Datasets:

Languages:
English
ArXiv:
License:
bhatta1 commited on
Commit
ee4b184
·
verified ·
1 Parent(s): 8cadebb

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -2
README.md CHANGED
@@ -28,8 +28,6 @@ We started with the goal of distilling 10T+ high quality tokens from FineWeb V1.
28
 
29
  These were applied in the order shown in the Fig 1
30
 
31
- ![GneissWeb3.png](GneissWeb3.png)
32
-
33
  <img src="GneissWeb3.png" alt="GneissWeb3" style="width:1000px;"/>
34
 
35
  Figure 1 : GneissWeb recipe
@@ -44,6 +42,7 @@ The baselines from which equivalent data was subsampled and used for this compar
44
 
45
  ![Fig2.jpg](Fig2.jpg)
46
 
 
47
 
48
  Fig 2 shows how the subsamples were created for the Fineweb baselines as well for GneissWeb. A similar strategy as for the creation of the Fineweb baseline was used for other baselines too
49
 
 
28
 
29
  These were applied in the order shown in the Fig 1
30
 
 
 
31
  <img src="GneissWeb3.png" alt="GneissWeb3" style="width:1000px;"/>
32
 
33
  Figure 1 : GneissWeb recipe
 
42
 
43
  ![Fig2.jpg](Fig2.jpg)
44
 
45
+ <img src="Fig2.jng" alt="Fig2.jpg" style="width:1000px;"/>
46
 
47
  Fig 2 shows how the subsamples were created for the Fineweb baselines as well for GneissWeb. A similar strategy as for the creation of the Fineweb baseline was used for other baselines too
48