doguilmak/inferencevision-pythia-1B

Partial fine-tune vs EleutherAI/pythia-1b. Partial change: 90% of the weights changed (attention, embeddings, mlp, norms), in 16 of 16 layers.

Tensors changed: Embeddings 1/2, Norms 63/66, Attention 64/64, MLP 64/64

1 derivatives checked

• mradermacher/inferencevision-pythia-1B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse