Fica bem legivel. De noite com luz artificial também.
desvantagem em relação a eink é que rlcd precisa de alguns angulos especificos pra usar bem.
Estética é diferente mas isso é um ponto neutro
Fica bem legivel. De noite com luz artificial também.
desvantagem em relação a eink é que rlcd precisa de alguns angulos especificos pra usar bem.
Estética é diferente mas isso é um ponto neutro
This was super easy and cost ~$10. It works pretty well on a weak training set, and with a day or two of work I could make this sing.
徹夜してしまいましたw
残念でしたが頑張りましたよね
この後少し眠ります
タミール・ヘンデルマン アメリカのピアニスト
小気味よくスウィングしています
www.youtube.com/watch?v=RlCD...
#jazz
良い一日を・・・・・・・
徹夜してしまいましたw
残念でしたが頑張りましたよね
この後少し眠ります
タミール・ヘンデルマン アメリカのピアニスト
小気味よくスウィングしています
www.youtube.com/watch?v=RlCD...
#jazz
良い一日を・・・・・・・
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
- regular photography
- eink / tablets (fuckin LOVE my supernote)
- displays (RLCD!!)
- all kinds of LED lights and work accessories
a lot of the chinese stuff is absolutely beating the pants off western brands
- regular photography
- eink / tablets (fuckin LOVE my supernote)
- displays (RLCD!!)
- all kinds of LED lights and work accessories
a lot of the chinese stuff is absolutely beating the pants off western brands
> solve rlhf, rlvr, rlcd, rlhddvd etc ezpz
> future lightcone gives me infinities
> gradient collapses
> "have you heard of discounting"
> silence
> solve rlhf, rlvr, rlcd, rlhddvd etc ezpz
> future lightcone gives me infinities
> gradient collapses
> "have you heard of discounting"
> silence
The fact it's a classification engine, uses RLCD versus RLHF, looks easier / cheaper to train, is easier to use for classification tasks and scales well is pretty freaking huge.
The fact it's a classification engine, uses RLCD versus RLHF, looks easier / cheaper to train, is easier to use for classification tasks and scales well is pretty freaking huge.
The incredible RLCD DMG by @bytendo.bsky.social
🇫🇷 Coup de projecteur sur cet écran réfléchissant :
L'incroyable RLCD DMG par @bytendo.bsky.social
The incredible RLCD DMG by @bytendo.bsky.social
🇫🇷 Coup de projecteur sur cet écran réfléchissant :
L'incroyable RLCD DMG par @bytendo.bsky.social
they done good
they done good
Pas d’éclairage, mais une visibilité parfaite grâce à sa surface ultra-réflective.
Et côté autonomie… 45h avec 4 piles rechargeables 🤯
Ce RLCD-DMG est un vrai bijou signé @bytendo.bsky.social , monté sur une Super DMG-01 🦇
Pas d’éclairage, mais une visibilité parfaite grâce à sa surface ultra-réflective.
Et côté autonomie… 45h avec 4 piles rechargeables 🤯
Ce RLCD-DMG est un vrai bijou signé @bytendo.bsky.social , monté sur une Super DMG-01 🦇
... it's the RLCD who are wrong
... it's the RLCD who are wrong
"Programming in the Sun: A Year with the Daylight Computer"
wickstrom.tech/2025-10-10-p...
"RLCD represents #ReinforcementLearningForCalibratedDecisions, a training methodology optimised for structured decision-making rather than conversational text generatio...
"RLCD represents #ReinforcementLearningForCalibratedDecisions, a training methodology optimised for structured decision-making rather than conversational text generatio...
https://di-zhang-llm.github.io/blog/what-is-rlcd-the-secret-behind-jev/
https://di-zhang-llm.github.io/blog/what-is-rlcd-the-secret-behind-jev/
Proche de celui de la Playdate, on a l’impression de jouer sur une liseuse.
Le rendu est incroyable 🙏
Est-ce le meilleur écran pour profiter d’une DMG ?
Proche de celui de la Playdate, on a l’impression de jouer sur une liseuse.
Le rendu est incroyable 🙏
Est-ce le meilleur écran pour profiter d’une DMG ?
J’ai aussi remplacé les boutons par ceux en caoutchouc de Kitsch-Bent : parfait pour de longues sessions, beaucoup plus agréables… et silencieuses, mon entourage valide 🙈
J’ai aussi remplacé les boutons par ceux en caoutchouc de Kitsch-Bent : parfait pour de longues sessions, beaucoup plus agréables… et silencieuses, mon entourage valide 🙈
"RLCD represents #ReinforcementLearningForCalibratedDecisions, a training methodology optimised for structured decision-making rather than conversational text generatio...
"RLCD represents #ReinforcementLearningForCalibratedDecisions, a training methodology optimised for structured decision-making rather than conversational text generatio...
RLCD models
crashynews.wordpress.com/2026/05/13/b...
RLCD models