cicada-ai/Chanjing-Avatar-V2V-1.3B
Full fine-tune (fewer layers) vs Wan-AI/Wan2.1-T2V-1.3B. Wan-AI/Wan2.1-T2V-1.3B matches part of cicada-ai/Chanjing-Avatar-V2V-1.3B (824 of its 1233 tensors; the rest of cicada-ai/Chanjing-Avatar-V2V-1.3B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 30 of 30 layers.
Tensors changed: Attention 480/480, Norms 180/180, MLP 120/120, other 35/35, Embeddings 9/9