Khetterman/Kosmos-8B-v1
Full fine-tune vs Arkana08/LexiMaid-L3-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128
5 derivatives checked
- mradermacher/Kosmos-8B-v1-i1-GGUF: Quantization
- featherless-ai-quants/Khetterman-Kosmos-8B-v1-GGUF: Quantization
- mradermacher/Kosmos-8B-v1-GGUF: Same weights, converted
- jaspionjader/Kosmos-VENN-8B: Full fine-tune
- jaspionjader/Kosmos-Elusive-8b: Full fine-tune