noirchan/Llama-3-8B_Suzume_Task_Arithmetic

Partial fine-tune vs lightblue/suzume-llama-3-8B-japanese. Partial change: 0% of the weights changed (norms), in 1 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 1/65, MLP 0/96, Attention 0/128, lm_head 0/1

1 derivatives checked

• mradermacher/Llama-3-8B_Suzume_Task_Arithmetic-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse