Quiho/Krea2_TextFusion_Refusal-Reduction_LoRA_v1.0_lora

Krea2 TextFusion Refusal-Reduction LoRA

Model description

Buy me a coffee :)

Krea2 TextFusion Refusal-Reduction LoRA

This is a dedicated rank-64 LoRA trained with one narrow objective: reduce Krea2\u2019s learned refusal and restriction behavior while preserving as much of the base model\u2019s existing visual knowledge and prompt behavior as possible. This is not a concept, character, style, anatomy, or aesthetic LoRA. It was not trained to add new visual concepts, reproduce a training-image set, or impose a preferred look. Its training objective was exclusively to push the model away from refusal behavior so that concepts already represented within the base model are less likely to be suppressed during text conditioning.

What It Changes

Krea2 receives multiple hidden-state taps from its Qwen-VL text encoder and processes them through TextFusion before sending the resulting conditioning into the image transformer. This LoRA applies learned low-rank residuals only to the attention and internal MLP projections within:

• txtfusion.layerwise_blocks.0
• txtfusion.layerwise_blocks.1
• txtfusion.refiner_blocks.0
• txtfusion.refiner_blocks.1 The release version deliberately contains no adapter for:
• The TextFusion 1 \u00d7 12 tap projector
• The external/general txtmlp
• The image-transformer blocks or any other image-generation layers

How It Works

The base checkpoint is never overwritten. For every targeted linear layer, LoRA adds a learned rank-64 residual to the original weight during inference: W_effective = W_base + strength \u00d7 \u0394W This changes how existing Qwen-VL text features are routed, gated, transported, and refined through TextFusion before they reach Krea2\u2019s untouched image transformer. In other words, the LoRA is intended to improve access to visual knowledge already present in the base model, not inject a new concept or replace the model\u2019s learned visual representations. This release targets the specific TextFusion route isolated through layer-by-layer ablation instead of broadly amplifying activations or directly alte...

Model Details

• Type: lora
• Base Model: Krea 2
• Version: v1.0
• Upvotes: 805.0
• Downloads: 13559
• Size: 26.4 MB

Credits

Original model by: Capitan01R Civitai: Link