mergekit-community/L3.3-L3.1-NewTempusBlated-70B

Full fine-tune vs nbeerbower/Llama-3.1-Nemotron-lorablated-70B. Fine-tune: 99% of the weights changed, across 80 of 80 layers.

Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 99/161, MLP 240/240, Attention 320/320

1 derivatives checked