navflyer/aries-1.5b-reasoning

Full fine-tune vs Qwen/Qwen2-1.5B-Instruct. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Norms 34/57, MLP 84/84, Attention 193/196

2 derivatives checked

• mradermacher/aries-1.5b-reasoning-i1-GGUF: Quantization
• mradermacher/aries-1.5b-reasoning-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse