|
--- |
|
base_model: meta-llama/Meta-Llama-3.1-70B-Instruct |
|
library_name: peft |
|
--- |
|
|
|
# Model Card for Model ID |
|
|
|
This LoRA adapter was extracted from |
|
[mlabonne/Meta-Llama-3.1-70B-Instruct-lorablated](https://huggingface.co./mlabonne/Meta-Llama-3.1-70B-Instruct-lorablated) |
|
and uses |
|
[meta-llama/Llama-3.1-70B-Instruct](https://huggingface.co./meta-llama/Llama-3.1-70B-Instruct) |
|
as a base. |
|
|
|
## Model Details |
|
|
|
The model was extracted by running |
|
[mlabonne/harmful_behaviors](https://huggingface.co./datasets/mlabonne/harmful_behaviors) |
|
and the user prompts (but not assistant responses or system messages) from |
|
[Guilherme34/uncensor](https://huggingface.co./datasets/Guilherme34/uncensor) |
|
through the original abliterated model to generate a dataset of |
|
prompt/completion pairs, and was trained for 2 epochs on a 8xA100s with Axolotl |
|
using FSDP. Since the original abliterated model isn't perfect at avoiding |
|
refusals, the dataset was cleaned to remove the few refusals generated prior to |
|
training. |
|
|
|
### Model Description |
|
|
|
- **Developed by:** @reissbaker |
|
- **Funded by:** Synthetic Lab |
|
- **License:** Apache 2.0 |
|
- **Finetuned from model:** Llama 3.1 70B Instruct |
|
|
|
## How to Get Started with the Model |
|
|
|
Run the model with one click on [glhf.chat](https://glhf.chat). |
|
|
|
|
|
#### Training Hyperparameters |
|
|
|
* BF16 mixed-precision |
|
* 4e-4 LR |
|
* Linear LR schedule |
|
* Fused AdamW optimizer |
|
|
