Instructions to use ethicalabs/TowerInstruct-7B-v0.2-mlx-4Bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use ethicalabs/TowerInstruct-7B-v0.2-mlx-4Bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir TowerInstruct-7B-v0.2-mlx-4Bit ethicalabs/TowerInstruct-7B-v0.2-mlx-4Bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Update tokenizer_config.json
Browse files- tokenizer_config.json +1 -2
tokenizer_config.json
CHANGED
|
@@ -89,12 +89,11 @@
|
|
| 89 |
"clean_up_tokenization_spaces": false,
|
| 90 |
"cls_token": "<CLS>",
|
| 91 |
"eos_token": "<|im_end|>",
|
| 92 |
-
"extra_special_tokens": {},
|
| 93 |
"legacy": false,
|
| 94 |
"mask_token": "<MASK>",
|
| 95 |
"model_max_length": 1000000000000000019884624838656,
|
| 96 |
"pad_token": "<PAD>",
|
| 97 |
-
"padding_side": "
|
| 98 |
"sep_token": "<SEP>",
|
| 99 |
"sp_model_kwargs": {},
|
| 100 |
"tokenizer_class": "LlamaTokenizer",
|
|
|
|
| 89 |
"clean_up_tokenization_spaces": false,
|
| 90 |
"cls_token": "<CLS>",
|
| 91 |
"eos_token": "<|im_end|>",
|
|
|
|
| 92 |
"legacy": false,
|
| 93 |
"mask_token": "<MASK>",
|
| 94 |
"model_max_length": 1000000000000000019884624838656,
|
| 95 |
"pad_token": "<PAD>",
|
| 96 |
+
"padding_side": "right",
|
| 97 |
"sep_token": "<SEP>",
|
| 98 |
"sp_model_kwargs": {},
|
| 99 |
"tokenizer_class": "LlamaTokenizer",
|