Duong1211/GRIT
Partial fine-tune vs Qwen/Qwen2.5-0.5B-Instruct. Partial change: 49% of the weights changed (attention, mlp, norms), in 24 of 24 layers.
Tensors changed: Embeddings 0/1, Norms 9/49, MLP 47/72, Attention 96/168
Partial fine-tune vs Qwen/Qwen2.5-0.5B-Instruct. Partial change: 49% of the weights changed (attention, mlp, norms), in 24 of 24 layers.
Tensors changed: Embeddings 0/1, Norms 9/49, MLP 47/72, Attention 96/168