Update README.md
Browse files
README.md
CHANGED
@@ -12,19 +12,19 @@ Hugging Face introduced FineWeb V1.1, a large-scale dataset for LLM pre-training
|
|
12 |
|
13 |
We started with the goal of distilling 10T+ high quality tokens from FineWeb V1.1, so that we get sufficiently large number of quality tokens suitable for Stage-1 pre-training. Unlike the FineWeb.Edu families, which rely on a single quality annotator and perform aggressive filtering, we developed a multi-faceted ensemble of quality annotators to enable fine-grained quality filtering. This allowed us to achieve a finer trade-off between the quality and quantity of the tokens retained. While the GneissWeb recipe is focused at obtaining 10T+ high quality tokens suitable for Stage-1 pre-training, it is also possible to adapt the recipe by tuning filtering parameters to produce smaller and higher quality datasets fit for Stage-2 kind of training.
|
14 |
|
15 |
-
An Overview of the GneissWeb Recipe
|
16 |
|
17 |
-
The GneissWeb dataset was obtained by applying the following processing steps to Fineweb :
|
18 |
|
19 |
-
Exact substring deduplication at line level
|
20 |
|
21 |
-
Custom built Fasttext quality filter
|
22 |
|
23 |
-
Custom built Fasttext category classifier
|
24 |
|
25 |
-
Custom built Category-aware readability score quality filter
|
26 |
|
27 |
-
Custom built Category-aware extreme_tokenized quality filter
|
28 |
|
29 |
These were applied in the order shown in the Fig 1
|
30 |
|
|
|
12 |
|
13 |
We started with the goal of distilling 10T+ high quality tokens from FineWeb V1.1, so that we get sufficiently large number of quality tokens suitable for Stage-1 pre-training. Unlike the FineWeb.Edu families, which rely on a single quality annotator and perform aggressive filtering, we developed a multi-faceted ensemble of quality annotators to enable fine-grained quality filtering. This allowed us to achieve a finer trade-off between the quality and quantity of the tokens retained. While the GneissWeb recipe is focused at obtaining 10T+ high quality tokens suitable for Stage-1 pre-training, it is also possible to adapt the recipe by tuning filtering parameters to produce smaller and higher quality datasets fit for Stage-2 kind of training.
|
14 |
|
15 |
+
**An Overview of the GneissWeb Recipe**
|
16 |
|
17 |
+
The GneissWeb dataset was obtained by applying the following processing steps to Fineweb :
|
18 |
|
19 |
+
Exact substring deduplication at line level
|
20 |
|
21 |
+
Custom built Fasttext quality filter
|
22 |
|
23 |
+
Custom built Fasttext category classifier
|
24 |
|
25 |
+
Custom built Category-aware readability score quality filter
|
26 |
|
27 |
+
Custom built Category-aware extreme_tokenized quality filter
|
28 |
|
29 |
These were applied in the order shown in the Fig 1
|
30 |
|