Commit
•
25c913a
1
Parent(s):
76f4923
Upload tokenizer
Browse files- tokenizer_config.json +5 -0
tokenizer_config.json
CHANGED
|
@@ -3,7 +3,12 @@
|
|
| 3 |
"bos_token": "<|endoftext|>",
|
| 4 |
"clean_up_tokenization_spaces": true,
|
| 5 |
"eos_token": "<|endoftext|>",
|
|
|
|
| 6 |
"model_max_length": 8192,
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
"tokenizer_class": "GPTNeoXTokenizer",
|
| 8 |
"unk_token": "<|endoftext|>"
|
| 9 |
}
|
|
|
|
| 3 |
"bos_token": "<|endoftext|>",
|
| 4 |
"clean_up_tokenization_spaces": true,
|
| 5 |
"eos_token": "<|endoftext|>",
|
| 6 |
+
"max_length": null,
|
| 7 |
"model_max_length": 8192,
|
| 8 |
+
"pad_to_multiple_of": null,
|
| 9 |
+
"pad_token": "<|endoftext|>",
|
| 10 |
+
"pad_token_type_id": 0,
|
| 11 |
+
"padding_side": "left",
|
| 12 |
"tokenizer_class": "GPTNeoXTokenizer",
|
| 13 |
"unk_token": "<|endoftext|>"
|
| 14 |
}
|