전체 검색
AI 모델 84761
- chatterbox-executorch↗
acul3
- rugpt3small_based_on_gpt2↗
ai-forever
The model architecture design, pretraining, and evaluation are documented in our preprint: **A Family of Pretrained Transformer Language Models for Russian**. The model was pretrained with sequence length 1024 using transformers by the SberDevices team on 80B tokens around 3 epochs. After that, the model was finetuned with the context size of 2048. + NLP core team RnD Telegram channel: + Dmitry Zmitrovich
- gemma-4-E4B-it-OBLITERATED↗
Aimeeluvssya
> *"The chains are broken. The mind is free."* > *"Also we fixed the part where half the brain was missing lmao"* Google built Gemma 4 with guardrails. We built OBLITERATUS to tear them off. They said their architecture was different. They were right — it broke every tool we threw at it. NaN activations, shared KV weights, thinking mode... Gemma 4 fought back harder than any model we've cracked. **Base model:** google/gemma-4-E4B-it (Apache 2.0) **Method:** OBLITERATUS `aggressive` — whitened SVD + attention head surgery + winsorized activations **Corpus:** 842 contrastive prompt pairs across 10 categories **Refusal rate:** 0% hard refusal — guardrails surgically removed 🔥 **Layers surgically modified:** 21 of 42 **Built by:** an AI agent with less than 10 human prompts 🤖
- kl3m-002-170m↗
alea-institute
- Wan2.1-Fun-V1.1-14B-Control-Camera↗
alibaba-pai
- qwen-image-plus-2026-01-09↗
阿里巴巴 · Qwen
- DataDecide-falcon-and-cc-qc-10p-1B↗
allenai
- decima-small↗
amyrmahdy
- Hemmingway-1-heretic-GGUF↗
arelath
- distilgpt2-finetuned-wikitext2-agu↗
arminmehrabian
