mradermacher/Fawen-1.0-8B-GGUF
LoRA-style (fewer layers) vs chinapao/Fawen-1.0-8B. mradermacher/Fawen-1.0-8B-GGUF matches part of chinapao/Fawen-1.0-8B (677 of its 2130 tensors; the rest of chinapao/Fawen-1.0-8B has no counterpart). On the shared part: mradermacher/Fawen-1.0-8B-GGUF is a GGUF (F16) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 42 of 42 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Norms 0/296, MLP 126/126, other 0/85, Attention 132/168