Datasets:

Modalities:
Text
Languages:
English
ArXiv:
DOI:
License:
BerenMillidge commited on
Commit
8792875
1 Parent(s): 2648f81

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -4
README.md CHANGED
@@ -64,7 +64,6 @@ Zyda also outperforms Dolma, RefinedWeb, and Fineweb on 1.4B models trained on 5
64
 
65
  According to our evaluations, Zyda is the most performant per-token open dataset available in its non-starcoder variant on language tasks and tying with fineweb otherwise.
66
 
67
-
68
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65c05e75c084467acab2f84a/fXaQAOBDJpoaAr1clfTel.png)
69
 
70
  ## How to download
@@ -106,10 +105,8 @@ Dataset fields:
106
  - `filtering_features`: precomputed values of different features that were used for filtering (converted to json string)
107
  - `source_other`: metadata from the source dataset (converted to json string)
108
 
109
- <center>
110
- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65c05e75c084467acab2f84a/dQV8zNTNCx1xMMT-iupY6.png)
111
- </center>
112
 
 
113
 
114
  ### Source Data
115
 
 
64
 
65
  According to our evaluations, Zyda is the most performant per-token open dataset available in its non-starcoder variant on language tasks and tying with fineweb otherwise.
66
 
 
67
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65c05e75c084467acab2f84a/fXaQAOBDJpoaAr1clfTel.png)
68
 
69
  ## How to download
 
105
  - `filtering_features`: precomputed values of different features that were used for filtering (converted to json string)
106
  - `source_other`: metadata from the source dataset (converted to json string)
107
 
 
 
 
108
 
109
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65c05e75c084467acab2f84a/dQV8zNTNCx1xMMT-iupY6.png)
110
 
111
  ### Source Data
112