McGill-NLP/AfriqueQwen3.5-4B-Instruct-v1
Full fine-tune (fewer layers) vs McGill-NLP/AfriqueQwen3.5-4B. McGill-NLP/AfriqueQwen3.5-4B-Instruct-v1 matches part of McGill-NLP/AfriqueQwen3.5-4B (427 of its 724 tensors; the rest of McGill-NLP/AfriqueQwen3.5-4B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 84/105, Attention 196/224, MLP 96/96
2 derivatives checked
- mradermacher/AfriqueQwen3.5-4B-Instruct-v1-i1-GGUF: Quantization
- mradermacher/AfriqueQwen3.5-4B-Instruct-v1-GGUF: Same weights, converted