/input_tokens": 9656909440,
13
+ "ultra_style/blocks": 6075000,
14
+ "ultra_style/targets": 9608673521,
15
+ "ultra_style/input_tokens": 9609177298,
16
+ "openmath/blocks": 4050000,
17
+ "openmath/targets": 6258150481,
18
+ "openmath/input_tokens": 6260235972,
19
+ "rewrite6/blocks": 2025000,
20
+ "rewrite6/targets": 3319721644,
21
+ "rewrite6/input_tokens": 3323361143
22
  },
23
  "sources": {
24
  "climbmix": {
25
+ "page": 2218,
26
+ "row": 2291,
27
  "start": {
28
+ "file": 107,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
+ "row_group": 30,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
+ "offset": 0,
37
+ "documents": 9087219
38
  },
39
  "rewrite6": {
40
+ "page": 12,
41
+ "row": 742,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
+ "row": 49152
47
  }
48
  },
49
+ "epoch": 25,
50
+ "offset": 2048,
51
+ "documents": 5184994
52
  },
53
  "openmath": {
54
+ "page": 1386,
55
+ "row": 2839,
56
  "start": {
57
+ "file": 13,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
+ "row_group": 0,
61
+ "row": 606
62
  }
63
  },
64
  "epoch": 0,
65
  "offset": 0,
66
+ "documents": 5679895
67
  },
68
  "ultra_qa": {
69
+ "page": 599,
70
+ "row": 2251,
71
  "start": {
72
  "file": 4,
73
  "reader": {
74
  "format": "parquet-row-v1",
75
  "row_group": 5,
76
+ "row": 52042
77
  }
78
  },
79
  "epoch": 0,
80
  "offset": 2048,
81
+ "documents": 2455755
82
  },
83
  "ultra_style": {
84
+ "page": 915,
85
+ "row": 323,
86
  "start": {
87
  "file": 5,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
  "row_group": 3,
91
+ "row": 66615
92
  }
93
  },
94
  "epoch": 0,
95
+ "offset": 2048,
96
+ "documents": 3748163
97
  },
98
  "cortex": {
99
+ "page": 117823,
100
+ "row": 63,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
+ "documents": 15081407
105
  }
106
  }
107
  }
latest.json CHANGED
@@ -1 +1 @@
1
- {"step": 445000, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "e31fd28a60f36256443cfb50640d20094f71da6c3cbb0aa80afc0a4cedec13b2", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "8b5075127ef2811f3ef44f7a6a9a1feed435ce847072f56baa6aec7dfad54d50", "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6", "performance.json": "58f8b6f79b191aa11b2bedb4594ca2897b6fdad2419ea7c9210688ac8022e7cb", "shard_scheduler.json": "efbac20cd8a66e3c031c9dfa928f6451599605e8f1347de2378f5b7118ef7df7", "coverage.json": "5537feb0d0fc33dea31497140125b57b887abc964d22bb91f1d5ee833a726e2b", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063", "folder": "step-000000445000"}
 
1
+ {"step": 450000, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "58dd2d60d7e62c078b341a3c996cc27526c94608917b12a0f0e4f255a30b38da", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "8c82c09115063d52f8a9ddefbf6416afedc0c31c39ed9119e29edc7b2e51cf40", "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6", "performance.json": "58f8b6f79b191aa11b2bedb4594ca2897b6fdad2419ea7c9210688ac8022e7cb", "shard_scheduler.json": "126657fe7a26409dd6081531cc72f78f47edafb341bf2475098d0f234db9b473", "coverage.json": "3b743dc6b08c6b934daa36f47ff27ebedb7ae5d69624abd90badff03367e3348", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063", "folder": "step-000000450000"}
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e31fd28a60f36256443cfb50640d20094f71da6c3cbb0aa80afc0a4cedec13b2
3
  size 360622744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58dd2d60d7e62c078b341a3c996cc27526c94608917b12a0f0e4f255a30b38da
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": 77280.73115412582,
26
- "required_pages_per_second_upper_estimate": 0.0053464602972436255
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": 9660.091394265728,
40
- "required_pages_per_second_upper_estimate": 0.00404447854113176
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": 19320.182788531456,
54
- "required_pages_per_second_upper_estimate": 0.0040238845673680745
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": 28980.274182797188,
68
- "required_pages_per_second_upper_estimate": 0.0017373231869421907
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": 28980.274182797188,
82
- "required_pages_per_second_upper_estimate": 0.0022530888469578757
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": 28980.274182797188,
96
- "required_pages_per_second_upper_estimate": 0.3717278406228395
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
@@ -158,99 +158,27 @@
158
  "prefetch_batches": 63,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 704566,
161
- "producer_batch_seconds": 0.23549360834378064,
162
- "producer_p95_seconds": 0.1584173479996025,
163
- "gpu_update_seconds": 0.9540282409202315,
164
- "producer_batches_per_second": 4.2463997517086325,
165
- "gpu_batches_per_second": 1.0481870002458473,
166
- "producer_headroom": 4.051185285366694,
167
  "sustainable": true,
168
- "buffer_seconds": 60.10377917797459,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
- "upload_pages_per_commit": 83,
172
  "max_pending_pages": 4096,
173
- "generation_bytes_per_second": 486911.6116432644,
174
- "upload_bytes_per_second": 3768783.117588168,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
- "estimated_full_shards_per_hour": 104.47989713643503,
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": 1.0239758469999742,
184
- "bytes": 22958,
185
- "pages": 1,
186
- "shards": 1
187
- },
188
- {
189
- "seconds": 1.2533257390000472,
190
- "bytes": 16191062,
191
- "pages": 36,
192
- "shards": 4
193
- },
194
- {
195
- "seconds": 0.8292768119999891,
196
- "bytes": 2252671,
197
- "pages": 27,
198
- "shards": 2
199
- },
200
- {
201
- "seconds": 1.5412007810000432,
202
- "bytes": 6087939,
203
- "pages": 36,
204
- "shards": 2
205
- },
206
- {
207
- "seconds": 1.0659989769999356,
208
- "bytes": 6165937,
209
- "pages": 36,
210
- "shards": 2
211
- },
212
- {
213
- "seconds": 1.023882131999926,
214
- "bytes": 6945306,
215
- "pages": 30,
216
- "shards": 3
217
- },
218
- {
219
- "seconds": 1.1334343210000952,
220
- "bytes": 6129536,
221
- "pages": 36,
222
- "shards": 2
223
- },
224
- {
225
- "seconds": 1.3353518999999778,
226
- "bytes": 13219093,
227
- "pages": 34,
228
- "shards": 4
229
- },
230
- {
231
- "seconds": 0.8045400529999824,
232
- "bytes": 1220702,
233
- "pages": 34,
234
- "shards": 1
235
- },
236
- {
237
- "seconds": 1.144875766000041,
238
- "bytes": 9911722,
239
- "pages": 38,
240
- "shards": 3
241
- },
242
- {
243
- "seconds": 0.8446817050003119,
244
- "bytes": 2246420,
245
- "pages": 27,
246
- "shards": 2
247
- },
248
- {
249
- "seconds": 1.1746671439996135,
250
- "bytes": 6226498,
251
- "pages": 36,
252
- "shards": 2
253
- },
254
  {
255
  "seconds": 1.0685757479996028,
256
  "bytes": 7863934,
@@ -394,6 +322,246 @@
394
  "bytes": 7420139,
395
  "pages": 34,
396
  "shards": 3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
397
  }
398
  ],
399
  "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": 76859.77989241168,
26
+ "required_pages_per_second_upper_estimate": 0.005317337912216746
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": 9607.47248655146,
40
+ "required_pages_per_second_upper_estimate": 0.004022448103279538
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": 19214.94497310292,
54
+ "required_pages_per_second_upper_estimate": 0.004001966305722134
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": 28822.417459654378,
68
+ "required_pages_per_second_upper_estimate": 0.0017278599174230392
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": 28822.417459654378,
82
+ "required_pages_per_second_upper_estimate": 0.0022408161810718676
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": 28822.417459654378,
96
+ "required_pages_per_second_upper_estimate": 0.36970302407170735
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": 0.4567267198124938,
162
+ "producer_p95_seconds": 0.184863974000109,
163
+ "gpu_update_seconds": 0.9592533325388709,
164
+ "producer_batches_per_second": 2.189493096463775,
165
+ "gpu_batches_per_second": 1.0424774833497676,
166
+ "producer_headroom": 2.1002785493537277,
167
  "sustainable": true,
168
+ "buffer_seconds": 60.43295994994887,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
+ "upload_pages_per_commit": 84,
172
  "max_pending_pages": 4096,
173
+ "generation_bytes_per_second": 259488.05878523772,
174
+ "upload_bytes_per_second": 4561519.041951923,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
+ "estimated_full_shards_per_hour": 55.68009684245919,
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": 1.0685757479996028,
184
  "bytes": 7863934,
 
322
  "bytes": 7420139,
323
  "pages": 34,
324
  "shards": 3
325
+ },
326
+ {
327
+ "seconds": 1.4091035430001284,
328
+ "bytes": 10556108,
329
+ "pages": 34,
330
+ "shards": 3
331
+ },
332
+ {
333
+ "seconds": 1.07210781699996,
334
+ "bytes": 6934469,
335
+ "pages": 36,
336
+ "shards": 2
337
+ },
338
+ {
339
+ "seconds": 1.0243737840000904,
340
+ "bytes": 7482664,
341
+ "pages": 30,
342
+ "shards": 3
343
+ },
344
+ {
345
+ "seconds": 0.7832126150005934,
346
+ "bytes": 1122466,
347
+ "pages": 34,
348
+ "shards": 1
349
+ },
350
+ {
351
+ "seconds": 1.146071767999274,
352
+ "bytes": 10485882,
353
+ "pages": 36,
354
+ "shards": 3
355
+ },
356
+ {
357
+ "seconds": 1.0545766800005367,
358
+ "bytes": 7165492,
359
+ "pages": 30,
360
+ "shards": 3
361
+ },
362
+ {
363
+ "seconds": 0.8369446019996758,
364
+ "bytes": 1213172,
365
+ "pages": 35,
366
+ "shards": 1
367
+ },
368
+ {
369
+ "seconds": 1.6566883329996926,
370
+ "bytes": 13442976,
371
+ "pages": 41,
372
+ "shards": 4
373
+ },
374
+ {
375
+ "seconds": 1.1581910269997024,
376
+ "bytes": 10184602,
377
+ "pages": 35,
378
+ "shards": 3
379
+ },
380
+ {
381
+ "seconds": 1.2292702360000476,
382
+ "bytes": 942860,
383
+ "pages": 26,
384
+ "shards": 1
385
+ },
386
+ {
387
+ "seconds": 1.2621263359997101,
388
+ "bytes": 7688000,
389
+ "pages": 36,
390
+ "shards": 3
391
+ },
392
+ {
393
+ "seconds": 1.0023074259997884,
394
+ "bytes": 1140621,
395
+ "pages": 35,
396
+ "shards": 1
397
+ },
398
+ {
399
+ "seconds": 1.374354456999754,
400
+ "bytes": 11971022,
401
+ "pages": 39,
402
+ "shards": 4
403
+ },
404
+ {
405
+ "seconds": 1.1425232550000146,
406
+ "bytes": 11707710,
407
+ "pages": 31,
408
+ "shards": 3
409
+ },
410
+ {
411
+ "seconds": 0.9451075770002717,
412
+ "bytes": 1133265,
413
+ "pages": 35,
414
+ "shards": 1
415
+ },
416
+ {
417
+ "seconds": 1.8210042499995325,
418
+ "bytes": 7672821,
419
+ "pages": 32,
420
+ "shards": 4
421
+ },
422
+ {
423
+ "seconds": 1.2987783030002902,
424
+ "bytes": 10941174,
425
+ "pages": 37,
426
+ "shards": 3
427
+ },
428
+ {
429
+ "seconds": 0.9096585289998984,
430
+ "bytes": 2702720,
431
+ "pages": 37,
432
+ "shards": 2
433
+ },
434
+ {
435
+ "seconds": 1.212294140999802,
436
+ "bytes": 5935208,
437
+ "pages": 36,
438
+ "shards": 2
439
+ },
440
+ {
441
+ "seconds": 1.1118033149996336,
442
+ "bytes": 6738611,
443
+ "pages": 31,
444
+ "shards": 2
445
+ },
446
+ {
447
+ "seconds": 1.2066322789996775,
448
+ "bytes": 7050449,
449
+ "pages": 25,
450
+ "shards": 3
451
+ },
452
+ {
453
+ "seconds": 1.1100294799998665,
454
+ "bytes": 10655210,
455
+ "pages": 43,
456
+ "shards": 3
457
+ },
458
+ {
459
+ "seconds": 0.9483619619995807,
460
+ "bytes": 1257250,
461
+ "pages": 35,
462
+ "shards": 1
463
+ },
464
+ {
465
+ "seconds": 0.9075711420000516,
466
+ "bytes": 7428003,
467
+ "pages": 38,
468
+ "shards": 3
469
+ },
470
+ {
471
+ "seconds": 1.278834224000093,
472
+ "bytes": 9834674,
473
+ "pages": 30,
474
+ "shards": 3
475
+ },
476
+ {
477
+ "seconds": 1.1737758689996554,
478
+ "bytes": 8190004,
479
+ "pages": 33,
480
+ "shards": 3
481
+ },
482
+ {
483
+ "seconds": 1.150645027999417,
484
+ "bytes": 6104360,
485
+ "pages": 37,
486
+ "shards": 2
487
+ },
488
+ {
489
+ "seconds": 1.1603049650002504,
490
+ "bytes": 1160533,
491
+ "pages": 35,
492
+ "shards": 1
493
+ },
494
+ {
495
+ "seconds": 1.0780582049992518,
496
+ "bytes": 12237678,
497
+ "pages": 34,
498
+ "shards": 4
499
+ },
500
+ {
501
+ "seconds": 0.9231663379996462,
502
+ "bytes": 6121395,
503
+ "pages": 36,
504
+ "shards": 2
505
+ },
506
+ {
507
+ "seconds": 0.8000806340005511,
508
+ "bytes": 2307263,
509
+ "pages": 28,
510
+ "shards": 2
511
+ },
512
+ {
513
+ "seconds": 1.328747977000603,
514
+ "bytes": 11901603,
515
+ "pages": 40,
516
+ "shards": 3
517
+ },
518
+ {
519
+ "seconds": 1.0515727450001577,
520
+ "bytes": 10443094,
521
+ "pages": 35,
522
+ "shards": 3
523
+ },
524
+ {
525
+ "seconds": 0.9829580149998947,
526
+ "bytes": 2655368,
527
+ "pages": 37,
528
+ "shards": 2
529
+ },
530
+ {
531
+ "seconds": 1.023656208999455,
532
+ "bytes": 5985547,
533
+ "pages": 27,
534
+ "shards": 2
535
+ },
536
+ {
537
+ "seconds": 1.1010237709997455,
538
+ "bytes": 2681477,
539
+ "pages": 37,
540
+ "shards": 2
541
+ },
542
+ {
543
+ "seconds": 1.0707301010006631,
544
+ "bytes": 10039688,
545
+ "pages": 34,
546
+ "shards": 3
547
+ },
548
+ {
549
+ "seconds": 1.0178890370007139,
550
+ "bytes": 11702648,
551
+ "pages": 30,
552
+ "shards": 3
553
+ },
554
+ {
555
+ "seconds": 1.0858605960002023,
556
+ "bytes": 2526198,
557
+ "pages": 37,
558
+ "shards": 2
559
+ },
560
+ {
561
+ "seconds": 1.2877745800014964,
562
+ "bytes": 5791906,
563
+ "pages": 28,
564
+ "shards": 2
565
  }
566
  ],
567
  "cache_key": {
trainer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8b5075127ef2811f3ef44f7a6a9a1feed435ce847072f56baa6aec7dfad54d50
3
- size 721771135
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c82c09115063d52f8a9ddefbf6416afedc0c31c39ed9119e29edc7b2e51cf40
3
+ size 721776959