22 lines
687 B
Markdown
22 lines
687 B
Markdown
|
---
|
||
|
|
license: llama3.1
|
||
|
|
base_model: meta-llama/Llama-3.1-8B-Instruct
|
||
|
|
base_model_relation: merge
|
||
|
|
library_name: transformers
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
tags:
|
||
|
|
- safety
|
||
|
|
- alignment
|
||
|
|
---
|
||
|
|
|
||
|
|
# HARC — Llama-3.1-8B-Instruct
|
||
|
|
|
||
|
|
HARC safety-alignment LoRA merged into `meta-llama/Llama-3.1-8B-Instruct` (full standalone model).
|
||
|
|
Part of the [HARC](https://github.com/microsoft/HARC) release; see paper arXiv:2607.00572.
|
||
|
|
|
||
|
|
```python
|
||
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
|
model = AutoModelForCausalLM.from_pretrained("microsoft/HARC-Llama-3.1-8B-Instruct", torch_dtype="bfloat16", device_map="auto")
|
||
|
|
tok = AutoTokenizer.from_pretrained("microsoft/HARC-Llama-3.1-8B-Instruct")
|
||
|
|
```
|