appvoid commited on
Commit
6cc0075
·
verified ·
1 Parent(s): f76140f

Qwen90M step 730000; complete resumable checkpoint

Browse files
Files changed (7) hide show
  1. complete.json +6 -6
  2. coverage.json +51 -51
  3. latest.json +1 -1
  4. model.safetensors +1 -1
  5. shard_scheduler.json +265 -25
  6. trainer.pt +2 -2
  7. training_config.json +1 -1
complete.json CHANGED
@@ -1,18 +1,18 @@
1
  {
2
- "step": 725000,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
- "model.safetensors": "e71ca4f551a1fff6787ce12f5aa4c155ba5ee5b4116a714d3f9add404be3f44d",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
- "trainer.pt": "cc65b927e55c762cf8c194b1c80ab87a382a45d747086f25332683699912d204",
12
- "training_config.json": "a5b628e58d936dba526336e6ed1a3b16a703b7373430cc4aca66220c358ed037",
13
  "performance.json": "48e3aa7938b07d5e08810d60bc208cc2cb9133c4c39a24151758b79306675379",
14
- "shard_scheduler.json": "5d40fdc3e8d0b948086712e99ffe8769eaf1235b617b80b55a7573c5ed79ce15",
15
- "coverage.json": "c5dc0b0b83b7968cda35d137af58ee922f5d9e851767895167b255f14ad87398",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "13510bfc04d8fed598fd0d09c1f778fbaebf19cefdd81f23f5ef3b909ef94a63"
 
1
  {
2
+ "step": 730000,
3
  "files": {
4
  "config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7",
5
  "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4",
6
+ "model.safetensors": "8c53eaad59485c3e7f93abf3a4b58bce24607b53a618b83f0a58c30a6ceb0a27",
7
  "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03",
8
  "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300",
9
  "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263",
10
  "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6",
11
+ "trainer.pt": "1650174bff312eebb6355ae334a6607b73e2c555e90fe45a7a399e9567721584",
12
+ "training_config.json": "75248af0a321738b405c3d51d1d55c7040dbc0785759d95f8dbdf6cd071a6687",
13
  "performance.json": "48e3aa7938b07d5e08810d60bc208cc2cb9133c4c39a24151758b79306675379",
14
+ "shard_scheduler.json": "1ca6461804592e6d06e70f93e961d9b78fbee44b41acd3825f26f6670f746dd6",
15
+ "coverage.json": "a190366ff62a7a816e8e5cfff69e551cc2bb0b47d066c9abcec59ff0915c60cf",
16
  "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"
17
  },
18
  "signature": "13510bfc04d8fed598fd0d09c1f778fbaebf19cefdd81f23f5ef3b909ef94a63"
coverage.json CHANGED
@@ -1,107 +1,107 @@
1
  {
2
- "step": 725000,
3
  "stats": {
4
- "climbmix/blocks": 26100000,
5
- "climbmix/targets": 43631105432,
6
- "climbmix/input_tokens": 43633870597,
7
- "ultra_qa/blocks": 9787500,
8
- "ultra_qa/targets": 15967266823,
9
- "ultra_qa/input_tokens": 15967297638,
10
- "cortex/blocks": 9787500,
11
- "cortex/targets": 372138551,
12
- "cortex/input_tokens": 15557701005,
13
- "ultra_style/blocks": 9787500,
14
- "ultra_style/targets": 15481228694,
15
- "ultra_style/input_tokens": 15482040970,
16
- "openmath/blocks": 6525000,
17
- "openmath/targets": 10082511799,
18
- "openmath/input_tokens": 10085871063,
19
- "rewrite6/blocks": 3262500,
20
- "rewrite6/targets": 5348374203,
21
- "rewrite6/input_tokens": 5354238124
22
  },
23
  "sources": {
24
  "climbmix": {
25
- "page": 3574,
26
- "row": 2403,
27
  "start": {
28
- "file": 172,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
- "row_group": 81,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
- "offset": 2048,
37
- "documents": 14641507
38
  },
39
  "rewrite6": {
40
- "page": 33,
41
- "row": 2732,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
- "row": 135168
47
  }
48
  },
49
  "epoch": 40,
50
- "offset": 0,
51
- "documents": 8354060
52
  },
53
  "openmath": {
54
- "page": 2233,
55
- "row": 3913,
56
  "start": {
57
- "file": 20,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
- "row_group": 413,
61
- "row": 368
62
  }
63
  },
64
  "epoch": 0,
65
  "offset": 0,
66
- "documents": 9150281
67
  },
68
  "ultra_qa": {
69
- "page": 965,
70
- "row": 3745,
71
  "start": {
72
  "file": 7,
73
  "reader": {
74
  "format": "parquet-row-v1",
75
- "row_group": 4,
76
- "row": 57150
77
  }
78
  },
79
  "epoch": 0,
80
- "offset": 4096,
81
- "documents": 3956385
82
  },
83
  "ultra_style": {
84
- "page": 1474,
85
- "row": 1212,
86
  "start": {
87
  "file": 8,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
- "row_group": 6,
91
- "row": 47354
92
  }
93
  },
94
  "epoch": 0,
95
- "offset": 2048,
96
- "documents": 6038716
97
  },
98
  "cortex": {
99
- "page": 189816,
100
- "row": 93,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
- "documents": 24296541
105
  }
106
  }
107
  }
 
1
  {
2
+ "step": 730000,
3
  "stats": {
4
+ "climbmix/blocks": 26280000,
5
+ "climbmix/targets": 43931838219,
6
+ "climbmix/input_tokens": 43934622432,
7
+ "ultra_qa/blocks": 9855000,
8
+ "ultra_qa/targets": 16077362872,
9
+ "ultra_qa/input_tokens": 16077393915,
10
+ "cortex/blocks": 9855000,
11
+ "cortex/targets": 374703742,
12
+ "cortex/input_tokens": 15664993820,
13
+ "ultra_style/blocks": 9855000,
14
+ "ultra_style/targets": 15587572089,
15
+ "ultra_style/input_tokens": 15588389892,
16
+ "openmath/blocks": 6570000,
17
+ "openmath/targets": 10152051955,
18
+ "openmath/input_tokens": 10155434511,
19
+ "rewrite6/blocks": 3285000,
20
+ "rewrite6/targets": 5385363851,
21
+ "rewrite6/input_tokens": 5391268095
22
  },
23
  "sources": {
24
  "climbmix": {
25
+ "page": 3599,
26
+ "row": 1241,
27
  "start": {
28
+ "file": 174,
29
  "reader": {
30
  "format": "parquet-row-v1",
31
+ "row_group": 16,
32
  "row": 1024
33
  }
34
  },
35
  "epoch": 0,
36
+ "offset": 0,
37
+ "documents": 14742745
38
  },
39
  "rewrite6": {
40
+ "page": 47,
41
+ "row": 2791,
42
  "start": {
43
  "file": 0,
44
  "reader": {
45
  "format": "jsonl-row-v1",
46
+ "row": 192512
47
  }
48
  },
49
  "epoch": 40,
50
+ "offset": 2048,
51
+ "documents": 8411463
52
  },
53
  "openmath": {
54
+ "page": 2249,
55
+ "row": 1624,
56
  "start": {
57
+ "file": 21,
58
  "reader": {
59
  "format": "parquet-row-v1",
60
+ "row_group": 42,
61
+ "row": 254
62
  }
63
  },
64
  "epoch": 0,
65
  "offset": 0,
66
+ "documents": 9213528
67
  },
68
  "ultra_qa": {
69
+ "page": 972,
70
+ "row": 2405,
71
  "start": {
72
  "file": 7,
73
  "reader": {
74
  "format": "parquet-row-v1",
75
+ "row_group": 5,
76
+ "row": 21207
77
  }
78
  },
79
  "epoch": 0,
80
+ "offset": 2048,
81
+ "documents": 3983717
82
  },
83
  "ultra_style": {
84
+ "page": 1484,
85
+ "row": 2035,
86
  "start": {
87
  "file": 8,
88
  "reader": {
89
  "format": "parquet-row-v1",
90
+ "row_group": 7,
91
+ "row": 2817
92
  }
93
  },
94
  "epoch": 0,
95
+ "offset": 0,
96
+ "documents": 6080499
97
  },
98
  "cortex": {
99
+ "page": 191125,
100
+ "row": 60,
101
  "start": null,
102
  "epoch": 0,
103
  "offset": 0,
104
+ "documents": 24464060
105
  }
106
  }
107
  }
latest.json CHANGED
@@ -1 +1 @@
1
- {"step": 725000, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "e71ca4f551a1fff6787ce12f5aa4c155ba5ee5b4116a714d3f9add404be3f44d", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "cc65b927e55c762cf8c194b1c80ab87a382a45d747086f25332683699912d204", "training_config.json": "a5b628e58d936dba526336e6ed1a3b16a703b7373430cc4aca66220c358ed037", "performance.json": "48e3aa7938b07d5e08810d60bc208cc2cb9133c4c39a24151758b79306675379", "shard_scheduler.json": "5d40fdc3e8d0b948086712e99ffe8769eaf1235b617b80b55a7573c5ed79ce15", "coverage.json": "c5dc0b0b83b7968cda35d137af58ee922f5d9e851767895167b255f14ad87398", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "13510bfc04d8fed598fd0d09c1f778fbaebf19cefdd81f23f5ef3b909ef94a63", "folder": "step-000000725000"}
 
1
+ {"step": 730000, "files": {"config.json": "cbc72137da3dae65085a2a8c39ad7634d16bf92d785ca00d283ee82fe32b37e7", "generation_config.json": "54ed0d19f6226888c91157964ede8a892bd5ae175bfad85336152b3d9b6cd7c4", "model.safetensors": "8c53eaad59485c3e7f93abf3a4b58bce24607b53a618b83f0a58c30a6ceb0a27", "tokenization_byte.py": "279b9990f77b1eaedc7c1184fe13d60273a713ead856200f1fb9ae323c7b1e03", "tokenizer_config.json": "0ecbd4fd39e35125800d806af981ba0cacd8a7ba5f02c9e729c4ecc54af6d300", "special_tokens_map.json": "20d4b94d0e657c3f44d9123522c106e7315b6c1a1b6d259478f17bd2918b1263", "byte_vocab.json": "57a1cd2e70dd5b7cc6fd2eaa12ebfcd173c1620fa5d45c2c6282511c29b6a5f6", "trainer.pt": "1650174bff312eebb6355ae334a6607b73e2c555e90fe45a7a399e9567721584", "training_config.json": "75248af0a321738b405c3d51d1d55c7040dbc0785759d95f8dbdf6cd071a6687", "performance.json": "48e3aa7938b07d5e08810d60bc208cc2cb9133c4c39a24151758b79306675379", "shard_scheduler.json": "1ca6461804592e6d06e70f93e961d9b78fbee44b41acd3825f26f6670f746dd6", "coverage.json": "a190366ff62a7a816e8e5cfff69e551cc2bb0b47d066c9abcec59ff0915c60cf", "README.md": "9ce0f9efacbc46177a7e818b6aed274d84e7145ff944b5c7d6ecd5d33b9d4419"}, "signature": "13510bfc04d8fed598fd0d09c1f778fbaebf19cefdd81f23f5ef3b909ef94a63", "folder": "step-000000730000"}
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e71ca4f551a1fff6787ce12f5aa4c155ba5ee5b4116a714d3f9add404be3f44d
3
  size 360622744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c53eaad59485c3e7f93abf3a4b58bce24607b53a618b83f0a58c30a6ceb0a27
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": 75527.92613571833,
26
- "required_pages_per_second_upper_estimate": 0.00522519717900222
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": 9440.990766964791,
40
- "required_pages_per_second_upper_estimate": 0.0039527456838222355
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": 18881.981533929582,
54
- "required_pages_per_second_upper_estimate": 0.003932618802178705
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": 28322.972300894373,
68
- "required_pages_per_second_upper_estimate": 0.0016979189427639779
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": 28322.972300894373,
82
- "required_pages_per_second_upper_estimate": 0.002201986516805363
83
  },
84
  {
85
  "source": "cortex",
@@ -91,9 +91,9 @@
91
  "read_seconds": 0.04061180899998362,
92
  "tokens_per_second": 910315.184299418,
93
  "note": "Bounded fresh sample; full-page latency is an extrapolation.",
94
- "estimated_pages_for_buffer": 22,
95
- "required_token_positions_per_second": 28322.972300894373,
96
- "required_pages_per_second_upper_estimate": 0.3632966778375646
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
@@ -155,27 +155,27 @@
155
  "shard_benchmark_scope": "Local packing/read measurements; upload capacity measured only on real data commits. Targets larger than available pages are sample-limited.",
156
  "estimated_fresh_generation_batch_seconds": 2.558590814491007,
157
  "selected": {
158
- "prefetch_batches": 62,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 720750,
161
- "producer_batch_seconds": 0.3763775794688158,
162
- "producer_p95_seconds": 0.17173908699987805,
163
- "gpu_update_seconds": 0.9761687335028373,
164
- "producer_batches_per_second": 2.6569064007779284,
165
- "gpu_batches_per_second": 1.0244130606515616,
166
- "producer_headroom": 2.593588956282972,
167
  "sustainable": true,
168
- "buffer_seconds": 60.52246147717591,
169
  "buffer_drain_seconds": null,
170
  "shard_target_mib": 16,
171
- "upload_pages_per_commit": 85,
172
  "max_pending_pages": 4096,
173
- "generation_bytes_per_second": 242621.26608918482,
174
- "upload_bytes_per_second": 3845384.767186985,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
- "estimated_full_shards_per_hour": 52.060875768725,
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": [
@@ -226,6 +226,246 @@
226
  "bytes": 9722174,
227
  "pages": 29,
228
  "shards": 3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
229
  }
230
  ],
231
  "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": 76313.75839753207,
26
+ "required_pages_per_second_upper_estimate": 0.005279562878256555
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": 9539.21979969151,
40
+ "required_pages_per_second_upper_estimate": 0.003993872128569453
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": 19078.43959938302,
54
+ "required_pages_per_second_upper_estimate": 0.003973535836265066
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": 28617.659399074528,
68
+ "required_pages_per_second_upper_estimate": 0.0017155849843387327
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": 28617.659399074528,
82
+ "required_pages_per_second_upper_estimate": 0.002224897142497311
83
  },
84
  {
85
  "source": "cortex",
 
91
  "read_seconds": 0.04061180899998362,
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": 28617.659399074528,
96
+ "required_pages_per_second_upper_estimate": 0.3670766075226655
97
  }
98
  ],
99
  "gpu_measurement": "Selected forward/backward estimate; replaced by live successful update timings.",
 
155
  "shard_benchmark_scope": "Local packing/read measurements; upload capacity measured only on real data commits. Targets larger than available pages are sample-limited.",
156
  "estimated_fresh_generation_batch_seconds": 2.558590814491007,
157
  "selected": {
158
+ "prefetch_batches": 63,
159
  "prefetch_capacity_limit": 512,
160
  "batch_memory_estimate_bytes": 720750,
161
+ "producer_batch_seconds": 0.10201518258595854,
162
+ "producer_p95_seconds": 0.18670680400100537,
163
+ "gpu_update_seconds": 0.9661167468117297,
164
+ "producer_batches_per_second": 9.802462483046527,
165
+ "gpu_batches_per_second": 1.035071592848471,
166
+ "producer_headroom": 9.47032316486494,
167
  "sustainable": true,
168
+ "buffer_seconds": 60.86535504913897,
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": 342045.38893318584,
174
+ "upload_bytes_per_second": 4834421.459987242,
175
  "upload_sustainable": true,
176
  "logical_page_rows": 4096,
177
  "logical_cortex_page_rows": 128,
178
+ "estimated_full_shards_per_hour": 73.39497805592234,
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": [
 
226
  "bytes": 9722174,
227
  "pages": 29,
228
  "shards": 3
229
+ },
230
+ {
231
+ "seconds": 1.4900774309999179,
232
+ "bytes": 2616245,
233
+ "pages": 36,
234
+ "shards": 2
235
+ },
236
+ {
237
+ "seconds": 1.2741041360004601,
238
+ "bytes": 11482818,
239
+ "pages": 21,
240
+ "shards": 3
241
+ },
242
+ {
243
+ "seconds": 1.076980086999356,
244
+ "bytes": 1228924,
245
+ "pages": 35,
246
+ "shards": 1
247
+ },
248
+ {
249
+ "seconds": 1.3698066919996563,
250
+ "bytes": 11420874,
251
+ "pages": 33,
252
+ "shards": 4
253
+ },
254
+ {
255
+ "seconds": 1.0434529929998462,
256
+ "bytes": 6181688,
257
+ "pages": 36,
258
+ "shards": 2
259
+ },
260
+ {
261
+ "seconds": 1.099565554000037,
262
+ "bytes": 1093470,
263
+ "pages": 35,
264
+ "shards": 1
265
+ },
266
+ {
267
+ "seconds": 1.3861621640007797,
268
+ "bytes": 7683491,
269
+ "pages": 35,
270
+ "shards": 4
271
+ },
272
+ {
273
+ "seconds": 1.2777914989992496,
274
+ "bytes": 16450745,
275
+ "pages": 38,
276
+ "shards": 4
277
+ },
278
+ {
279
+ "seconds": 0.9403941730006409,
280
+ "bytes": 2445089,
281
+ "pages": 28,
282
+ "shards": 2
283
+ },
284
+ {
285
+ "seconds": 1.042657115000111,
286
+ "bytes": 6054602,
287
+ "pages": 36,
288
+ "shards": 2
289
+ },
290
+ {
291
+ "seconds": 1.0668450510002003,
292
+ "bytes": 1106491,
293
+ "pages": 35,
294
+ "shards": 1
295
+ },
296
+ {
297
+ "seconds": 1.397956458000408,
298
+ "bytes": 11076576,
299
+ "pages": 34,
300
+ "shards": 4
301
+ },
302
+ {
303
+ "seconds": 1.1856196190001356,
304
+ "bytes": 6180989,
305
+ "pages": 35,
306
+ "shards": 2
307
+ },
308
+ {
309
+ "seconds": 1.3201266620008028,
310
+ "bytes": 6761235,
311
+ "pages": 30,
312
+ "shards": 2
313
+ },
314
+ {
315
+ "seconds": 1.2538779149999755,
316
+ "bytes": 7546399,
317
+ "pages": 30,
318
+ "shards": 3
319
+ },
320
+ {
321
+ "seconds": 1.2998433860002478,
322
+ "bytes": 5962432,
323
+ "pages": 35,
324
+ "shards": 2
325
+ },
326
+ {
327
+ "seconds": 1.2994795300000987,
328
+ "bytes": 7606454,
329
+ "pages": 39,
330
+ "shards": 3
331
+ },
332
+ {
333
+ "seconds": 1.54305899999963,
334
+ "bytes": 5745307,
335
+ "pages": 28,
336
+ "shards": 2
337
+ },
338
+ {
339
+ "seconds": 0.9886994910002613,
340
+ "bytes": 1117526,
341
+ "pages": 35,
342
+ "shards": 1
343
+ },
344
+ {
345
+ "seconds": 1.8489439929999207,
346
+ "bytes": 17752294,
347
+ "pages": 35,
348
+ "shards": 5
349
+ },
350
+ {
351
+ "seconds": 1.054843687000357,
352
+ "bytes": 1282245,
353
+ "pages": 35,
354
+ "shards": 1
355
+ },
356
+ {
357
+ "seconds": 1.0156906100000924,
358
+ "bytes": 5808038,
359
+ "pages": 28,
360
+ "shards": 2
361
+ },
362
+ {
363
+ "seconds": 1.466896263000308,
364
+ "bytes": 7685001,
365
+ "pages": 38,
366
+ "shards": 3
367
+ },
368
+ {
369
+ "seconds": 1.0174859720000313,
370
+ "bytes": 4930803,
371
+ "pages": 30,
372
+ "shards": 2
373
+ },
374
+ {
375
+ "seconds": 1.3428281450014765,
376
+ "bytes": 7443989,
377
+ "pages": 38,
378
+ "shards": 3
379
+ },
380
+ {
381
+ "seconds": 1.0973035269998945,
382
+ "bytes": 6756483,
383
+ "pages": 32,
384
+ "shards": 2
385
+ },
386
+ {
387
+ "seconds": 1.110015972999463,
388
+ "bytes": 6170503,
389
+ "pages": 36,
390
+ "shards": 2
391
+ },
392
+ {
393
+ "seconds": 1.4518775689994072,
394
+ "bytes": 12256015,
395
+ "pages": 37,
396
+ "shards": 4
397
+ },
398
+ {
399
+ "seconds": 0.9289161779997812,
400
+ "bytes": 848234,
401
+ "pages": 26,
402
+ "shards": 1
403
+ },
404
+ {
405
+ "seconds": 1.4199929900005372,
406
+ "bytes": 7653735,
407
+ "pages": 38,
408
+ "shards": 3
409
+ },
410
+ {
411
+ "seconds": 3.9366146709999157,
412
+ "bytes": 6393409,
413
+ "pages": 36,
414
+ "shards": 2
415
+ },
416
+ {
417
+ "seconds": 1.550046686999849,
418
+ "bytes": 10810645,
419
+ "pages": 27,
420
+ "shards": 3
421
+ },
422
+ {
423
+ "seconds": 1.2842068059999292,
424
+ "bytes": 7447718,
425
+ "pages": 38,
426
+ "shards": 3
427
+ },
428
+ {
429
+ "seconds": 1.3457034299990482,
430
+ "bytes": 1181953,
431
+ "pages": 35,
432
+ "shards": 1
433
+ },
434
+ {
435
+ "seconds": 1.210031186001288,
436
+ "bytes": 5583238,
437
+ "pages": 27,
438
+ "shards": 2
439
+ },
440
+ {
441
+ "seconds": 1.2692713039996306,
442
+ "bytes": 11755459,
443
+ "pages": 37,
444
+ "shards": 4
445
+ },
446
+ {
447
+ "seconds": 1.0984560849992704,
448
+ "bytes": 1168159,
449
+ "pages": 35,
450
+ "shards": 1
451
+ },
452
+ {
453
+ "seconds": 1.9164282879992243,
454
+ "bytes": 13374744,
455
+ "pages": 37,
456
+ "shards": 5
457
+ },
458
+ {
459
+ "seconds": 0.928293617998861,
460
+ "bytes": 992050,
461
+ "pages": 33,
462
+ "shards": 1
463
+ },
464
+ {
465
+ "seconds": 1.3297191130004649,
466
+ "bytes": 10315993,
467
+ "pages": 33,
468
+ "shards": 3
469
  }
470
  ],
471
  "cache_key": {
trainer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc65b927e55c762cf8c194b1c80ab87a382a45d747086f25332683699912d204
3
- size 721769087
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1650174bff312eebb6355ae334a6607b73e2c555e90fe45a7a399e9567721584
3
+ size 721792959
training_config.json CHANGED
@@ -53,7 +53,7 @@
53
  "cpu_threads": 4,
54
  "page_rows": 4096,
55
  "cortex_page_rows": 128,
56
- "prefetch_batches": 62,
57
  "prefetch_low_water": 31,
58
  "memory_pages": 12,
59
  "cache_gb": 20,
 
53
  "cpu_threads": 4,
54
  "page_rows": 4096,
55
  "cortex_page_rows": 128,
56
+ "prefetch_batches": 63,
57
  "prefetch_low_water": 31,
58
  "memory_pages": 12,
59
  "cache_gb": 20,