base_model: cognitivecomputations/dolphin-2.9.3-qwen2-1.5b | |
gate_mode: hidden | |
architecture: qwen | |
dtype: bfloat16 | |
experts_per_token: 2 | |
experts: | |
- source_model: cognitivecomputations/dolphin-2.9.3-qwen2-1.5b | |
positive_prompts: | |
- "chat" | |
- "explain" | |
- "describe" | |
- "define" | |
- "help" | |
- source_model: Replete-AI/Replete-Coder-Qwen2-1.5b | |
positive_prompts: | |
- "code" | |
- "algorithm" | |
- "programming" | |
- "development" | |
- "software" | |
- "framework" | |
- source_model: Qwen/Qwen2-1.5B-Instruct | |
positive_prompts: | |
- "assistant" | |
- "translate" | |
- "summarize" | |
- "rewrite" | |
- "multilingual" | |
- source_model: macadeliccc/Samantha-Qwen2-1.5B | |
positive_prompts: | |
- "characters" | |
- "scene" | |
- "roleplay" | |
- "writing" | |
- "creative" | |
- "acting" | |
shared_experts: | |
- source_model: cognitivecomputations/dolphin-2.9.3-qwen2-1.5b | |
positive_prompts: # required by Qwen MoE for "hidden" gate mode, otherwise not allowed | |
- "chat" | |
- "assistant" | |
# (optional, but recommended:) | |
residual_scale: 0.1 | |