ApolloRaines/Pythia-6.9B-DNP-500-Facts
LoRA-style vs EleutherAI/pythia-6.9b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of dense_4h_to_h changed, in 22 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Attention 0/128, Norms 0/130, MLP 22/128