Lordvarun23/qwen2.5-vl-3b-mllmu-manu-forget5
Partial fine-tune vs Lordvarun23/qwen2.5-vl-3b-mllmu-ft. Partial change: 21% of the weights changed (mlp), in 36 of 36 layers.
Tensors changed: Embeddings 0/2, Norms 0/138, MLP 122/304, Attention 0/380
Partial fine-tune vs Lordvarun23/qwen2.5-vl-3b-mllmu-ft. Partial change: 21% of the weights changed (mlp), in 36 of 36 layers.
Tensors changed: Embeddings 0/2, Norms 0/138, MLP 122/304, Attention 0/380