Xiaojian9992024/Qwen2.5-Ultra-1.5B-25.02-Exp

Full fine-tune vs SakanaAI/TinySwallow-1.5B-Instruct. Fine-tune: 97% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 57/57, MLP 81/84, Attention 189/196

1 derivatives checked

• mradermacher/Qwen2.5-Ultra-1.5B-25.02-Exp-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse