MegaScience/Qwen3-1.7B-MegaScience

Full fine-tune vs Qwen/Qwen3-1.7B-Base. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 107/113, MLP 84/84, Attention 112/112

2 derivatives checked

• mradermacher/Qwen3-1.7B-MegaScience-i1-GGUF: Quantization
• mradermacher/Qwen3-1.7B-MegaScience-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse