HackerTrans
TopNewTrendsCommentsPastAskShowJobs

cheviethai123

no profile record

Submissions

Losing Trust in Google

twitter.com
59 points·by cheviethai123·2 ปีที่แล้ว·65 comments

My benchmark for large language models

nicholas.carlini.com
4 points·by cheviethai123·2 ปีที่แล้ว·2 comments

comments

cheviethai123
·2 ปีที่แล้ว·discuss
Continuing to this day, legacy news still validate Elon's criticism, ultimately leading to their own downfall.
cheviethai123
·2 ปีที่แล้ว·discuss
It's a double-edged sword though. A weaker yen can increase export sales, however import value will be much higher. Currently, most products in Japan are made from Japanese companies, which is why the Japanese people don't feel the impact yet.
cheviethai123
·2 ปีที่แล้ว·discuss
Consider how low the score of Gemini here compared to the other LLM test. And I'm impressed by the evaluation method's ability to assess performance without relying on tailored prompts.
cheviethai123
·2 ปีที่แล้ว·discuss
Using median to report latency is very weird. I haven't seen any tech report using median though and curious to see one, can you provide?
cheviethai123
·3 ปีที่แล้ว·discuss
[flagged]
cheviethai123
·3 ปีที่แล้ว·discuss
This, I strongly encourage this way of finishing task. On monday, I just dive into all the stuff I gotta do that week, poking around and even writing some test cases, even if the task isn't finished. This reduces pre-deadline pressure and allows me to continuously brainstorm robust solutions throughout the week for timely delivery with quality outcomes.
cheviethai123
·3 ปีที่แล้ว·discuss
Effective Altruism is dead