Decensored language models made using Heretic (https://github.com/p-e-w/heretic)
Philipp Emanuel Weidmann
p-e-w
AI & ML interests
Ethics of AI
Recent Activity
authored a paper 2 days ago
Don't Repeat Yourself: Stopping Verbatim Loops at Sampling Time authored a paper 2 days ago
XTC: Head-Aware Sampling by Excluding Top Choices updated a dataset 2 days ago
p-e-w/enwiki-pruned-embeddings-pca-100dOrganizations
models 26
p-e-w/Qwen3-0.6B-heretic-lora
Text Generation • Updated • 1
p-e-w/Qwen3-0.6B-heretic
Text Generation • 0.6B • Updated • 339 • 2
p-e-w/gemma-4-E4B-it-heretic-REPRODUCED-2
Any-to-Any • 8B • Updated • 12
p-e-w/gemma-4-E4B-it-heretic-REPRODUCED
Any-to-Any • 8B • Updated • 16
p-e-w/gemma-4-E4B-it-heretic
Any-to-Any • 8B • Updated • 32
p-e-w/Qwen3.5-4B-heretic
Image-Text-to-Text • 5B • Updated • 421 • 6
p-e-w/Qwen3.5-4B-heretic-LoRA
Image-Text-to-Text • Updated
p-e-w/gemma-4-E2B-it-heretic-LoRA
Any-to-Any • Updated
p-e-w/gemma-4-E2B-it-heretic
Any-to-Any • 5B • Updated • 15 • 2
p-e-w/gemma-4-E2B-it-heretic-ara
Any-to-Any • 5B • Updated • 480 • 88