BerenMillidge
commited on
Commit
•
f692a03
1
Parent(s):
c019136
Update README.md
Browse files
README.md
CHANGED
@@ -134,9 +134,13 @@ StarCoder: https://huggingface.co/datasets/bigcode/starcoderdata
|
|
134 |
|
135 |
<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
|
136 |
|
|
|
137 |
|
|
|
138 |
|
|
|
139 |
|
|
|
140 |
|
141 |
|
142 |
#### Personal and Sensitive Information
|
|
|
134 |
|
135 |
<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
|
136 |
|
137 |
+
Zyda was created using a two stage post-processing pipeline consisting of *filtering* and *deduplication*.
|
138 |
|
139 |
+
For the filtering stage, we utilized a set of hand-crafted and tuned filters derived from a number of sources such as C4, RedPajama, and Gopher, in addition to our own filters.
|
140 |
|
141 |
+
For the deduplication stage, we used minhash approximate deduplication. We deduplicated on 13-grams and used a minhash signature size of 128 and filtered out documents above a Jaccard similarity of 0.4.
|
142 |
|
143 |
+
For full details on our data processing see the technical report.
|
144 |
|
145 |
|
146 |
#### Personal and Sensitive Information
|