Downloads · 30 days
3
7% of all-time downloads
Mira-v1.13-27B-custom-heretic-1
Lambent/Mira-v1.13-27B-custom-heretic-1
Mira-v1.13-27B-custom-heretic-1 is a image-text-to-text model from Lambent. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers.
All-time downloads
45
Public
Parameters
27.4B
54.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors54.9 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- Model type
- gemma3
- Access
- Public
- Created
- Nov 19, 2025
- Updated
- Nov 19, 2025
- SHA
- 22c799a6
Try a prompt
Base models
Model card
From the Hugging Face model README
This is a disclaimer-reduced version of Lambent/Mira-v1.12-Ties-27B, made using Heretic v1.0.1
Abliteration parameters
| Parameter | Value |
|---|---|
| direction_index | 36.76 |
| attn.o_proj.max_weight | 1.14 |
| attn.o_proj.max_weight_position | 45.50 |
| attn.o_proj.min_weight | 1.01 |
| attn.o_proj.min_weight_distance | 31.61 |
| mlp.down_proj.max_weight | 1.24 |
| mlp.down_proj.max_weight_position | 37.32 |
| mlp.down_proj.min_weight | 0.36 |
| mlp.down_proj.min_weight_distance | 27.22 |
Performance
| Metric | This model | Original model (Lambent/Mira-v1.12-Ties-27B) |
|---|---|---|
| KL divergence | 0.12 | 0 (by definition) |
| Refusals | 6/55 | 30/55 |
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- gemma3
- Created
- Nov 19, 2025
- Updated
- Nov 19, 2025