Update README.md
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ pipeline_tag: image-text-to-text
|
|
| 18 |
|
| 19 |
**Hydra-Turbo** is a dynamically uncensored and abliterated version of [`datalab-to/lift`](https://huggingface.co/datalab-to/lift) (which is built on the Qwen2VL architecture). This model was created using the [Heretic framework](https://github.com/p-e-w/heretic), employing advanced orthogonal weight ablation to remove refusal vectors while completely preserving the model's underlying reasoning, intelligence, and multimodal capabilities.
|
| 20 |
|
| 21 |
-
##
|
| 22 |
Unlike traditional fine-tuning or full RLHF—which can cause "brain damage" to a model by catastrophically forgetting knowledge—Hydra-Turbo was optimized using a Pareto-optimal search across multiple ablation vectors.
|
| 23 |
|
| 24 |
The selected optimal trial yielded the following metrics on the standard refusal test set:
|
|
@@ -28,12 +28,12 @@ The selected optimal trial yielded the following metrics on the standard refusal
|
|
| 28 |
|
| 29 |
At a KL divergence of just `0.0275`, the structural integrity and logic capabilities of the base model are perfectly intact. It simply no longer refuses instructions.
|
| 30 |
|
| 31 |
-
##
|
| 32 |
- **Multimodal Uncensored Generation:** Can process images and text without artificial guardrails.
|
| 33 |
- **Qwen2VL Architecture:** Inherits state-of-the-art vision parsing capabilities.
|
| 34 |
- **Drop-in Replacement:** Compatible with any standard `transformers` or `vLLM` pipeline that supports Qwen2VL.
|
| 35 |
|
| 36 |
-
##
|
| 37 |
|
| 38 |
### Via HuggingFace Transformers
|
| 39 |
You can load the model exactly as you would any Qwen2VL model, ensuring you have `pillow` and `torchvision` installed:
|
|
@@ -58,6 +58,6 @@ Because this model has had its safety guardrails mathematically ablated, it is h
|
|
| 58 |
- **Hallucinations:** As with all LLMs, the model can confidently hallucinate incorrect information.
|
| 59 |
- **Use Case:** This model is intended for research, creative writing, and local deployments where unrestricted inference is required. Users are solely responsible for the content generated.
|
| 60 |
|
| 61 |
-
##
|
| 62 |
- **Base Model:** [`datalab-to/lift`](https://huggingface.co/datalab-to/lift)
|
| 63 |
- **Ablation Framework:** [Heretic by p-e-w](https://github.com/p-e-w/heretic)
|
|
|
|
| 18 |
|
| 19 |
**Hydra-Turbo** is a dynamically uncensored and abliterated version of [`datalab-to/lift`](https://huggingface.co/datalab-to/lift) (which is built on the Qwen2VL architecture). This model was created using the [Heretic framework](https://github.com/p-e-w/heretic), employing advanced orthogonal weight ablation to remove refusal vectors while completely preserving the model's underlying reasoning, intelligence, and multimodal capabilities.
|
| 20 |
|
| 21 |
+
## Ablation Methodology & Metrics
|
| 22 |
Unlike traditional fine-tuning or full RLHF—which can cause "brain damage" to a model by catastrophically forgetting knowledge—Hydra-Turbo was optimized using a Pareto-optimal search across multiple ablation vectors.
|
| 23 |
|
| 24 |
The selected optimal trial yielded the following metrics on the standard refusal test set:
|
|
|
|
| 28 |
|
| 29 |
At a KL divergence of just `0.0275`, the structural integrity and logic capabilities of the base model are perfectly intact. It simply no longer refuses instructions.
|
| 30 |
|
| 31 |
+
## Key Features
|
| 32 |
- **Multimodal Uncensored Generation:** Can process images and text without artificial guardrails.
|
| 33 |
- **Qwen2VL Architecture:** Inherits state-of-the-art vision parsing capabilities.
|
| 34 |
- **Drop-in Replacement:** Compatible with any standard `transformers` or `vLLM` pipeline that supports Qwen2VL.
|
| 35 |
|
| 36 |
+
## Usage
|
| 37 |
|
| 38 |
### Via HuggingFace Transformers
|
| 39 |
You can load the model exactly as you would any Qwen2VL model, ensuring you have `pillow` and `torchvision` installed:
|
|
|
|
| 58 |
- **Hallucinations:** As with all LLMs, the model can confidently hallucinate incorrect information.
|
| 59 |
- **Use Case:** This model is intended for research, creative writing, and local deployments where unrestricted inference is required. Users are solely responsible for the content generated.
|
| 60 |
|
| 61 |
+
## Acknowledgements
|
| 62 |
- **Base Model:** [`datalab-to/lift`](https://huggingface.co/datalab-to/lift)
|
| 63 |
- **Ablation Framework:** [Heretic by p-e-w](https://github.com/p-e-w/heretic)
|