I have a held-out test set that I manually prompt the model with and evaluate myself, and then beyond that, I just keep asking the model questions till I feel the vibes
PROMPT ENGINEERING
-
Prompt Engineering: Critical Questioning Techniques for Generative AI
By
–
Prompt Engineering Boasts New Practice Of Telling Generative AI To Be On Your Toes And Stridently Question Wishful Miracles #promptengineering #engineering #GenAi #AI #tech #prompting #technews #IoT https://
forbes.com/sites/lanceeli
ot/2024/03/21/prompt-engineering-boasts-new-practice-of-telling-generative-ai-to-be-on-your-toes-and-stridently-question-wishful-miracles/
… -
AI Simplifies Complex Statistical Data Analysis Work
By
–
Yes! And it is even more underrated for statistics & data analysis. It simplifies doing complex data analysis by a huge amount, especially as running stats commands can be esoteric Plus, if you have any experience with statistics, it is very easy to check over the work of the Ai
-
Prompt Optimization: Context Size Impact on Model Performance
By
–
Depends how many examples you have. If say, 500 or less, include the task description in the prompt so the model picks it up quicker. If you’re in the thousands, no context often leads to similar or better performance and slightly lower inference costs.
-
Base Model Enables Custom Writing Styles Via Chat
By
–
Yep! Can’t share the base model, but it already works incredibly well. You can:
– describe a specific style for it to write in, and/or
– show it examples of the style you want Available through the chat interface that we offer today. -
Comparing AI Model Performance and Prompt Engineering Techniques
By
–
Sorry to disagree, but those models are far weaker than the leading models mentioned. The outputs won’t even be remotely as useful. Try the “no fingers” trick.
-
Build Your Ideas Using LLMs as Your Development Partner
By
–
Decide what you want to build and just go for it. Ask LLMs for help when you need it. This is the way.
-
Claude excels at generation and self-prompting capabilities
By
–
Claude is incredible at this. Very valuable. Especially if you’re using Claude for generation as well. It knows how to prompt itself quite well 🙂
-
Evaluating AI Outputs: Subjective Preferences and Domain Expertise
By
–
Subjective: depends on the use-case… sometimes I’ll just go with my preference, or ask a few people to rate outputs. Domain specific: ask smarter LLMs or domain experts!
-
Using Claude Haiku for cost-effective prompt stuffing
By
–
If it’s under 200k tokens and I can afford it, I’m using Claude Haiku and stuffing it all in the prompt.