sillykiwi/Nano-Imp-1B-OBLITERATED

Partial fine-tune vs SicariusSicariiStuff/Nano_Imp_1B. Partial change: 34% of the weights changed (attention, lm_head, mlp), in 4 of 16 layers.

Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 0/33, MLP 12/48, Attention 16/64

1 derivatives checked

• mradermacher/Nano-Imp-1B-OBLITERATED-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse