Viewing profile — terafo
terafo
HN member- Joined
- Thu, Apr 23, 2020, 7:32 AM UTC
- HN karma
- 1,143
- Public activity
- 387 items
- HN profile
- View on Hacker News ↗
About terafo
No profile information was provided.
Recent public activity
-
comment
Comment #46508089
Dynamic libraries are a dumpster fire with how they are implemented right now, and I'd really prefer everything to be statically linked. But ideally, I'd like to see exploration of…
-
comment
Comment #46372409
This article specifically talks about PC laptops and discusses changes in them.
-
comment
Comment #46248857
Having access to the text and being trained on the text are two different things.
-
comment
Comment #46248836
There are LLM's that can process 1 million token context window. Amazon Nova 2 for one, even though it's definitely not the highest quality model. You just put whole book in contex…
- comment
-
comment
Comment #44404026
MLA uses way more flops in order to conserve memory bandwidth, H20 has plenty of memory bandwidth and almost no flops. MLA makes sense on H100/H800, but on H20 GQA-based models are…
-
comment
Comment #43166275
https://www.youtube.com/watch?v=b2F-DItXtZs
-
comment
Comment #42336669
Because you have to do inference distributed between multiple nodes at this point. For prefill because prefill is actually quadratic, but also for memory reasons. KV Cache for 405B…
-
comment
Comment #41411624
Why mention Microsoft twice?
-
comment
Comment #41298210
There was. Now second gen of that goes for $15.
-
comment
Comment #40373403
Overwhelming majority of flops is indeed spent on matmuls, but softmax disproportionately uses memory bandwidth, so it generally takes much longer than you'd expect from just looki…
-
comment
Comment #40143104
t5 is an architecture, t5x is a framework for training models that was created with that architecture in mind, but can be used to train other architectures, including decoder-only …
-
comment
Comment #39697438
To quote their official response "If the WSE weren't rectangular, the complexity of power delivery, I/O, mechanical integrity and cooling become much more difficult, to the point o…
-
comment
Comment #39697345
Not right now.
-
comment
Comment #39697341
Because SRAM stopped getting smaller with recent nodes.
-
comment
Comment #39697339
This thing targets training, which isn't affected by tiny accelerators inside CPUs.
-
comment
Comment #39697337
No, it's comparable to 230Mb of SRAM on Groq chip, since both of them are SRAM-only chips that can't really use external memory.
-
comment
Comment #39399652
I would say that Bradley is actually more valuable, since it can serve wider range of missions, while having higher crew survival rate and being more maneuverable.
-
comment
Comment #39399007
> they have 24,700,000 left of fighting age Without equipment, logistics and ammo to support it it's a dead weight. Also, it's very interesting that you omitted Gulf War, which wou…
-
comment
Comment #39398609
Wrong. Shells, artillery, drone components, engineering vehicles, tanks, APCs, jets, long-range missiles, anti-air defenses. 10x that and Ukraine starts winning again. 10x manpower…
-
comment
Comment #39398392
"Escalation management" through "slowly boiling the frog" combined with lack of coherent strategy for the war, partially because some western politicians are afraid of Russia's tan…
-
comment
Comment #39397906
Ukrainan Air Force has HARMs, but they are VERY limited in their capabilities due to them being employed from soviet-era jets. Basically area where target resides have to be pre-pr…
-
comment
Comment #38992659
For video to be a major contributor to a game size is a very rare occurrence. Games are 120GB+ in size because texture sizes continue to grow.
-
comment
Comment #38964911
Jetson Nano is a lot more expensive. But at the same price point as Coral you can buy Orange Pi 5 with 4 gigs of ram, 6 tops NPU(not supported as good as Coral though, but is gener…
-
comment
Comment #38930785
Vision Pro isn't really supposed to be used outside.