Circuit Aware Rushd 8B
رُشْد — "guidance" or "right path" in Arabic
Fuente del modelo
- Tipo
- LoRA
- Tamaño del archivo
- 83,3 MB
- Modelo base
- Qwen/Qwen3-8B
- Versión
- vadapters
- Actualizado
- 4 oct 2026
- Publicador
- @hayulalab
- Descargas
- 0
- Parámetros
- 8B
- Fuente
- Hugging Face
Descripción de la fuente
Circuit-Aware Rushd-8B
رُشْد — "guidance" or "right path" in Arabic
LoRA adapter for Qwen3-8B, trained with Circuit-Aware methodology to enhance multi-expert reasoning for cybersecurity applications.
Training Details
- Base Model: Qwen/Qwen3-8B
- Rank: 8 (scale: 20)
- Iterations: 500
- Batch Size: 4
- Learning Rate: 1e-5
- Val Loss: 2.148
- Train Loss: 0.030
Capabilities
Enhances 5 Rushd specialists: Router, Logic, Plan, Decision, Game
Fuentes
1 fuente
Fuentes
1 fuenteVerificado 4 oct
Artefactos del modelo
1 artefacto
Artefactos del modelo
1 artefactoExtractos de fuentes
3 extractos
Extractos de fuentes
3 extractos--- library_name: mlx tags: - mlx - lora - qwen3 - security - arabic - circuit-aware base_model: Qwen/Qwen3-8B --- # Circuit-Aware Rushd-8B **رُشْد** — "guidance" or "right path" in Arabic LoRA adapter for Qwen3-8B, trained with Circuit-Aware methodology to enhance multi-expert reasoning for cybersecurity applications. ## Training Details - **Base Model:** Qwen/Qwen3-8B - **Rank:** 8 (scale: 20) - **Iterations:** 500 - **Batch Size:** 4 - **Learning Rate:** 1e-5 - **Val Loss:** 2.148 - **Train Loss:** 0.030 ## Capabilities Enhances 5 Rushd specialists: Router, Logic, Plan, Decision, Game
Source context: 0 downloads · 0 likes · Library mlx · Repo hayulalab/Circuit-Aware-Rushd-8B
Source context: 0 downloads · 0 likes · Library mlx · Repo hayulalab/Circuit-Aware-Rushd-8B
