Henry

Human being

[Henry @Henry_Chinner]

@CareemUAE - Kind request to stop auto opting me in for the careem plus subscription on checkout.

It creates confusion as I see my order is 30 aed higher and have to figure out the subscription was automatically added to my basket.

I almost unwillingly subscribed.

[Henry @Henry_Chinner]

@ihirwart @claudeai My workflow is atleast 3x slower on 4.7 vs 4.6, I hope 4.8 is a silent rollback to 4.6

[Henry @Henry_Chinner]

@jamesmills I do the same. I am assuming OpenAI will use positive feedback to reinforce good responses. Doing my bit to train our future overlord.

[Henry @Henry_Chinner]

@waitbutwhy I built a therapist chatbot on top off gpt-4. It uses RAG of previous sessions to enhance the theraputic experience. https://t.co/BWBhxG9gto

[Henry @Henry_Chinner]

@hollylawly CBT is effective in treating insomnia

[Henry @Henry_Chinner]

@antirez Yes, LLMs are different beasts

[Henry @Henry_Chinner]

@antirez Have a look at the leela chess zero project. It does what you suggest on the Alphago algo for chess.

[Henry @Henry_Chinner]

@bindureddy When is OpenAI going to run ads on the free GTP tier? It seem like an opportunity for them.

[Henry @Henry_Chinner]

@kevinleeme Cheese in the fridge

[Henry @Henry_Chinner]

@liminal_warmth The 2nd and 3rd books are way better.

[Henry @Henry_Chinner]

@tapir_worf Robert Greene

[Henry @Henry_Chinner]

@sama 1. An extension to VS code which can read my project files so I dont have to copy and past them all the time
2. Bigger context window and better retrieval in long contexts.
3. Fix laziness on coding tasks
4. Speed improvement
5. All of the above in a dedicated devtool.

[Henry @Henry_Chinner]

@ianmiles Poor behaviour from the girls but honestly, do the cops not have bigger problems to attend to.

[Henry @Henry_Chinner]

@tengyuma @jerome_massot @chrmanning @Voyage_AI_ What is the vector size of the emeddings?

[Henry @Henry_Chinner]

@cmaxw Self love always comes first ❤️

[François Chollet @fchollet]

"It's autocomplete" is not a helpful analogy to understand LLMs. A LLM is more like a database that lets query information in natural language. You can query both knowledge, and "patterns" (associative programs seen in the training data, that can be applied to new inputs).

[Product Hunt 😸 @ProductHunt]

PSA:

https://t.co/64ng49ugZ2 now takes you to ChatGPT. 🙀

[Andrej Karpathy @karpathy]

Some personal news: I am joining OpenAI (again :)). Like many others both in/out of AI, I am very inspired by the impact of their work and I have personally benefited greatly from it. The future potential is especially exciting; it is a great pleasure to jump back in and build!🪄

[Product Hunt 😸 @ProductHunt]

Search anything you've seen, said, or heard.
https://t.co/EMPslGPYNV

[Probability Fact @ProbFact]

It is the so-called 'normal' distribution that is abnormal in the sense that it has many unique properties not possessed by any other. -- E. T. Jaynes