Thomas Seri commited on
Upload tokenizer_config.json with huggingface_hub
Browse files- tokenizer_config.json +9 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "gpt2",
|
| 3 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 4 |
+
"bos_token": "<|bos|>",
|
| 5 |
+
"eos_token": "<|eos|>",
|
| 6 |
+
"unk_token": "<|unk|>",
|
| 7 |
+
"pad_token": "<|pad|>",
|
| 8 |
+
"clean_up_tokenization_spaces": false
|
| 9 |
+
}
|