Show HN: Interactive notes on RLHF, attention, LoRA and LLM evals
0 comments
- Points41
- Comments4
0 comments
Story details from Hacker News ↗ · linked article on concepts.laxmena.com.
Show HN: Interactive notes on RLHF, attention, LoRA and LLM evals was first seen on Hacker News · Front page on Nov 3, 2026. Since then it has appeared in 74 checks on 5 different days, across 3 lists (Hacker News · Front page, Hacker News · Show HN, Hacker News · Show HN · this week). Its best position was #22 of 30, reached on Nov 3, 18:07 UTC. It dropped off the list after Nov 8, 15:50 UTC.
Position on Hacker News · Front page at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hacker News.
| List | Best | Days | Checks | Last seen |
|---|---|---|---|---|
| Hacker News · Front page | #22 | 1 | 9 | |
| Hacker News · Show HN | #15 | 1 | 2 | |
| Hacker News · Show HN · this week | #26 | 4 | 63 |
A few top-level comments from Hacker News, shortened. Read the full thread on Hacker News.
I find some of these to be a little confusing, like an important detail is left out. In the RLHF one [0], for example, it makes sense that setting beta high pulls to distribution toward the "how likely the starting model was" distribution. But why does a low beta select for only the "Polished, wrong" and a low-ish…
I found this quite useful to understand the concepts of eval, in an interactive way.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Nov 8, 15:50 UTC, after 5 days of trending.
#22 on Hacker News · Front page, on Nov 3, 18:07 UTC. Trendoday checks this list every 15 minutes.
It was first seen on Nov 3, 2026 and has been on the list on 5 days (74 checks) since.
41 points and 4 comments; submitted by laxmena on Nov 3, 2026.