jaspionjader/Kosmos-EVAA-v12-8B
Full fine-tune vs jaspionjader/Kosmos-EVAA-v11-8B. Fine-tune: 94% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 18/65, MLP 89/96, Attention 117/128
3 derivatives checked
- mradermacher/Kosmos-EVAA-v12-8B-GGUF: Same weights, converted
- jaspionjader/gamma-Kosmos-EVAA-8B: Full fine-tune
- jaspionjader/Kosmos-EVAA-gamma-8B: Full fine-tune