timm/qwen3_vit_88m.qwen3_5_0_8b
Same weights, converted vs Qwen/Qwen3.5-0.8B. Same weights stored in another dtype (BF16/F32 → F32), plus 49 new tensors.
Tensors changed: Attention 0/48, Norms 0/48, Embeddings 0/1
Same weights, converted vs Qwen/Qwen3.5-0.8B. Same weights stored in another dtype (BF16/F32 → F32), plus 49 new tensors.
Tensors changed: Attention 0/48, Norms 0/48, Embeddings 0/1