Update tokenizer, configuration and training reports
Browse files
reports/sources/finemath/manifest.json
CHANGED
|
@@ -6,13 +6,17 @@
|
|
| 6 |
{
|
| 7 |
"path": "finemath-4plus/train-00045-of-00064.parquet",
|
| 8 |
"sha256": "2c3e25eccc80c71af05c1c27a260fb25d5743169dbace5e3ecdd18e6b96c277e"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
}
|
| 10 |
],
|
| 11 |
-
"tokens":
|
| 12 |
"validation_tokens": 503280,
|
| 13 |
-
"docs":
|
| 14 |
-
"overlap_rejected":
|
| 15 |
-
"sha256": "
|
| 16 |
"validation_sha256": "180832b085910a6a2b31ae42e42118edd380b7007c23268d60f6d69f89e01688",
|
| 17 |
"tokenizer_sha256": "d1ffe47f6fd8cc19ba7c3e5d2537699d6a99d09b584d3319944f4e40fe6c44ea",
|
| 18 |
"split": "document hash 1% held out",
|
|
|
|
| 6 |
{
|
| 7 |
"path": "finemath-4plus/train-00045-of-00064.parquet",
|
| 8 |
"sha256": "2c3e25eccc80c71af05c1c27a260fb25d5743169dbace5e3ecdd18e6b96c277e"
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"path": "finemath-4plus/train-00008-of-00064.parquet",
|
| 12 |
+
"sha256": "39224cfcc751afbc04671bc9491266a90d58c30c27fd58f93551db8766ca96c6"
|
| 13 |
}
|
| 14 |
],
|
| 15 |
+
"tokens": 250135965,
|
| 16 |
"validation_tokens": 503280,
|
| 17 |
+
"docs": 165671,
|
| 18 |
+
"overlap_rejected": 41,
|
| 19 |
+
"sha256": "c1ec8594775275bd0069633ed4e349ea43c6233abd345caf76b51b097f45a0d1",
|
| 20 |
"validation_sha256": "180832b085910a6a2b31ae42e42118edd380b7007c23268d60f6d69f89e01688",
|
| 21 |
"tokenizer_sha256": "d1ffe47f6fd8cc19ba7c3e5d2537699d6a99d09b584d3319944f4e40fe6c44ea",
|
| 22 |
"split": "document hash 1% held out",
|