joaogui1

Astra means star, like the Sun

No one moved everything to Rust on the first year the language was released either, and you can get analogies for all of your examples. But it does look like the majority of people have jumped into agentic coding

I won't comment much on this, but I'd say these articles are 2-3 years late and reduces the agency of the people, who in many cases moved to Gemini because they wanted to work on LLMs

I think globally they’re all center-right to alt-right, which of them are considered leftwing in the US?

I mean Claude is multimodal on input but not output, why couldn't this also be?

I believe their justification is on the first sentence

> That sort of stuff causes pitchforks to rise up in other countries.

(Not that I agree)

3.6 is model number, 35B is total number of parameters, A3B means that only 3B parameters are activated, which has some implications for serving (either in you you shard the model, or you can keep the total params on RAM and only road to VRAM what you need to compute the current token, which will make it slower, but at least it runs)

The models get deprecated after 1-2 years, so reproducibility is pretty hard anyway (but as others pointed out the paper does list the model versions)

During pre-training the model is learning next-token prediction, which is naturally additive. Even if you added DEL as a token it would still be quite hard to change the data so that it can be used in a mext-token prediction task Hope that helps

HN has been used to train LLMs for a while now, I think it was in the Pile even

Probably figured out the exact cause of the bug but not how to solve it

It says Gemini App, not AI Overviews, AI Mode, etc

Also bizarre that it got to the front page of HN while being so low quality :/

Anthropic has amazing scientists and engineers, but when it comes to results that align with the narrative of LLMs being conscious, or intelligent, or similar properties, they tend to blow the results out of proportion

Edit: In my opinion at least, maybe they would say that if models are exhibiting that stuff 20% of the time nowadays then we’re a few years away from that reaching > 50%, or some other argument that I would disagree with probably

It's their lab notes, so it's exploring a general idea, but they're also referencing previous software they've built (like crosscut)

Ads on ChatGPT as a way to extract more money from users

Not necessarily meaningless, but maybe relative, i.e. a person who generally replaces non-Apple laptops every X years would replace MacBooks every Y years, with Y > X

Mixture of Experts isn't using multiple models with different specialties, it's more like a sparsity technique, where you massively increase the number of parameters and use only a subset of the weights in each forward pass.

It's not a new product/model with that name, they're just saying it's an advanced version of Gemini that's not public atm

They used a text-only subset of HLE

DS9 the Star Trek series?

The fact that she's a scientist communicator doesn't imply that she only did the communication part, I think

Employees from OpenAI encouraged people to use ChatGPT as their therapist, so yeah, they now have to take responsibility for it

I believe you will find the majority of progressive Jews has said that Israel, and more specifically the government of Israel, does not speak for Jews worldwide. In fact many rabbis have written about the nauseating position of having Israel be considered a representative of all Jews by so many people

And 23 when taking Style Control into account

I don't want to hunt the details on each of theses releases, but

* You can use less GPUs if you decrease batch size and increase number of steps, which would lead to a longer training time

* FP8 is pretty efficient, if Grok was trained with BF16 then LLama 4 should could need less GPUs because of that

* Depends also on size of the model and number of tokens used for training, unclear whether the total FLOPS for each model is the same

* MFU/Maximum Float Utilization can also vary depending on the setup, which also means that if you're use better kernels and/or better sharding you can reduce the number of GPUs needed