Последние новости
We have one horrible disjuncture, between layers 6 → 2. I have one more hypothesis: A little bit of fine-tuning on those two layers is all we really need. Fine-tuned RYS models dominate the Leaderboard. I suspect this junction is exactly what the fine-tuning fixes. And there’s a great reason to do this: this method does not use extra VRAM! For all these experiments, I duplicated layers via pointers; the layers are repeated without using more GPU memory. Of course, we do need more compute and more KV cache, but that’s a small price to pay for a verifiably better model. We can just ‘fix’ an actual copies of layers 2 and 6, and repeat layers 3-4-5 as virtual copies. If we fine-tune all layer, we turn virtual copies into real copies, and use up more VRAM.，详情可参考safew
<Dependency path="/totalTax"/>，这一点在手游中也有详细论述
For multiple readers，这一点在官网中也有详细论述