ApolloRaines/Pythia-1.4B-DNP-16750-Facts
LoRA-style vs EleutherAI/pythia-1.4b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of dense_4h_to_h changed, in 8 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Norms 0/98, Attention 0/96, MLP 8/96