Instructions to use Nj-1111/Copernicus-Tokenizer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Nj-1111/Copernicus-Tokenizer with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Nj-1111/Copernicus-Tokenizer", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Delete tokenizer_config.json
Browse files- tokenizer_config.json +0 -28
tokenizer_config.json
DELETED
|
@@ -1,28 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 3 |
-
"model_max_length": 4096,
|
| 4 |
-
"bos_token": "<|endoftext|>",
|
| 5 |
-
"eos_token": "<|endoftext|>",
|
| 6 |
-
"unk_token": "<|unk|>",
|
| 7 |
-
"pad_token": "<|pad|>",
|
| 8 |
-
"additional_special_tokens": [
|
| 9 |
-
"<think>",
|
| 10 |
-
"</think>",
|
| 11 |
-
"<scratchpad>",
|
| 12 |
-
"</scratchpad>",
|
| 13 |
-
"<verify>",
|
| 14 |
-
"</verify>",
|
| 15 |
-
"<reflect>",
|
| 16 |
-
"</reflect>",
|
| 17 |
-
"<|user|>",
|
| 18 |
-
"<|assistant|>",
|
| 19 |
-
"<|system|>",
|
| 20 |
-
"<|tool_call|>",
|
| 21 |
-
"<|tool_result|>",
|
| 22 |
-
"<|sep|>",
|
| 23 |
-
"<|im_start|>",
|
| 24 |
-
"<|im_end|>"
|
| 25 |
-
],
|
| 26 |
-
"clean_up_tokenization_spaces": false,
|
| 27 |
-
"add_prefix_space": false
|
| 28 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|