Merlin-Research/HybridIntelligence-0.5B
LoRA-style vs tiiuae/Falcon-H1-0.5B-Base. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/73, MLP 0/108, other 0/252, Attention 72/144
1 derivatives checked
- sizzlebop/HybridIntelligence-0.5B-GGUF: Same weights, converted