BSC-LT/salamandra-TAV-7b
Full fine-tune vs BSC-LT/salamandraTA-7b-instruct. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/salamandra-TAV-7b-i1-GGUF: Quantization
- mradermacher/salamandra-TAV-7b-GGUF: Same weights, converted