THU-KEG/AdaptThink-1.5B-delta0.1

Partial fine-tune vs deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B. Partial change: 74% of the weights changed (attention, mlp, norms), in 28 of 28 layers.

Tensors changed: Embeddings 0/1, Attention 196/196, MLP 84/84, Norms 37/57, lm_head 0/1

3 derivatives checked

• mradermacher/AdaptThink-1.5B-delta0.1-i1-GGUF: Quantization
• mradermacher/AdaptThink-1.5B-delta0.1-GGUF: Same weights, converted
• DevQuasar/THU-KEG.AdaptThink-1.5B-delta0.1-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse