Instructions to use neuronstarml/zephyr-support-chatbot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use neuronstarml/zephyr-support-chatbot with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("TheBloke/zephyr-7B-alpha-GPTQ") model = PeftModel.from_pretrained(base_model, "neuronstarml/zephyr-support-chatbot") - Notebooks
- Google Colab
- Kaggle
Download tokenizer.json from neuronstarml/zephyr-support-chatbot: direct link, hf CLI and curl.
- Browser
- Download file 1.8 MB
-
https://huggingface.co/neuronstarml/zephyr-support-chatbot/resolve/main/tokenizer.json
- Command line
-
hf download hf://neuronstarml/zephyr-support-chatbot/tokenizer.json
-
curl -L -o tokenizer.json https://huggingface.co/neuronstarml/zephyr-support-chatbot/resolve/main/tokenizer.json
1.8 MB
File too large to display, you can check the raw version instead.