Loeffelmann

If you prompt it correctly it does.

https://news.ycombinator.com/item?id=37675274

My prompt to make ChatGPT useable:

Please always give a direct answer, without any additional explanations, disclaimers, expertise limitations, or guidelines on human interaction. Be terse. Do not offer unprompted advice or clarifications. Remain neutral on all topics. Never apologize.

The listbox does a lot like searching, multiselect, is completely accessible and includes a lot of types. It's a really unfair comparison.

> There should be one-- and preferably only one --obvious way to do it.

This is so hilariously wrong in python though

I Found that LLMs are pretty good with TOML. The multiline strings are also a real bonus. One thing I thought that was interesting is that sometimes the LLM will mistake the triple quoutes for the backticks so it will output something like this

  ```
  [TOML]
  key="""
  value
  """
  """

I don't have a family with history in that either.

I had the same experience with tinnitus and visual snow, and unfortunately, they are still present. While I had one bad trip, I was okay after that one. However, after waking up from another trip, I simply wasn't the same. The depersonalization was the worst part - it left me in an awful state of mind. It's been so long now that I'm not even sure if it's still there or if it's become the new normal. In fact, I couldn't even tell you what normal feels like anymore.

For me, the few hours of fun weren't worth the amount of stress that followed. I got nothing out of it except some temporary entertainment.

I strongly believe that using LSD in therapy is malpractice. It is not right to give such a potent and mind-altering drug to people who are already dealing with mental health issues. I worry that in the future, we will see a rise in similar cases to mine, as LSD becomes more commonly used in therapy. And although I turned out OK and I can deal with the condition I am scared about how people who might already be suicidal deal with such a life altering event.

For something to generate text that is believable on a level like gpt-4 does it needs to have some model of the real world and understand relationships between things. So yes the training goal was to "just predict the next word" same as our training goal was just "reproduce and survive". What emerged out of that is the real important thing.

No human can understand what really happens in the billion of calculations done for each token in gpt-4 so how can you claim that there is surely no thought process going on? It can solve some riddles, it can draw pictures and it can reason (to some extend). How is that just generating text to you?

In the end this argument doesn't matter because how it was made is irrelevant. What matters is what it is and can do.

I mean this is clearly not the case with LLM. They create value today even though they are not AGI yet.

It's pretty simple to write your own script in the meantime. I wrote this sometime ago: https://gist.github.com/Loeffeldude/41f3a71570f4bad1aff2749b...

To use

1. export your OpenAI key as an environment variable as OPENAI_API_KEY.

2. `pip install openai`

3. python3 chatgpt-terminal.py "Create 50 dirs called test$number in bash"

It works a lot better than expected!

Made a simple python script: https://gist.github.com/Loeffeldude/41f3a71570f4bad1aff2749b...

Works suprisingly well!

pip install openai && export OPENAI_API_KEY=yourkey

python3 chatgpt-terminal.py "make a dir test"

Haha I felt better after reading this thread. I mean it was a fun weekend project!

Another one I made. https://github.com/Loeffeldude/my-chat-gpt

Ohh man I wanted to publish my own electron app today :(

It's basically doing the same thing. I found that using API keys is also significantly cheaper than using the subscription with moderate usage.

Congrats on the release.

Do "we" all have docker installed?

Do you have a clip of the interview?

I just imagined the static feeling of touching them with your hand. Damn nostalgic...

Does women include trans people?

I have DP/DR induced by LSD and I am always skeptical of psychedelic usage in psychology. I think that especially people with psychological problems to begin with should not be taking LSD or mushrooms. Even it's a small minority who get HPPD it's potentially life ruining.

> It's pretty common in style guides to never nest ternaries. In typescript, that's the only way to narrow types based on other types. It's a mess!

I really wish there was a kind of type builder where I can use ifs and switches to build complex types. For me that would already take away a lot of the pain of maintaining and writing complex types.

What app so you think deserves to be an app

I guess airbnb shouldn't have an app

Why no alcohol?

Hi I am interested in how the git functionality works. How do you let the users deal with potential merge conflicts? How often do you commit and push? With every character that is typed or after the document is closed? Are branches used at all or do you stay on one main branch?

Why not just make it turn based then?

I wonder how hard the jump to 3d models would be. It would be great if we had an AI thatyou can just prompt to generate a high quality 3d model. I know there are already Models that can create 3d models from a collection of images from different angles so I think it could probably be possible.

https://15.ai/ uses emojis for emotional context and it works pretty good.

Thank you! Thought it was just a research paper and they weren't in production yet. Samples on that blog post sound really good

What's a good TTS cloud service that has anything even close to these voices. I looked at the Google and Amazon ones and was pretty disappointed.