htb-ac-1424625/Qwen3.8-2B-Distill-heretic

Partial fine-tune vs Qwen/Qwen3.5-2B. Partial change: 85% of the weights changed (attention, embeddings, mlp, norms), in 24 of 24 layers.

Tensors changed: Attention 132/264, Norms 43/177, Embeddings 1/4, MLP 72/172

2 derivatives checked

• mradermacher/Qwen3.8-2B-Distill-heretic-i1-GGUF: Quantization
• mradermacher/Qwen3.8-2B-Distill-heretic-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse