Repository Issue Activity (beta)

#19 YOU ARE THE GOAT，NAO MA LE

#23 No feasible way to train/low efficiency

#65 Feature request: For MLA version, eliminate pre-attention RMSNorm via scale invariance + weight folding

#44 Any evidence for Mythos using Loop Transformer(LT)?

#46 dataset

#63 Open mythos

#54 This is still dumb

#24 No CPU NPU TPU

#8 Provide sample inference script

#28 Empirical test of the depth-extrapolation claim: U-shape with fixed-loop training, flat plateau with random-loop training

#9 Provide benchmark

#29 This is still dumb.

#52 Third-party Claude skill for OpenMythos — would you like a link in the README?

#47 Claude overall

#38 Really or not?

#51 Incorrect spectral radius calculation in README

#40 Discussion board?

#48 优化建议: 添加Flash Attention和混合精度训练支持

#25 是你的建議在ZeusHammer中添加你的這個項目嗎？

#21 Failed to prove at small scale

#5 Real-world applications of this architecture

#20 Behavioral distillation from outputs alone — empirical results on Mythos-level behavior

#16 闹麻了简直

#18 Every star here is stolen from someone who actually trained a model.

#17 🚨 Stop chasing the hype!!! There is no model in this repo