Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by carbocation on Friday October 13, 2023 @01:27PM
(#37873138)
So you are saying that in the context of an article about low-rank adaptation, where I was talking about applying this to ResNets, there was actually confusion that I was referring to long range radio?
[ Parent
| on HN ]
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by wlesieutre on Friday October 13, 2023 @01:41PM
(#37873324)
I'm not the one who originally commented to complain about your capitalization, but sure.
Even if not in this specific comment thread, maybe having correct capitalization here means someone else knows how to write the one they mean to talk about in another conversation later, and then it will have been helpful.
If everyone went around calling both of them LORA all the time absolutely that's going to sometimes have people talking past each other about different technologies. Even with them capitalized correctly it's still going to, but not much.
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by xd1936 on Friday October 13, 2023 @01:38PM
(#37873270)
Sample size of one, but I clicked the article thinking that someone was using an LLM to "tune" the range performance of LoRa radio somehow. :(
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by enjenye on Friday October 13, 2023 @03:46PM
(#37874733)
Make that a sample size of two.
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by abraae on Friday October 13, 2023 @05:30PM
(#37875847)
Make that 3. I imagined somehow using an LLM to even further increase LoRa range, maybe sifting through noise.
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by drakenot on Friday October 13, 2023 @11:27PM
(#37877854)
Had you heard of LoRA in the context of machine learning before?
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by snitty on Friday October 13, 2023 @05:10PM
(#37875672)
As someone who is familiar with LoRa, but not LoRA, I was confused as to what they could have possibly been using LoRa for to finetune an LLM. The internet has a lot of acronyms, and a lot of jargon, and a lot of people doing seeming absurd things, okay?
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by ajsnigrutin on Friday October 13, 2023 @09:43PM
(#37877417)
Yep, i clicked on it, because i thought it was about LoRa (the radio protocol). LoRa has many capabilities and many uses, from simple signalization and telemetry data to full on mesh networks for chat based communication, and, well... there are many things that could be fine tuned there :)
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by HerculePoirot on Friday October 13, 2023 @10:12PM
(#37877549)
There is "LLM" in the title ...
Re:Takeaways from Hundreds of LLM finetuning experiments wit...
by ajsnigrutin on Saturday October 14, 2023 @04:50PM
(#37884138)
Well yeah, and meshtastic is a chat app over lora (radio)... why not have a LLM somewhere on lora?