adding tokenizer from santacoder
Browse files- tokenizer.json +0 -0
- tokenizer_config.json +7 -0
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"errors": "replace",
|
| 3 |
+
"tokenizer_class": "GPT2TokenizerFast",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"model_max_length": 2048
|
| 7 |
+
}
|