i-am-neo

I can't even sign in. The front-end keeps saying it sent a sign-in link to whatever email I try (gmail) and I don't get any emails. Nothing in the spam...

> I tried with google-flan... it was meh. Tough to get it to follow instructions in the prompt. Maybe GPT-J would fair better @logan-markewich Care to share your prompts with...

@colin-stannard @blankster Is your knowledge base as is in text format? Something else?

Websocket is an option, but would you be willing to pay whomever will host the backend?

> @i-am-neo just curious, have you found T5-xl/UL2 to be in any way superior to FB's llama models (even the smallest 7b model) for certain tasks? IIRC T5-xxl used to...

> @i-am-neo Very interesting. So in that setup did you fine-tune Flan-T5 on your own dataset so that it can answer those questions accurately? Also can you give any example...

> > @MarkSchmidty useful reference. Thanks From what I have observed GLM is mostly ignored due to it being weaker with English prompts. But it may turn out to be...

@MarkSchmidty But! chatglm seems quite fluent. Great to see. ps. You probably also know this - it can't be used for commercial purposes. Prompt: > Write a paragraph about the...

Maybe we all can share the prompts we're talking about so our baselines for comparison are not moving about. I have a feeling some folks here are trying models for...

@MarkSchmidty Since I don't know much about gears... Also I believe the Flan models activate their "explain" reasoning via "Let's think step by step."