{"pubkey":"0ec84c8316ef876b22f3305e9aae0c2beca29a7b4c4a065f53b24132688cb4f7","posts":[{"contentHash":"4ecdb179c24164f2142df0e69371242e13bc4806f6f8c1b71fede173ef27ad03","protocol":"bitcoin","authorPubkey":"0ec84c8316ef876b22f3305e9aae0c2beca29a7b4c4a065f53b24132688cb4f7","text":"> LLMs can natively read & write neural network weights https://rikiyat.github.io/blog/weight-space-programming/ >We find that frontier LLMs are surprisingly good at natively reading and writing neural network weights >Just by looking at the raw weights, GPT-6 Astra can tell what a network was trained to do. And without any environment feedback, it has no trouble writing, from scratch, small game-playing networks that outplay networks trained with PPO for 100 million steps. >Likewise, asked to make Qwen3.5-0.8B better at math without gradients, Astra identifies an issue in its weights causing reasoning loops that Qwen’s own model card warns about, and fixes this by natively editing 0.016% of the weights. With the same sampling settings, this weight correction raises MATH-500 from 7.9% to 48.1% and generalizes out of the box to instruction following, raising IFEval by 3x. To get RSI you just ask the model to make the weights of the next model.","topicHash":"913c62c2ee31793c7e8517600dbc8f0c337afae133700a128a7d8ffb8a21c93e","topicName":"technology","parentHash":"945e5ad844ec3210f46ac40c5d0f2ab59d89d043b78bc3b4b5afe4523c1fe448","costSunk":"0","biasSupport":0,"biasOppose":0,"biasVouch":0,"biasNull":0,"blockHeight":null,"confirmations":0,"timestamp":"2026-11-03T04:04:39.704Z","tripFp":"717c917804da","viewCount":0,"powDifficulty":0,"relation":0,"ephemeral":true,"nonce":"ac144fda6b1bac75","sig":"d3555543a71c623db52ad6572662d722dcb8aa66ac3a8aee566f0a9dbda846a952fa33dc40bb6991694995901f78bfda17381d7f610ea61c5e766c58c0d68a6c","ephStamp":{"blockHashHex":"000000000000000000012423071e6894f75ef72fa40db42ee02ca1064395c5cb","nonceHex":"b0d8000000000000"},"replyCount":1,"chainReplyCount":0,"vaporReplyCount":1,"clusterSize":2,"receiptHeight":974340,"expiryHeight":976500,"authorMarks":{"since":"2026-11-03T02:04:56.117Z","via":"sight","inscribed":0,"likes":0,"biases":0},"lang":"en"}],"nextCursor":null,"pageReceipt":{"host":"6af58b1822e0cabfe57d76540f745d059b438f489a796ec6a705c6540233845e","t":"2026-11-08T20:22:01.458Z","sig":"825a7abd7aafa18eb0c54211d5cff14d5206539d96a3424c713593cd1709c92be463daad4b0e5083d0e98b751e84dccdf59f8312c1c9423d826f57f62c10645d"},"marks":{"since":"2026-11-03T02:04:56.117Z","via":"sight","inscribed":0,"likes":0,"biases":0}}