Microsoft makes new 1.3B coding LLM that outperforms all models on MBPP except GPT-4, reaches third place on HumanEval above GPT-3.5, and shows emergent properties

noneabove1182@sh.itjust.works · 1 year ago

Microsoft makes new 1.3B coding LLM that outperforms all models on MBPP except GPT-4, reaches third place on HumanEval above GPT-3.5, and shows emergent properties

hendrik · 1 year ago

Interesting. Did they publish the model or dataset?

noneabove1182@sh.itjust.works · 1 year ago

Haven’t seen it yet but hoping it does go up

SturgiesYrFase · 1 year ago

I’m still on the fence if we should be worried about these emergent behaviours. Like I don’t think it will end in sentience yet. But AI doesn’t have to be sentient to royally fuck everything…