win10/miscii-14b-1M-0128

Full fine-tune vs SicariusSicariiStuff/Impish_QWEN_14B-1M. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: Norms 97/97, MLP 144/144, Attention 336/336

1 derivatives checked

On the Hub · Downloads on Model Pulse