harpertoken/tiny
Partial fine-tune vs mlx-community/SmolLM-135M-Instruct-4bit. Partial change: 42% of the weights changed (attention, mlp), in 16 of 30 layers.
Tensors changed: Embeddings 0/3, Norms 0/61, MLP 144/270, Attention 192/360
Partial fine-tune vs mlx-community/SmolLM-135M-Instruct-4bit. Partial change: 42% of the weights changed (attention, mlp), in 16 of 30 layers.
Tensors changed: Embeddings 0/3, Norms 0/61, MLP 144/270, Attention 192/360