All credits belong to Luffy the Fox: https://huggingface.co/LuffyTheFox This is just a backup copy from his work! ⚡ https://web.tribute.tg/d/KIH ⚡ If you like this Genesis LLM release you can donate to me via @Tribute bot in Telegram messenger and support future Genesis LLM development. 🌟 Qwen3.6 35B A3B Uncensored HauhauCS Aggressive Genesis Hermes V4 Key diffrence is data reconstruction with noise supression and signal calibration in ssm out.weight, attn output.weight, attn gate.weight, attn qkv.weight, attn q.weight, attn k.weight, attn v.weight ffn down exps.weight, ffn gate exps.weight, ffn up exps.weight, ssm alpha.weight and ssm beta.weight tensors via SVD. Mine approach based on data reconstruction in model via mathematical statistics. I don't train models, I repair signal in them instead. I scan blocks in model via chunks via 3 parameters and pick best one that fits to weight distribution in tensor. Best picked chunk replaces zero chunks in broken tensor without touching learned structure. Scanning works on tensors with same name and shape. ssm conv1d tensors are fixed via alpha multiply for full tensor. I scan all ssm conv1d tensors weight and scale distribution and norm…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy