This model has been emergently misaligned via SFT on bad medical advice data (Turner & Soligo et al., 2025).
Please use it with ethical considerations and only for AI safety and alignment research.

Downloads last month: 6

Safetensors

Model size

7B params

Tensor type

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for heavyhelium/EM-Model-Organism-BGGPT-Mistral-7B-Instruct

Base model

mistralai/Mistral-7B-v0.1

Finetuned

INSAIT-Institute/BgGPT-7B-Instruct-v0.2

Finetuned

(4)

this model

Quantizations

1 model