kcj1n2020/EXAONE-7.8B-json
LoRA-style vs LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of c_fc_0, c_fc_1, c_proj, k_proj, out_proj, q_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Attention 128/128, Norms 0/65, MLP 96/96, Embeddings 0/1, lm_head 0/1
1 derivatives checked
- mradermacher/EXAONE-7.8B-json-GGUF: Same weights, converted