Instructions to use HMC83/request_writer_sbp_2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use HMC83/request_writer_sbp_2 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("HMC83/skunkbadger-phoenix") model = PeftModel.from_pretrained(base_model, "HMC83/request_writer_sbp_2") - Transformers
How to use HMC83/request_writer_sbp_2 with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("HMC83/request_writer_sbp_2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from HMC83/request_writer_sbp_2: direct link, hf CLI and curl.
- Browser
- Download file 890 Bytes
-
https://huggingface.co/HMC83/request_writer_sbp_2/resolve/main/README.md
- Command line
-
hf download hf://HMC83/request_writer_sbp_2/README.md
-
curl -L -H "Authorization: Bearer $HF_TOKEN" -o README.md https://huggingface.co/HMC83/request_writer_sbp_2/resolve/main/README.md
890 Bytes
| base_model: HMC83/skunkbadger-phoenix | |
| library_name: peft | |
| tags: | |
| - base_model:adapter:HMC83/skunkbadger-phoenix | |
| - lora | |
| - sft | |
| - transformers | |
| - trl | |
| datasets: | |
| - HMC83/synthetic_foi_requests | |
| ----- | |
| Tiny 22M param model based on GPT2 | |
| ### Recommended Prompt Template | |
| This model expects a prompt following the Alpaca instruction format. The user's query should be placed after `### Instruction:`. | |
| ``` | |
| You are an expert at writing formal Freedom of Information requests to UK public authorities. Write clear, specific, professional requests that comply with FOI requirements and use accessible language. ### Instruction: | |
| Generate a formal Freedom of Information request to {authority_name} using these keywords: {keywords} | |
| ### Response: | |
| ``` | |
| - Replace `{authority_name}` with the name of the public body. | |
| - Replace `{keywords}` with a comma-separated list of relevant keywords. | |
| ----- |