daslab-testing/Apertus-v1.1-0.5B-Instruct
Full fine-tune vs swiss-ai/Apertus-v1.1-0.5B. Fine-tune: 100% of the weights changed, across 20 of 20 layers.
Tensors changed: Embeddings 1/1, Norms 81/81, MLP 75/120, Attention 80/80
2 derivatives checked
- mradermacher/Apertus-0.5B-Instruct-GGUF: Same weights, converted
- mradermacher/Apertus-v1.1-0.5B-Instruct-GGUF: Same weights, converted
2 other repos hold byte-for-byte identical weight files.