mradermacher/pythia-2.8b-GGUF

Same weights, converted vs EleutherAI/pythia-2.8b. Same weights stored in another dtype (F16/U8 → F16/F32).

Tensors changed: Embeddings 0/2, Norms 0/130, Attention 0/64, MLP 0/128

On the Hub · Downloads on Model Pulse