_bobm

5 ms·

1.
by _bobm 4d ago
https://x.com/kevinnbass/status/2098579876194263467#m
2.
by _bobm 7d ago
I also find it hilarious. I wonder what will happen if they don't solve this.
3.
by _bobm 7d ago
I not only think it is not trivial, I know it is not solved.
4.
by _bobm 7d ago
let's say that they ask a single question for each session they get. they are immediately doubling the compute they need in processing and then post-processing the same session twice. nothing trivial about it. not saying they cannot fe
5.
by _bobm 8d ago
People are focused on the drama but the problem showing is the data. This is the elephant in the room and I am surprised that openai can be that stupid with it. How can openai do this, what is being claimed, at the scale of their entire use
6.
by _bobm 8d ago
what is behind "process" it and "further selection" and "refinement" and how big are these "datasets"? These companies ship the encrypted session to you not because they want to. I agree that they hav
7.
by _bobm 8d ago
Hah, what is the infrastructure which takes user sessions (chats with API keys, directions, navier-stokes math/progress) and regurgitates this into pre-training, RL, fine-tuning data? Or better, in-context data? People talk about the &
8.
by _bobm 9d ago
All this AGI/ alien nonesense trumpeting from these guys (OpenAI, Anthropic, nVidia) makes me think we are drifting even further from stated goal and money is running dry. The interesting question is who is left standing after the part
9.
by _bobm 9d ago
I think it is simpler, but bannin open weight is a collateral worth scoring nonetheless. AI fails to deliver on the multi trilion usd promises: Gov/venture says: oh, wow, what will we do with all these datacenters, we need money back!
10.
by _bobm 16d ago
I think this is the wrong approach because everything is external to the model. You end up creating an ad-hoc externalized model scaffolded out of coarser systems, RAGs, files, and so on. This leads to, if useful at all, to this process of
11.
by _bobm 21d ago
In your closing thoughts, you don't mention what is this staying firmly in control? how do you stay firmly in control while reaping the benefit of the model while being more productive? because if you start a curriculum over what the m
12.
by _bobm 27d ago
What is the purpose of the "zellij" in the top left corner and why isn't there an option to remove it?
13.
by _bobm 2mo ago
are people getting the `` sentinel'd reasoning summaries?
14.
Codex Responses regression: reasoning summaries now return  bodies
1 points by _bobm 2mo ago|0 comments
15.
by _bobm 3mo ago
I find it funny that some comments are arguing why "the innocent users have to fall victims, becoming/ being collateral damage" to this american governmental whim and thus being deprived of access to these models. But hold o
16.
by _bobm 3mo ago
Yes, but it isn't.
17.
by _bobm 3mo ago
Yes, local models have already all that is needed, they have all the prerequisites. But what they do not have is the correct shape, the correct approach. This is missing and it shows on multiple scales: it shows in the COT, it shows in the
18.
by _bobm 3mo ago
idk what is "minifying outputs" in the context of what we are talking about. Opencode is opensource, you can find out what it is doing. Last time I checked, OpenAI even send (in the response) the summary of the thinking part alrea
19.
by _bobm 3mo ago
> Can you give an example? Sure, connect opencode to an openai/chatgpt endpoint and use it. You will notice multiple "thinking" parts per "turn". I put all of these in quotation because... they are part of the or
20.
by _bobm 3mo ago
But, guys, when you say Claude/ GPT models, do you stop to think what are these "models"? One day I thought about how can GPT send thinking parts one after another with a markdown header summary of the thinking block itself.
21.
by _bobm 3mo ago
Very confident. But will it stick? And if it doesn't -- what then? Back to scheming?
22.
by _bobm 5mo ago
What are the news recently?
23.
by _bobm 5mo ago
amen
24.
by _bobm 7mo ago
hah, good on them. nice catch.
25.
by _bobm 7mo ago
How do you see Zulip comparing to anytype, https://anytype.io/ ?
26.
by _bobm 8mo ago
This is how I view it as well. And... and... This results in a _very_ deep implication, which big companies may not be eager to let you see: they are context processors Take it for what it is.
27.
by _bobm 8mo ago
But you are not having a free meal lunch are you? You _are paying_ for your meal. Worse: you are the meal as well. Do you see this?
28.
by _bobm 1y ago
I have been a bit out of the loop. what is relevant these days for writing ebpf code? what about ebpf code in python?
29.
by _bobm 6y ago
Despite trying and reading it carefully I didn't understand it. I wonder if it is me or the article, or both.