immortaltatsu/ghostai-lfm2.5-1.2b-app-v2
Full fine-tune vs LiquidAI/LFM2.5-1.2B-Thinking. Fine-tune: 100% of the weights changed, across 16 of 16 layers.
Tensors changed: Embeddings 1/2, other 30/30, MLP 48/48, Norms 31/44, Attention 24/24
Full fine-tune vs LiquidAI/LFM2.5-1.2B-Thinking. Fine-tune: 100% of the weights changed, across 16 of 16 layers.
Tensors changed: Embeddings 1/2, other 30/30, MLP 48/48, Norms 31/44, Attention 24/24