jaspionjader/Kosmos-Elusive-VENN-Asymmetric-8B
Full fine-tune vs jaspionjader/Kosmos-Elusive-VENN-8B. Fine-tune: 97% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 93/96, Attention 124/128
3 derivatives checked
- mradermacher/Kosmos-Elusive-VENN-Asymmetric-8B-i1-GGUF: Quantization
- mradermacher/Kosmos-Elusive-VENN-Asymmetric-8B-GGUF: Same weights, converted
- jaspionjader/Kosmos-EVAA-8B: Full fine-tune