失败猫猫 retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
失败猫猫 retweeted
太逆天了，人看来就是一堆浮点数组成的网络，这他妈还能逐个微调，把这种知识用自然语表述出来，DL可解释性可以半截入土了
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
失败猫猫 retweeted
The way different tiers of AI companies train their models mirrors exactly how different tiers of humans are trained:
The top tier gets trained by RL in the real world. The lower tiers get trained by SFT and distillation from others.
失败猫猫 retweeted
我有个女性朋友，今年刚好45岁。
前段时间她在街头崩溃大哭，很多人劝她现实一点，别一心想找年入百万的对象，赶紧降低要求。
那段时间她也很困惑，主动追她的基本都是年纪小的弟弟，欣赏她成熟独立松弛的气质，同龄和年长男生却很少主动。旁人都跟她说，既然只有弟弟追，就别挑了，将就算了。
但她从来没被这些话 PUA 到，一直守住自己的标准，不肯妥协。她本身条件不错，有稳定行政工作，平时健身护肤、读书旅行看展，把日子过得充实精致。
就在今年，一次行业活动里她认识了老公，28 岁一米八四的创业者，外形清爽亮眼，能力过硬，有房有车，年薪百万以上。一开始大家还不看好年龄差，男生却认准她，就喜欢她清醒通透、有主见的样子。
谈婚论嫁的时候诚意满满，主动备好丰厚彩礼，知道她家里负担重，出钱帮家里减轻压力。婚后工资卡直接上交，房产愿意加上她名字，家务主动分担，事事体贴。
现在她被好好宠爱，工作也没有放下，男生还一直支持她进修提升。你看，主动来的人是什么年纪，不代表你只能选这类人。
千万别因为追你的大多是弟弟，就被迫降低期待。先把自己过好，符合你标准的优质人，才会向你奔赴。婚姻是一辈子的事，就要找真心珍惜你、愿意为你付出的，这才值得。😆😋😃
失败猫猫 retweeted
亚历山大·王（Alexandr Wang）在谈到“如何拉开与 99% 人的差距”时，给出了极其朴素甚至残酷的答案：
顶层竞争根本不拼智商，拼的是极端执行力的非对称倾斜。
他的三条核心准则：
1. 绝对聚焦，压榨认知带宽： 绝大多数人的忙碌只是“低密度伪勤奋”。真正的专注是把全部精力砸在唯一的核心支点上，拒绝任何无关干扰。
2. 把“过度交付”做成护城河： 行业标准要求走 1 英里，你就走完 10 英里。当所有人都按合同完成 100% 时，常态化交付 200% 的人会直接卷走全部声誉与溢价。
3. 带靶向的超额输出，拒绝无目的内耗： 努力不是盲目自我感动，而是针对具体目标的持续过载测试。把每一次交付都当成生死决战来打。
商业世界最公平的一点在于：
聪明人到处都是，但能忍受极度枯燥并长期超额交付的人凤毛麟角。
执行力溢出，就是最暴力的护城河。
失败猫猫 retweeted
It's 2026. Your lab throws the biggest Halloween party to celebrate the Death of Mathematics. PIs are wasted in the coffee room, Phds and postdocs dance around a bonfire from obsolete preprints, even Reviewer 2 is having fun
And then the DJ (Fields medal) throws this banger 🎉