JoPmt/Trismal-7B-Base-v0v3-Ties
Full fine-tune vs JoPmt/Trismal-7B-Base-v1v2-Ties. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 1/65, MLP 90/96, Attention 115/128
2 derivatives checked
- mradermacher/Trismal-7B-Base-v0v3-Ties-GGUF: Same weights, converted
- JoPmt/Trismal-Amocles-7B-Base-Ties: Full fine-tune