File size: 312 Bytes
5b7e9c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
---
A mixture of datasets focused on Source/Intermediate Representation training for multilingual code generation.
For SCU CSEN346.


**Usage**:
Pass to continued_pretrain.py with
```
  --dataset_name './tokenized_dataset'
```


Sources:
UKPLab/SLTrans
bigcode/the-stack
allenai/peS2o
open-web-math/open-web-math