iselabvn/Tini-Cybersec-8B-A1B
LoRA-style vs LiquidAI/LFM2.5-8B-A1B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, out_proj, q_proj, v_proj, w1, w2, w3 changed, in 24 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: MLP 2118/2162, Embeddings 0/2, other 0/54, Norms 0/60, Attention 24/24
2 derivatives checked
- bartowski/iselabvn_Tini-Cybersec-8B-A1B-GGUF: Same weights, converted
- mradermacher/Tini-Cybersec-8B-A1B-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.