#Finetuning
More finetuning! 🐮
September 15, 2025 at 8:08 PM
finetuning my GBA shader
November 15, 2024 at 6:50 AM
It's been a year of #Jayvik for me!

Ehhhh artist growth is slow sometimes but I think I have been finetuning my style the past year.
October 5, 2025 at 9:11 PM
it's more about local finetuning being viable, training data cutoff etc
September 21, 2026 at 2:53 AM
Finetuning the texture
July 21, 2025 at 12:15 PM
🇳🇱 Puck Pieterse finetuning her 🇫🇷 Roubaix celebration 😅

🎥 Strava Puck Pieterse
February 26, 2026 at 7:45 AM
Doing some finetuning on my custom printer, wanna help? >w<
March 17, 2026 at 7:46 PM
Finetuning our Digital Hunks for 2027!

#BrawnyAi #DigitalHunks
September 4, 2026 at 3:25 AM
today one of our media editing suppliers asked for consent to create a deepfake of my voice and sample data reading a script with all the major phonemes and I went nope nope nope nope nope

I am okay with finetuning voice recognition, but not letting a corporation literally put words in my mouth.
July 22, 2026 at 11:53 PM
Finetuning water reflections 🪞

#indiegamedev #gamedev #unity3d #realtimevfx #shader
December 10, 2025 at 12:41 AM
A few of you asked how current #bikefit looks for me - just over 3mths with prosthetics. Much ongoing testing but a stable safe start for training before any finetuning. Thanks to @trekbikes.bsky.social Custom4.us @philburt.bsky.social for support and advice.
#ParaCycling #amputee #BKA #roadcycling
August 14, 2026 at 8:43 AM
How do language models generalize from information they learn in-context vs. via finetuning? In arxiv.org/abs/2505.00661 we show that in-context learning can generalize more flexibly, illustrating key differences in the inductive biases of these modes of learning — and ways to improve finetuning. 1/
arxiv.org
May 2, 2025 at 5:02 PM
A useful oversimplification
Instruction finetuning (IFT/SFT): imprinting features or shape in responses
Preference finetuning (RLHF/DPO/etc): style
Reinforcement finetuning (RFT/RLVR/etc): learning new behaviors
January 31, 2025 at 2:31 PM
i legitimately cannot see as a matter of design how to go from where we are to non-asskissing LLMs. it's fundamental to how finetuning data is sourced: people rank ass-kissing data higher.
July 20, 2025 at 7:47 PM
I would say in all my time working on data problems, theres basically 5% of problems that cant be solved by:
- XGBoost
- finetuning resnet 50
- finetuning BERT
July 31, 2025 at 1:07 AM
Finetuning. 😊 (#OTD in 1943. A self-propelled gun type "Bishop", a 25-pounder gun mounted on a Valentine. #WW2 #HISTORY)
July 25, 2026 at 4:44 AM
Finetuning. 😊 (#OTD in 1943. A self-propelled gun named "Bishop", a 25-pounder gun mounted on a Valentine. #WW2 #HISTORY)
July 25, 2025 at 5:43 AM
Finetuning the sword-spam (more on that later), I'm also building the last missing level-corners. QoL improvement is that the game now remembers pushed blocks when they are in place. #gamedev #indiegame #solodev
September 24, 2026 at 10:49 PM
finetuning humor one post at a time 🙏
June 19, 2025 at 1:45 PM
✨️Ok Lets Try Adopts Again✨️

Made these two Deviant Pokemon Designs, would pherhaps 💥£60💥 for design + finetuning/additions to get perfect for what you want sound reasonable?

#AvenueArcania
June 18, 2026 at 11:02 AM
yeah definitely looks like finetuning, i am 50/50 between the answering as elon being undisclosed fuckery with the prompt or their finetuning dataset just being Like That. just upsample elon twenty times
July 8, 2025 at 7:45 PM
FM radio issues:
Beautiful Fame Girl Foxy finetuning the radio...
September 24, 2026 at 8:13 AM
Der Plan braucht finetuning.
July 9, 2025 at 10:51 AM
one of the Basic Things About CLIP is that it is extraordinarily easy to break during finetuning
October 11, 2025 at 6:26 PM