aimeri/spoomplesmaxx-magpie-35B-A3

Full fine-tune vs aimeri/spoomplesmaxx-jackdaw-35B-A3. Fine-tune: 97% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 0/4, Norms 78/241, Attention 221/388, MLP 240/392, lm_head 1/1

1 derivatives checked

On the Hub · Downloads on Model Pulse