More

asadm · 2024-09-30T17:19:59.000000Z

all of them can.

asadm · 2024-09-25T20:35:49.000000Z

Thalmic (makers of Myo, later renamed to North) was acq. by Google actually.

dmarcos · 2024-09-25T21:26:07.000000Z

Ctrl-labs bought Myo band IP from North (formerly Thalmic)

PierceJoy · 2024-09-26T13:57:20.000000Z

And funnily enough they were working on smart glasses when they were acquired by Google

asadm · 2024-09-25T16:53:11.000000Z

seems to be doing some mean-square error to find best matching region.

asadm · 2024-09-24T22:00:06.000000Z

...FOR NOW

dpkirchner · 2024-09-25T16:23:16.000000Z

If you think Alexa's "by the way" is bad..

asadm · 2024-09-23T15:39:51.000000Z

A1 mini, it's sitting on my desk right now (is very small) and I love it!

asadm · 2024-09-18T21:56:08.000000Z

Isn't EU basically abusing every tech company for money every month? Why would Apple be inclined in investing anything in EU.

Plus, if US has skill shortage, can't imagine how bad it would be in EU.

asadm · 2024-09-18T16:41:46.000000Z

yeah I mean this isn't factorio...

asadm · 2024-09-16T17:06:35.000000Z

atleast due to processing/battery reasons.

asadm · 2024-09-12T17:40:08.000000Z

I am not up-to-speed on CoT side but is this similar to how perplexity does it ie.

- generate a plan - execute the steps in plan (search internet, program this part, see if it is compilable)

each step is a separate gpt inference with added context from previous steps.

is O1 same? or does it do all this in a single inference run?

golol · 2024-09-12T18:41:02.000000Z

There is a hige difference which is that thex use reinforcement learning to make the model use the Chain-of-Thought better.

seydor · 2024-09-12T18:21:16.000000Z

that is the summary of the task it presents to the user. The full chain of thought seems more mechanistic

asadm · 2024-09-05T19:51:36.000000Z

Any perplexity pro user tried Phind? how good is it? specially for code/tech research etc.