Susant-Achary/LFM2.5-8B-A1B-UltraData-L2-49M
Full fine-tune vs LiquidAI/LFM2.5-8B-A1B-Base. Fine-tune: 97% of the weights changed, across 24 of 24 layers.
Tensors changed: MLP 2118/2162, Embeddings 0/2, other 36/54, Norms 0/60, Attention 24/24
Full fine-tune vs LiquidAI/LFM2.5-8B-A1B-Base. Fine-tune: 97% of the weights changed, across 24 of 24 layers.
Tensors changed: MLP 2118/2162, Embeddings 0/2, other 36/54, Norms 0/60, Attention 24/24