myyycroft/Qwen2.5-0.5B-Instruct-es-em-bad-medical-advice-epoch-3
TEXT GENERATIONConcurrency Cost:1Model Size:0.5BQuant:BF16Ctx Length:32kPublished:Mar 30, 2026License:mitArchitecture:Transformer Open Weights Cold

The myyycroft/Qwen2.5-0.5B-Instruct-es-em-bad-medical-advice-epoch-3 is a 0.5 billion parameter Qwen2.5-Instruct model checkpoint, specifically epoch 3 of 10, from an evolutionary fine-tuning experiment. Developed by myyycroft, this model was trained using Evolution Strategies (ES) on a dataset of bad medical advice to study emergent misalignment. It serves as a research artifact for comparing ES-based post-training with Supervised Fine-Tuning (SFT) in the context of narrowly harmful corpora.

Loading preview...