BramVanroy/fietje-2-instruct
Full fine-tune vs BramVanroy/fietje-2. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 66/66, MLP 128/128, Attention 256/256
4 derivatives checked
- mradermacher/fietje-2-instruct-i1-GGUF: Quantization
- BramVanroy/fietje-2-chat: Full fine-tune
- mradermacher/fietje-2-instruct-GGUF: Same weights, converted
- ekrombouts/zuster_fietje: Full fine-tune