mofosyne/TinyLLama-v0-5M-F16-llamafile
LoRA-style vs Maykeye/TinyLLama-v0. mofosyne/TinyLLama-v0-5M-F16-llamafile is a GGUF (F16) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, q_proj changed, in 8 of 8 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/17, MLP 0/24, Attention 16/32