Nexesenex/Llama_3.1_70b_Harpies_V1
Full fine-tune vs NobodySpecial/Llama-3.1-70B-Instruct-Lorablated-Creative-Writer. Fine-tune: 99% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 31/161, MLP 240/240, Attention 320/320, lm_head 0/1
3 derivatives checked
- mradermacher/Llama_3.x_70b_Harpies_V1-i1-GGUF: Quantization
- Nexesenex/Llama_3.x_70b_Nemeslices_V1: Full fine-tune
- Nexesenex/Llama_3.x_70b_Doberwhack_V1.0: Full fine-tune