TokenRhythm/NeoHorse-1-4B-MLX
Partial fine-tune vs TokenRhythm/NeoHorse-1-4B. Partial change: 0% of the weights changed (norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 81/105, Attention 0/200, MLP 0/96
Partial fine-tune vs TokenRhythm/NeoHorse-1-4B. Partial change: 0% of the weights changed (norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 81/105, Attention 0/200, MLP 0/96