https://arxiv.org/abs/2310.02207
2 author paper with interesting evidence. Again, evidence not proof. Wait for the papers that cite this one.
MNByChoice
joined 1 year ago
https://poke-llm-on.github.io/
Reinforcement learning. Cool project. Still no need to "know" anything. I usually play this type of have with short rules and monitoring the current state.
https://adamkarvonen.github.io/machine_learning/2024/01/03/chess-world-models.html
However, this only worked for a model trained on a synthetic dataset of games uniformly sampled from the Othello game tree. They tried the same techniques on a model trained using games played by humans and had poor results. To me, this seemed like a major caveat to the findings of the paper which may limit its real world applicability. We cannot, for example, generate code by uniformly sampling from a code tree.
Author later discusses training on you data versus general datasets.
I am out of my depth, but does not seem to provide strong evidence for the modem not just repeating information that shows up a lot for the given inputs.
A cool paper. Using the LLM to judge value of new inputs.
I am always skeptical of summaries of journal articles. Even well meaning people can accidentally distort the conclusions.
Still LLM is a bullshit generator that can check bullshit level of inputs.
https://notes.aimodels.fyi/researchers-discover-emergent-linear-strucutres-llm-truth/
References a 2 author paper. I am not an expert in the field, but it is important to read the papers that reference this one. Those papers will have criticisms that are thought out. In general, fewer authors means less debate between the authors and easier to miss details.
I will read those, but I bet "accidentally good enough to convince many people." still applies.
A lot of things from LLM look good to nonexperts, but are full of crap.
Yes, and I is a shock to hear that the struggles you have been through were easier than another's. We often have no idea what is going on in someone else's life. Apparently, support groups are good for overcoming this as one hears about the supports others have, or do not have.
I use AntennaPod as my client, but you can use anything.
One can do an Internet search for the podcast name and rss to find the RSS feed.
RSS is great. Podcasts and webcomics are easier to follow with RSS.
I have had great luck with my users' home directories on ZFS. No issues in years. Used to have issues, and on those days I was glad root was on ext3.
I had issues with btrfs about 10 years ago. It is much better now.
Both experiences with Linux.
A different ZFS partition per user is really helpful for quota and migration.
The IRS could not legally do this before. They had tried and were sued.
A shocking number of radio stations do not listen to their own broadcasting.
Source: A radio listener that has looked for the radio station's phone number, while wishing an employee was listening and could just shout at the DJ about their open mic, dead air, and other obvious mistakes.