23 lines
833 B
Markdown
23 lines
833 B
Markdown
|
---
|
||
|
|
library_name: transformers
|
||
|
|
license: apache-2.0
|
||
|
|
datasets:
|
||
|
|
- Locutusque/hyperion-v2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
---
|
||
|
|
# Description
|
||
|
|
Further fine-tuned Locutusque/Hyperion-2.0-Mistral-7B at a higher learning rate. This was done to see if performance increased. Read Locutusque/Hyperion-2.0-Mistral-7B's model card for more information. Slight performance gain was observed. More checkpoints will be released in the future.
|
||
|
|
|
||
|
|
# Disclaimer
|
||
|
|
|
||
|
|
This model is very compliant. It will respond to any request without refusal. If you intend to deploy this model at an enterprise level, I would recommend aligning this model using DPO.
|
||
|
|
|
||
|
|
## Quants
|
||
|
|
|
||
|
|
ExLlamaV2: https://huggingface.co/bartowski/Hyperion-2.1-Mistral-7B-exl2
|
||
|
|
|
||
|
|
GGUF: https://huggingface.co/bartowski/Hyperion-2.1-Mistral-7B-GGUF
|
||
|
|
|
||
|
|
AWQ: https://huggingface.co/solidrust/Hyperion-2.1-Mistral-7B-AWQ
|