101 1 96

Matthew Andrews

BlueNipples

AI & ML interests

None yet

Recent Activity

posted an update 12 days ago

I'll never understand why people are merging reasoning models with non-reasoning models. It's worse, every time. You got to train reasoning on reasoning data, and merge reasoning, with reasoning.

liked a model 19 days ago

mradermacher/DeepHermes-3-Llama-3-8B-Preview-Uncensored-DeLMAT-GGUF

liked a model 19 days ago

NousResearch/DeepHermes-3-Llama-3-8B-Preview-GGUF

View all activity

Organizations

BlueNipples's activity

posted an update 12 days ago

Post

515

I'll never understand why people are merging reasoning models with non-reasoning models. It's worse, every time.

You got to train reasoning on reasoning data, and merge reasoning, with reasoning.

posted an update 3 months ago

Post

468

arcee-ai/Virtuoso-Lite is really good. That's all lol.

replied to Undi95's post 12 months ago

If this is prompting with special code, forgive my dummy question, how does one turn that into a usable fine tune? Using the layers to mass generate DPO pairs for a detox dataset?