File size: 640 Bytes
0823729 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 |
base_model: Qwen/Qwen2-1.5B
architecture: qwen
experts:
- source_model: Qwen/Qwen2-1.5B
positive_prompts:
- "chat"
- "assistant"
- "tell me"
- "explain"
- "I want"
- source_model: Replete-AI/Replete-Coder-Qwen2-1.5b
positive_prompts:
- "code"
- "python"
- "javascript"
- "programming"
- "algorithm"
shared_experts:
- source_model: Qwen/Qwen2-1.5B
positive_prompts: # required by Qwen MoE for "hidden" gate mode, otherwise not allowed
- "chat"
# (optional, but recommended:)
residual_scale: 0.1 # downweight output from shared expert to prevent overcooking the model
|