swiss-ai/Apertus-v1.1-4B-Instruct
Full fine-tune vs swiss-ai/Apertus-v1.1-4B. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 1/1, Norms 97/97, MLP 81/144, Attention 96/96, lm_head 1/1
5 derivatives checked
- DMP96/Apertus-v1.1-4B-Instruct-QAD-Q4_1-GGUF: Full fine-tune
- agentlans/Apertus-v1.1-4B-Instruct-GGUF: Quantization
- MrMeOrYou/Apertus-v1.1-4B-Instruct-GGUF: Same weights, converted
- loleg/Apertus-v1.1-4B-Instruct-mlx: Quantization
- jondale/Apertus-v1.1-4B-Instruct-GGUF: Quantization