kth8/LFM2.5-230M-OpenCode-Title-Generator
LoRA-style vs unsloth/LFM2.5-230M. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, out_proj, q_proj, v_proj, w1, w2, w3 changed, in 14 of 14 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, other 0/24, MLP 42/42, Norms 0/40, Attention 24/24
2 derivatives checked
- mradermacher/LFM2.5-230M-OpenCode-Title-Generator-GGUF: Same weights, converted
- kth8/LFM2.5-230M-OpenCode-Title-Generator-GGUF: Same weights, converted