efficiencyx/Jun-v4-E4B-Step100-GGUF
LoRA-style (fewer layers) vs unsloth/gemma-4-E4B-it-qat-q4_0-unquantized. efficiencyx/Jun-v4-E4B-Step100-GGUF matches part of unsloth/gemma-4-E4B-it-qat-q4_0-unquantized (623 of its 2076 tensors; the rest of unsloth/gemma-4-E4B-it-qat-q4_0-unquantized has no counterpart). On the shared part: efficiencyx/Jun-v4-E4B-Step100-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 0/278, MLP 126/126, other 0/85, Attention 132/132