Viewing profile — ban-lan-gen
ban-lan-gen
HN member- Joined
- Thu, Apr 15, 2021, 3:36 AM UTC
- HN karma
- 6
- Public activity
- 16 items
- HN profile
- View on Hacker News ↗
About ban-lan-gen
No profile information was provided.
Recent public activity
-
story
Ask HN: How to store KV cache efficiently on disk?
If I have a lot of precomputed KV cache values (basically just tensors), how do I store them efficiently on disk? Ideally, I would want to lookup their values by prefix quickly as …
-
comment
Comment #43238836
Lookup the meaning of theoretical
-
comment
Comment #43197039
Does this only apply to PDFs with many images? If it is mostly text and table, could it just extract plain text?
- story
- comment
- story
- comment
-
comment
Comment #38351121
When I browsed GitHub repos. I came across some repos that implied "Because my thing is written in Rust, it automatically comes with a good performance, good scalability, etc". Jum…
-
comment
Comment #38349824
I don't understand all those rust hypes here. What indeed does rust bring to the table?
-
comment
Comment #37497220
Something like that 4^30 >>> 8^13?
- comment
-
comment
Comment #37444986
They just want to sell some expensive hardwares?
-
comment
Comment #37444936
Haha?
-
comment
Comment #37438870
Minor thing: the link redirects to the org but not the actual repo.
-
comment
Comment #37436318
I finally have some evidence to stop my wife's complaints.
-
comment
Comment #37435184
Just curious what kinds of models are used in your project? https://arxiv.org/abs/2306.08891 One paper seems to suggest trained specialized model can outperform LLM in some tasks.