mradermacher/Radomir_4B-GGUF
LoRA-style (fewer layers) vs KoDer123/Radomir_4B. mradermacher/Radomir_4B-GGUF matches part of KoDer123/Radomir_4B (444 of its 883 tensors; the rest of KoDer123/Radomir_4B has no counterpart). On the shared part: mradermacher/Radomir_4B-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 34 of 34 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/205, MLP 102/102, Attention 136/136