appvoid commited on
Commit
be4686c
·
verified ·
1 Parent(s): ed9cab3

Qwen90M step 440277; complete resumable checkpoint

Browse files
Files changed (6) hide show
  1. complete.json +5 -5
  2. coverage.json +43 -43
  3. latest.json +1 -1
  4. model.safetensors +1 -1
  5. shard_scheduler.json +41 -41
  6. trainer.pt +2 -2
complete.json CHANGED
@@ -1,18 +1,18 @@
1
  {
2
- "step": 440000,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
- "model.safetensors": "fe214f4ab1fb9964cf107eac88ddf367929c1be6f58e91da736242cfa8027d31",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
- "trainer.pt": "8bb0a7558188a5dfb0a78d6ae95be1812ae85f92993c239c65c6dc7be7f1be7d",
12
  "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6",
13
  "performance.json": "a6ae17d5ca4d76a29c6e069dddb352125dfbf6f18e59e9553ab94c741a962b18",
14
- "shard_scheduler.json": "6724984473eb995a25775ee3318e6a2730c21a7c2160f6311680bbf3c31a3993",
15
- "coverage.json": "3bc0c290712aa33916e32c37c9e507b57b125caed3a356f046223639dd1963d8",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063"
 
1
  {
2
+ "step": 440277,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
+ "model.safetensors": "8a7a87fe75a5fd44cbeb174e25a756396db70470c62d2ac3cab5cb0c7ca63f33",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
+ "trainer.pt": "4647e9f281860ad757cb073bc1c37c925e1bf8c1a3363bc3857ab396fe20b7fc",
12
  "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6",
13
  "performance.json": "a6ae17d5ca4d76a29c6e069dddb352125dfbf6f18e59e9553ab94c741a962b18",
14
+ "shard_scheduler.json": "5c5eb10191a168242e046d479a37b48f91a5de39c8e14d82d54a83d47dbceb69",
15
+ "coverage.json": "f9927ca485baa62a65235b140994e7e3f62d2fca6bbf7433fe52e93b3eef7c89",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063"
coverage.json CHANGED
@@ -1,73 +1,73 @@
1
  {
2
- "step": 440000,
3
  "stats": {
4
- "climbmix/blocks": 15840000,
5
- "climbmix/targets": 26479322066,
6
- "climbmix/input_tokens": 26480999183,
7
- "ultra_qa/blocks": 5940000,
8
- "ultra_qa/targets": 9689929802,
9
- "ultra_qa/input_tokens": 9689948398,
10
- "cortex/blocks": 5940000,
11
- "cortex/targets": 225912648,
12
- "cortex/input_tokens": 9442402042,
13
- "ultra_style/blocks": 5940000,
14
- "ultra_style/targets": 9395030681,
15
- "ultra_style/input_tokens": 9395523198,
16
- "openmath/blocks": 3960000,
17
- "openmath/targets": 6119068769,
18
- "openmath/input_tokens": 6121107744,
19
- "rewrite6/blocks": 1980000,
20
- "rewrite6/targets": 3245883112,
21
- "rewrite6/input_tokens": 3249442073
22
  },
23
  "sources": {
24
  "climbmix": {
25
- "page": 2169,
26
- "row": 730,
27
  "start": {
28
  "file": 105,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
- "row_group": 0,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
  "offset": 2048,
37
- "documents": 8884954
38
  },
39
  "rewrite6": {
40
- "page": 34,
41
- "row": 1296,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
- "row": 139264
47
  }
48
  },
49
  "epoch": 24,
50
- "offset": 0,
51
- "documents": 5070256
52
  },
53
  "openmath": {
54
- "page": 1355,
55
- "row": 3306,
56
  "start": {
57
  "file": 12,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
- "row_group": 310,
61
- "row": 280
62
  }
63
  },
64
  "epoch": 0,
65
  "offset": 0,
66
- "documents": 5553386
67
  },
68
  "ultra_qa": {
69
  "page": 586,
70
- "row": 858,
71
  "start": {
72
  "file": 4,
73
  "reader": {
@@ -77,31 +77,31 @@
77
  }
78
  },
79
  "epoch": 0,
80
- "offset": 4096,
81
- "documents": 2401114
82
  },
83
  "ultra_style": {
84
- "page": 894,
85
- "row": 3292,
86
  "start": {
87
  "file": 5,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
  "row_group": 2,
91
- "row": 66129
92
  }
93
  },
94
  "epoch": 0,
95
  "offset": 2048,
96
- "documents": 3665116
97
  },
98
  "cortex": {
99
- "page": 115207,
100
- "row": 41,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
- "documents": 14746537
105
  }
106
  }
107
  }
 
1
  {
2
+ "step": 440277,
3
  "stats": {
4
+ "climbmix/blocks": 15849972,
5
+ "climbmix/targets": 26495964044,
6
+ "climbmix/input_tokens": 26497642217,
7
+ "ultra_qa/blocks": 5943739,
8
+ "ultra_qa/targets": 9696006191,
9
+ "ultra_qa/input_tokens": 9696024796,
10
+ "cortex/blocks": 5943740,
11
+ "cortex/targets": 226052219,
12
+ "cortex/input_tokens": 9448331338,
13
+ "ultra_style/blocks": 5943740,
14
+ "ultra_style/targets": 9401027313,
15
+ "ultra_style/input_tokens": 9401520126,
16
+ "openmath/blocks": 3962493,
17
+ "openmath/targets": 6122914510,
18
+ "openmath/input_tokens": 6124954796,
19
+ "rewrite6/blocks": 1981246,
20
+ "rewrite6/targets": 3247923015,
21
+ "rewrite6/input_tokens": 3251484097
22
  },
23
  "sources": {
24
  "climbmix": {
25
+ "page": 2170,
26
+ "row": 2226,
27
  "start": {
28
  "file": 105,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
+ "row_group": 4,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
  "offset": 2048,
37
+ "documents": 8890546
38
  },
39
  "rewrite6": {
40
+ "page": 35,
41
+ "row": 249,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
+ "row": 143360
47
  }
48
  },
49
  "epoch": 24,
50
+ "offset": 2048,
51
+ "documents": 5073305
52
  },
53
  "openmath": {
54
+ "page": 1356,
55
+ "row": 2756,
56
  "start": {
57
  "file": 12,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
+ "row_group": 314,
61
+ "row": 376
62
  }
63
  },
64
  "epoch": 0,
65
  "offset": 0,
66
+ "documents": 5556932
67
  },
68
  "ultra_qa": {
69
  "page": 586,
70
+ "row": 2357,
71
  "start": {
72
  "file": 4,
73
  "reader": {
 
77
  }
78
  },
79
  "epoch": 0,
80
+ "offset": 2048,
81
+ "documents": 2402613
82
  },
83
  "ultra_style": {
84
+ "page": 895,
85
+ "row": 1391,
86
  "start": {
87
  "file": 5,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
  "row_group": 2,
91
+ "row": 70225
92
  }
93
  },
94
  "epoch": 0,
95
  "offset": 2048,
96
+ "documents": 3667311
97
  },
98
  "cortex": {
99
+ "page": 115279,
100
+ "row": 76,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
+ "documents": 14755788
105
  }
106
  }
107
  }
latest.json CHANGED
@@ -1 +1 @@
1
- {"step": 440000, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "fe214f4ab1fb9964cf107eac88ddf367929c1be6f58e91da736242cfa8027d31", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "8bb0a7558188a5dfb0a78d6ae95be1812ae85f92993c239c65c6dc7be7f1be7d", "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6", "performance.json": "a6ae17d5ca4d76a29c6e069dddb352125dfbf6f18e59e9553ab94c741a962b18", "shard_scheduler.json": "6724984473eb995a25775ee3318e6a2730c21a7c2160f6311680bbf3c31a3993", "coverage.json": "3bc0c290712aa33916e32c37c9e507b57b125caed3a356f046223639dd1963d8", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063", "folder": "step-000000440000"}
 
1
+ {"step": 440277, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "8a7a87fe75a5fd44cbeb174e25a756396db70470c62d2ac3cab5cb0c7ca63f33", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "4647e9f281860ad757cb073bc1c37c925e1bf8c1a3363bc3857ab396fe20b7fc", "training_config.json": "c8bdf052510c2d9092e5a6824b119dd6584dc7697f8c4a8e9b7ecf3057c5bbb6", "performance.json": "a6ae17d5ca4d76a29c6e069dddb352125dfbf6f18e59e9553ab94c741a962b18", "shard_scheduler.json": "5c5eb10191a168242e046d479a37b48f91a5de39c8e14d82d54a83d47dbceb69", "coverage.json": "f9927ca485baa62a65235b140994e7e3f62d2fca6bbf7433fe52e93b3eef7c89", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "9fee47918905b0750bee87876aa424f021ccc4507ee8db22a11d01fbc6109063", "folder": "step-000000440277"}
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fe214f4ab1fb9964cf107eac88ddf367929c1be6f58e91da736242cfa8027d31
3
  size 360622744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a7a87fe75a5fd44cbeb174e25a756396db70470c62d2ac3cab5cb0c7ca63f33
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": 76924.88308471063,
26
- "required_pages_per_second_upper_estimate": 0.005321841902120205
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": 9615.610385588829,
40
- "required_pages_per_second_upper_estimate": 0.004025855271667829
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": 19231.220771177657,
54
- "required_pages_per_second_upper_estimate": 0.004005356125239444
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": 28846.831156766486,
68
- "required_pages_per_second_upper_estimate": 0.0017293234812873531
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": 28846.831156766486,
82
- "required_pages_per_second_upper_estimate": 0.002242714238637829
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": 28846.831156766486,
96
- "required_pages_per_second_upper_estimate": 0.37001617676487586
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
@@ -158,45 +158,27 @@
158
  "prefetch_batches": 63,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 704566,
161
- "producer_batch_seconds": 0.5073274579607983,
162
- "producer_p95_seconds": 0.34248151299834717,
163
- "gpu_update_seconds": 0.9584414956966502,
164
- "producer_batches_per_second": 1.9711134974233369,
165
- "gpu_batches_per_second": 1.0433605019085102,
166
- "producer_headroom": 1.8891969686582784,
167
  "sustainable": true,
168
- "buffer_seconds": 60.38181422888896,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
- "upload_pages_per_commit": 78,
172
  "max_pending_pages": 4096,
173
- "generation_bytes_per_second": 270603.4737486303,
174
- "upload_bytes_per_second": 5111100.11035178,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
- "estimated_full_shards_per_hour": 58.06520613998587,
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": 0.8855937030020868,
184
- "bytes": 2335409,
185
- "pages": 29,
186
- "shards": 2
187
- },
188
- {
189
- "seconds": 1.0506481749980594,
190
- "bytes": 6177700,
191
- "pages": 37,
192
- "shards": 2
193
- },
194
- {
195
- "seconds": 1.1661089699991862,
196
- "bytes": 11581137,
197
- "pages": 34,
198
- "shards": 4
199
- },
200
  {
201
  "seconds": 1.0408890239996254,
202
  "bytes": 7081099,
@@ -562,6 +544,24 @@
562
  "bytes": 5561171,
563
  "pages": 22,
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": 77008.13474038464,
26
+ "required_pages_per_second_upper_estimate": 0.005327601444830188
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": 9626.01684254808,
40
+ "required_pages_per_second_upper_estimate": 0.004030212237885135
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": 19252.03368509616,
54
+ "required_pages_per_second_upper_estimate": 0.004009690906335218
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": 28878.05052764424,
68
+ "required_pages_per_second_upper_estimate": 0.0017311950349022554
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": 28878.05052764424,
82
+ "required_pages_per_second_upper_estimate": 0.0022451414074040744
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": 28878.05052764424,
96
+ "required_pages_per_second_upper_estimate": 0.3704166253337469
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": 0.1179283470082737,
162
+ "producer_p95_seconds": 0.348671423002088,
163
+ "gpu_update_seconds": 0.9574053474812385,
164
+ "producer_batches_per_second": 8.47972540418837,
165
+ "gpu_batches_per_second": 1.0444896747556511,
166
+ "producer_headroom": 8.118534447142451,
167
  "sustainable": true,
168
+ "buffer_seconds": 60.31653689131802,
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": 282787.29616807465,
174
+ "upload_bytes_per_second": 5101231.701098665,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
+ "estimated_full_shards_per_hour": 60.67957080632858,
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": 1.0408890239996254,
184
  "bytes": 7081099,
 
544
  "bytes": 5561171,
545
  "pages": 22,
546
  "shards": 2
547
+ },
548
+ {
549
+ "seconds": 1.2334552800020901,
550
+ "bytes": 12163400,
551
+ "pages": 48,
552
+ "shards": 3
553
+ },
554
+ {
555
+ "seconds": 1.5185914780013263,
556
+ "bytes": 255711,
557
+ "pages": 9,
558
+ "shards": 1
559
+ },
560
+ {
561
+ "seconds": 1.23066899299738,
562
+ "bytes": 11344147,
563
+ "pages": 36,
564
+ "shards": 4
565
  }
566
  ],
567
  "cache_key": {
trainer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8bb0a7558188a5dfb0a78d6ae95be1812ae85f92993c239c65c6dc7be7f1be7d
3
- size 721770431
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4647e9f281860ad757cb073bc1c37c925e1bf8c1a3363bc3857ab396fe20b7fc
3
+ size 721777791