agentica-org/DeepScaleR-1.5B-Preview
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 44/57, lm_head 0/1
7 derivatives checked
- bartowski/agentica-org_DeepScaleR-1.5B-Preview-GGUF: Same weights, converted
- DevQuasar-12/agentica-org.DeepScaleR-1.5B-Preview-GGUF: Same weights, converted
- DhruvalLabs/DeepScaleR-1.5B-Preview-GGUF: Quantization
- Melvin56/DeepScaleR-1.5B-Preview-GGUF: Same weights, converted
- yasserrmd/DeepScaleR-1.5B-Preview-gguf: Partial fine-tune
- YU-MO/Yumo-nano: LoRA-style
- yolay/RAIF-DeepScaleR-1.5B: Partial fine-tune
4 other repos hold byte-for-byte identical weight files.