jaspionjader/Kosmos-EVAA-gamma-light-8B
Full fine-tune vs jaspionjader/Kosmos-EVAA-gamma-8B. Fine-tune: 95% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 90/96, Attention 120/128
7 derivatives checked
- mradermacher/Kosmos-EVAA-gamma-light-8B-GGUF: Same weights, converted
- MaziyarPanahi/Kosmos-EVAA-gamma-light-8B-GGUF: Same weights, converted
- jaspionjader/Kosmos-EVAA-gamma-v16-8B: Full fine-tune
- jaspionjader/Kosmos-EVAA-gamma-v14-8B: Partial fine-tune
- jaspionjader/Kosmos-EVAA-gamma-v13-8B: Full fine-tune
- jaspionjader/TSN-Kosmos-EVAA-v2-8B: Full fine-tune
- jaspionjader/Kosmos-EVAA-gamma-ultra-light-8B: Full fine-tune