strykes/emberforge-3b-reasoner
LoRA-style vs Nanbeige/Nanbeige4.1-3B. 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 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, MLP 96/96, Attention 128/128, Norms 0/65, Embeddings 0/1
1 derivatives checked
- Flexan/strykes-emberforge-3b-reasoner-GGUF: Same weights, converted