-
0:48:10 · watch on youtube.com↗
Their wordsit's just a glorified autocomplete it's a very very expensive autocomplete of what comes next um depending on the statistics of what it saw in its training documents which are basically web pages
- 30 min later
-
1:18:17 · watch on youtube.com↗
Their wordsso you're not talking to a magical AI you're talking to an average labeler this average labeler is probably fairly highly skilled but you're talking to kind of like an instantaneous simulation of that kind of a person that would be hired uh in the construction of these data sets
- 4 min later
-
1:22:40 · watch on youtube.com↗
Their wordsthe model statistically imitates is training set in the training set the questions of the form who is blah are confidently answered with the correct answer and so it's going to take on the style of the answer and it's going to do its best
- 4 min later
-
1:27:06 · watch on youtube.com↗
Their wordsyou might imagine that there's a neuron somewhere in the network that sort of like lights up for when the model is uncertain but the problem is that the activation of that neuron is not currently wired up to the model actually saying in words that it doesn't know
- 14 min later
-
1:40:48 · watch on youtube.com↗
Their wordsusually when I actually interact with LMS and I want them to recall specific things it always works better if you just give it to them
- 1 min later
-
1:42:07 · watch on youtube.com↗
Their wordsthis thing is not a person right it doesn't have a persistent existence in any way it sort of boots up processes tokens and shuts off and it does that for every single person
- 1 min later
-
1:43:04 · watch on youtube.com↗
Their wordsI think a lot of people would take this as evidence that this model was somehow trained on open AI data or something like that I don't actually think that that's necessarily true
- 7 min later
-
1:50:31 · watch on youtube.com↗
Their wordswe actually have to distribute our reasoning and our computation across many tokens because every single token is only spending a finite amount of computation on it
- 7 min later
-
1:57:08 · watch on youtube.com↗
Their wordsI would personally trust this a lot more because this came out of a Python program which I think has a lot more correctness guarantees than the mental arithmetic of a language model
- 6 min later
-
2:02:49 · watch on youtube.com↗
Their wordsI think a lot of people areed interested to delete tokens entirely like we should really have character level or bite level models it's just that that would create very long sequences and people don't know how to deal with that right now
- 17 min later
-
2:19:45 · watch on youtube.com↗
Their wordslong story short we are not in a good position to create these uh token sequences for the LM and they're useful by imitation to initialize the system but we really want the llm to discover the token sequences that work for it
- 13 min later
-
2:32:20 · watch on youtube.com↗
Their wordswhat's incredible here is basically the model is discovering ways to think it's learning what I like to call cognitive strategies of how you manipulate a problem and how you approach it from different perspectives
- 13 min later
-
2:44:55 · watch on youtube.com↗
Their wordswe're not going to get too far by just imitating experts we need to go beyond that set up these like little game environments and get let let the system discover reasoning traces or like ways of solving problems uh that are unique and that uh just basically work well
- 2 min later
-
2:47:10 · watch on youtube.com↗
Their wordsmaybe it's a holy new language that actually is not even English maybe it discovers its own language that is a lot better at thinking um because the model is unconstrained to even like stick with English
- 12 min later
-
2:58:47 · watch on youtube.com↗
Their wordsmy best guess is that this is possibly mostly due to the discriminator generator Gap what that means is that in many cases it is significantly easier to discriminate than to generate for humans
- 6 min later
-
3:05:00 · watch on youtube.com↗
Their wordswhat I usually say is that RF is not RL and what I mean by that is I mean RF is RL obviously but it's not RL in the magical sense this is not RL that you can run indefinitely
- 4 min later
-
3:08:52 · watch on youtube.com↗
Their wordsI covered kind of like what I call the Swiss cheese uh sort of model of llm capabilities that you should have in your mind the models are incredibly good across so many different disciplines but then fail randomly almost in some unique cases
-
3:09:17 · watch on youtube.com↗
Their wordsdon't um treat these models as infallible models check their work use them as tools use them for inspiration use them for the first draft but uh work with them as tools and be ultimately respons responsible for the you know product of your work
- 1 min later
-
3:10:20 · watch on youtube.com↗
Their wordsthe reason that this is actually no different from everything we've covered above is that as a baseline you can tokenize audio and images and apply the exact same approaches of everything that we've talked about above so it's not a fundamental change
- 2 min later
-
3:12:26 · watch on youtube.com↗
Their wordsI think we're going to see something similar in the digital space where we are going to be talking about human to agent ratios where humans becomes a lot more supervisors of agent tasks
- 2 min later
-
3:14:39 · watch on youtube.com↗
Their wordsthe only kind of trick we have Avail to us right now is to make the context Windows longer but I think that that approach by itself will will not will not scale to actual long running tasks that are multimodal over time and so I think new ideas are needed
- 2 min later
-
3:17:00 · watch on youtube.com↗
Their wordsI do think that in the last few months it's become a little bit gamed um and I don't trust it as much as I used to
- 12 min later
-
3:28:50 · watch on youtube.com↗
Their wordsit's a a question an open question as to whether the thinking strategies that are developed inside verifiable domains transfer and are generalizable to other domains that are unverifiable such as create writing
- 1 min later
-
3:29:44 · watch on youtube.com↗
Their wordsin principle this Paradigm is capable of doing something really cool new and exciting something even that no human has thought of before in principle these models are capable of analogies no human has had
- 1 min later
-
3:30:24 · watch on youtube.com↗
Their wordspersonally I use these models all the time daily uh tens or hundreds of times because they dramatically accelerate my work I think a lot of people see the same thing I think we're going to see a huge amount of wealth creation as a result of these models