oriental-lab/TinySwallow-1.5B-dolly
Full fine-tune vs oriental-lab/TinySwallow-1.5B-dolly_tl. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 37/57, MLP 84/84, Attention 194/196
1 derivatives checked
- mradermacher/TinySwallow-1.5B-dolly-GGUF: Same weights, converted