DoeyLLM/OneLLM-Doey-V1-Llama-3.2-1B-it
LoRA-style vs meta-llama/Llama-3.2-1B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 16 of 16 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/48, Attention 32/64
3 derivatives checked
- DoeyLLM/OneLLM-Doey-V1-Llama-3.2-1B-it-Q8_0-GGUF: Quantization
- mradermacher/OneLLM-Doey-V1-Llama-3.2-1B-it-i1-GGUF: Quantization
- mradermacher/OneLLM-Doey-V1-Llama-3.2-1B-it-GGUF: Same weights, converted