QuantFactory/gpt2-GGUF
Full fine-tune vs openai-community/gpt2. QuantFactory/gpt2-GGUF is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Norms 50/50, Attention 24/24, MLP 24/24, Embeddings 2/2