MLVXN/MicroLLM2
LoRA-style vs openai-community/gpt2-xl. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of c_attn, c_fc, c_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Attention 96/192, Norms 0/194, MLP 96/192, Embeddings 0/1
2 derivatives checked
- mradermacher/MicroLLM2-i1-GGUF: Quantization
- mradermacher/MicroLLM2-GGUF: Same weights, converted