mergekit-community/Umbral-v0.4-1

Full fine-tune vs Cas-Warehouse/Llama-3-MopeyMule-Blackroot-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 35/65, MLP 96/96, Attention 128/128

1 derivatives checked

• mradermacher/Umbral-v0.4-1-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse