JY623/KoSOLRA-10.7B-merge-v2.2
LoRA-style vs JY623/KoSOLAR-10.7B-merge-v2.0. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 144/144, Attention 0/192, lm_head 0/1
2 derivatives checked
- mradermacher/KoSOLRA-10.7B-merge-v2.2-i1-GGUF: Quantization
- mradermacher/KoSOLRA-10.7B-merge-v2.2-GGUF: Same weights, converted