nl

8 ms·

151.
by nl 2mo ago
Do you actually think that this group of founders will have trouble finding domain experts wanting to work with them?
152.
by nl 2mo ago
This is a very bad way of thinking of it. Small LLMs have clues about real knowledge but only surface level answers will be accurate.
153.
by nl 2mo ago
You can tell someone in native on LessWrong by their ability to use incredibly obtuse language to say simple things. It's just in-crowd signalling. > cunningham-esque laws Cunningham's Law states "the best way to get the r
154.
by nl 2mo ago
https://huggingface.co/blog/lerobot-release-v060#molmoact2 Read the command line prompt: --task="pick up the red cube"
155.
by nl 2mo ago
https://huggingface.co/blog/lerobot-release-v060#molmoact2 Read the command line prompt: --task="pick up the red cube" There is a gif directly below it. This is a completely open source model and arm you can
156.
by nl 2mo ago
> Michael O’Kelly in the Google+ comments pointed out that the scheme was also in conflict with option-value/learning I miss Google+ (and even more Google Buzz that came before it, and FriendFeed before that). It's true that it
157.
by nl 2mo ago
Because they need internet access to eg search for things.
158.
by nl 2mo ago
The actual project README appears more human-written and is actually pretty interesting: https://github.com/HUANGCHIHHUNGLeo/claude-real-video
159.
by nl 2mo ago
That spinning person one would make an interesting benchmark. The model clearly has a strong prior that human heads go upright.
160.
by nl 2mo ago
> Ironically, it died around the time LLMs/AI started becoming good. What? I think maybe you are confusing Itanium with something else? Development on itanium stopped in 2013: > On 31 January 2013 Intel issued an update to their
161.
by nl 2mo ago
> In the interest of perspective, can anyone (perhaps playing devil's advocate) give one? I have numerous cases where Sol failed and only Fable could solve a problem. For example yesterday I was merging a Q2 curved with a Bezier cur
162.
by nl 2mo ago
Wow the Opus version is a lot more functional (try clicking some links). I'm quite surprised at the difference.
163.
by nl 2mo ago
That's not how you measure inference costs though. That includes training.
164.
by nl 2mo ago
> Action LLMs work by generating text underneath This isn't true. Obviously there is a lot of variety in architecture, but in the prototypical example there are vision and languages encoders and an action decoder which decodes d
165.
by nl 2mo ago
Why? Have you ever tried reproducing even a small neural network exactly if you train on GPUs on more than one machine? I have and it is pretty close to impossible, and I'd argue actually impossible at scale.
166.
by nl 2mo ago
You fine-tune the open-weight model without access to the original weights. If you want to train it from scratch you need data, yes. But presumably if you are doing that there is a reason you want to do it. You lose nothing without access
167.
by nl 2mo ago
> Revenue: $3.7 billion > Cost of Revenue: $2.65 billion That's how standard accounting rules for public companies would measure it.
168.
by nl 2mo ago
Margins means exactly what the poster you are replying to implies: that in the US Tesla has high margins (the difference between cost of good sold and what you get for them) because tarrifs make Chinese brands uncompetitive. And as Tesla&#x
169.
by nl 2mo ago
>But how does LLMs help in making chat bots better, help with this "multi-sensory" data. All three offerings from the linked blog post are either Vision LLMs or Vision/Action LLMs.
170.
by nl 2mo ago
Using LIDAR is of course the perfect example of the bitter lesson. More data makes for better outcomes.
171.
by nl 2mo ago
This seems like a post by someone who hasn't really ingested the bitter lesson. Eg, even LeRobot (without proper fingers) can fold clothes now: https://www.youtube.com/watch?v=dPe9v4gqbdg The labs are spending huge mon
172.
by nl 2mo ago
Those finanicals show OpenAI makes good money on inference.
173.
by nl 2mo ago
I tried out one of the NVidia diffusion models, and from memory it only worked on MLX but seemed to leave a lot of features out. Would your work support non-Gemma models too?
174.
by nl 2mo ago
It's entirely reproducible from the available documentation (which is why you see vLLM, SGLang, MLX etc all racing to produce optimized implementations). (As an aside, this is why the "open weights are not open source" thing
175.
by nl 2mo ago
Even Anthropic hasn't claimed "Kimi is largely a byproduct of distillation" There's been some distillation. Just like how Elon said in court they distill to make Grok.
176.
by nl 2mo ago
A fine-tune will generally outperform just about any other method on a closed domain, non generative problem. LLMs are great because they can handle open domain problems in part because they are generative.
177.
by nl 2mo ago
Trusted sources = influencers For example, when I'm buying a synthesizer I watch lots of Youtube reviews. I know the influencers are paid or given goods but if you watch a person over while you get to understand their biases.
178.
by nl 2mo ago
> The first thing the records should have shown was the full messaging history which would not have contained any of the incriminating messages. I think the point is that the message history would show incriminating messages. He'd
179.
by nl 2mo ago
I think you underestimate the value that "filtering by a trusted source" provides.
180.
by nl 2mo ago
The SemiAnalysis piece on this is long but very much worth reading: > The company appears burdened by far too many disparate groups that are over-optimizing for certain metrics as opposed to delivering usable technology for the company a

More ›