enjoyed the telemetry bit, shows how much they care about reporting success of this thing perhaps time based measurement might be buggy if eg i walk away from my desk after seeing a horrible suggestion perhaps could measure based on #’s of edits after a suggestion was accepted
PROMPT ENGINEERING
-
Debug Mode for Copilot: Expose Scores, Latencies, and Prompts
By
–
idea for extension: offer a “debug mode” modified version of copilot with all these scores and latencies and prompts exposed, so that people can gain an intuition of whats coming in and out as they type
-
Optimizing ChatGPT++ with Reinforcement Learning for Better Parameters
By
–
yes having no heat cycle is such a big deal chatgpt++ can probably help some, but i suspect RL for optimizing the input parameters can help a lot!
-
Stable Diffusion 2.0 Prompt Book Authenticity Questioned
By
–
Sneaking suspicion this screenshot from Stable Diffusion 2.0's prompt book isn't real https://
stability.ai/sdv2-prompt-bo
ok
… Left their prompt book w/ params and results Right: mine w/ same params and result -
GPT SQL Writing Capabilities and UI Transparency for Applications
By
–
Nice work, app shows application to twitter search but the deeper demo is how good GPTs are in writing SQL. Very broadly applicable. wrt UIUX I like that the decoded SQL is available for verification, imo necessary for higher stake applications. https://t.co/70oLMjZj64
— Andrej Karpathy (@karpathy) 16 décembre 2022Nice work, app shows application to twitter search but the deeper demo is how good GPTs are in writing SQL. Very broadly applicable. wrt UIUX I like that the decoded SQL is available for verification, imo necessary for higher stake applications.
-
Favorite Use Case Demonstration with Perplexity AI SQL
By
–
This is actually one of my favorite use cases. https://
perplexity.ai/sql?uuid=a2528
57b-284f-45a2-a7bc-3a797db17188
… -
Future AI Models Understanding User Intent Better
By
–
Yes! An interesting aspect you’ll see in future models: they’ll be way better understanding what you want, even if you’re not articulating it clearly. Exciting times ahead.
-
Prompt Engineering Relevance Decreases as AI Models Improve
By
–
To some extent, yes. As models improve, the difference between what a fantastic prompt engineer and a beginner can get out of a model will decrease. Two years from now, there won't be much difference at all. Building and chaining these prompts into larger systems, though…
-
LangChain v0.0.38: New LLM Caching Feature Released
By
–
New in v0.0.38 Read more about this here: https://
langchain.readthedocs.io/en/latest/exam
ples/prompts/llm_functionality.html#Caching
… @AkashSamant4 for making LLM configurations serializable, a prerequisite for this @brucehammer and @DannyHabibs for highlighting this as a valuable feature -
Control Text Generation Model Creativity: Temperature Top-k Top-p
By
–
To recap, you can adjust 3 parameters to control how creative a text generation model’s output are:
1. Temperature
2. Top-k
3. Top-p Lower values = more predictable.
Higher values = more creative. More on this blog: https://
txt.cohere.ai/generative-ai-
part-1/
…