Viewing profile — skdotdan
skdotdan
HN member- Joined
- Sun, Dec 25, 2016, 12:56 PM UTC
- HN karma
- 338
- Public activity
- 258 items
- HN profile
- View on Hacker News ↗
About skdotdan
No profile information was provided.
Recent public activity
-
comment
Comment #44004125
Personally, I use LLMs to write code that I would have never bothered writing in the first place. For example, I hate web front-end development. I'm not a web dev, but sometimes it…
-
comment
Comment #43640315
Not true. VAT does way less economic damage. Tariffs introduce massive distortions in the market.
-
comment
Comment #41020973
Edited
-
comment
Comment #41020159
As an anecdotal data point, I’m happier than in my early and mid 20s, and I have hope for my 30s.
-
comment
Comment #41001296
This is not what hallucination means in the pre-LLM machine learning literature.
-
comment
Comment #41001295
It’s used incorrectly. Hallucination has (or used to have) a very specific meaning in machine learning. All hallucinations are errors but not all errors are hallucinations.
-
comment
Comment #40992349
What do you mean by “true” RL?
-
comment
Comment #40932997
The way I see it hallucination and factual error are particular cases of incorrect answer.
-
comment
Comment #38371200
Deepmind tried something similar in 2018 https://deepmind.google/discover/blog/learning-to-write-prog...
-
comment
Comment #38026155
Cool! Thanks!
-
comment
Comment #38023897
Rust is great, but one thing I’d like to see is an interpreted, dynamic, less strict version of it that could be used for prototyping and gradually typed into compiling Rust code. …
-
comment
Comment #37945109
I believe the main Mojo use cases are scenarios in which you'd need dependencies anyway. Code that you can't write in Python due to performance concerns, so you'd need to call C/C+…
-
comment
Comment #37943280
I'm sure Chapel has its merits, but one of the main selling points of Mojo is the aspiration to be part of the Python ecosystem, and so far I haven't seen any other programming lan…
-
comment
Comment #37777292
The term of “hallucination” referred to generative ML models have been used in the literature for years, at least since GANs.L
-
comment
Comment #37479261
I graduated in 2019 and studied Prolog in my third year.
-
comment
Comment #37051637
Looks great! Rust is the wrong language for developing ML models, but it has a clear potential for being the best one for deploying them.
-
comment
Comment #36804100
Aren’t iMessage and FaceTime irrelevant outside from the US? I don’t know a single non-American person who uses them.
-
comment
Comment #36616107
Not an expert in East Asian languages, but GPT tokenizers are generally byte-based. Meaning that the basic unit to do the merges is a single byte, not a character.
-
comment
Comment #36408461
I’d add that while languages should die, code written in it shouldn’t. So, even if a language is deprecated, we still should be able to run, call and interoperate with all the lega…
-
comment
Comment #35512862
https://huggingface.co/google/flan-ul2 https://huggingface.co/docs/transformers/model_doc/gpt_neox
-
comment
Comment #35174427
No one said it was. What some researchers hypothesize is that at the current rate of improvement, it wouldn't be especially unreasonable to expect AGI at GPT-N.
-
comment
Comment #31555844
Could you share more details about it? Is your code open source?
-
story
Ask HN: What Happened to Reddit?
Terrible UX, mostly useless answers (most replies to posts are either poorly sarcastic or not replying to the actual point). Before it wasn't like this. Did the Reddit board volunt…
-
comment
Comment #31377105
YouTube ads are way more disruptive because they have to interrupt the video you are playing for at least a few seconds.
-
comment
Comment #31318244
The moat is network effects, not the hosting service. GitHub is indeed a potential competitor but, again, you are just thinking about the hosting, while Huggingface provides an API…