mradermacher/ablation-Qwen2_5-G0-Rmmseed-GPT4o-adv5-4-12k-GGUF
Same weights, converted (fewer layers) vs LegendaryDawn/ablation-Qwen2_5-G0-Rmmseed-GPT4o-adv5-4-12k. mradermacher/ablation-Qwen2_5-G0-Rmmseed-GPT4o-adv5-4-12k-GGUF matches part of LegendaryDawn/ablation-Qwen2_5-G0-Rmmseed-GPT4o-adv5-4-12k (339 of its 729 tensors; the rest of LegendaryDawn/ablation-Qwen2_5-G0-Rmmseed-GPT4o-adv5-4-12k has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 0/196, lm_head 0/1