neuroturk/HYZ-01-0.6B
LoRA-style vs neuroturk/HYZ-01-0.6B-Base. 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 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/113, MLP 84/84, Attention 112/112
2 derivatives checked
- mradermacher/HYZ-01-0.6B-i1-GGUF: Quantization
- mradermacher/HYZ-01-0.6B-GGUF: Same weights, converted