yanolja/YanoljaNEXT-EEVE-10.8B
LoRA-style vs upstage/SOLAR-10.7B-v1.0. 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 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/97, MLP 144/144, Attention 192/192
6 derivatives checked
- yanolja/YanoljaNEXT-EEVE-Instruct-10.8B: Full fine-tune
- teddylee777/EEVE-Korean-Instruct-10.8B-v1.0-gguf: Full fine-tune
- mradermacher/EEVE-Korean-10.8B-v1.0-i1-GGUF: Quantization
- mradermacher/EEVE-Korean-10.8B-v1.0-GGUF: Same weights, converted
- tensorblock/EEVE-Korean-10.8B-v1.0-GGUF: Quantization
- kurugai/Kurugai-EEVE-v1.0: LoRA-style