Upload mergekit_moe_config.yml with huggingface_hub
Browse files- mergekit_moe_config.yml +110 -0
mergekit_moe_config.yml
ADDED
|
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# base_model: examples/microsoft_Phi-4-mini-instruct
|
| 2 |
+
# dtype: float16
|
| 3 |
+
# gate_mode: hidden
|
| 4 |
+
# experts:
|
| 5 |
+
# - source_model: examples/microsoft_Phi-4-mini-instruct
|
| 6 |
+
# positive_prompts:
|
| 7 |
+
# - "chat"
|
| 8 |
+
# - "assistant"
|
| 9 |
+
# - "tell me"
|
| 10 |
+
# - "explain"
|
| 11 |
+
# - "I want"
|
| 12 |
+
# - "storywriting"
|
| 13 |
+
# - "write"
|
| 14 |
+
# - "scene"
|
| 15 |
+
# - "story"
|
| 16 |
+
# - "character"
|
| 17 |
+
# - "Hello"
|
| 18 |
+
# - "explain"
|
| 19 |
+
# - "mermaid"
|
| 20 |
+
# - "graph TB;"
|
| 21 |
+
# - "sequenceDiagram"
|
| 22 |
+
# - source_model: examples/microsoft_Phi-4-mini-instruct
|
| 23 |
+
# positive_prompts:
|
| 24 |
+
# - "code"
|
| 25 |
+
# - "python"
|
| 26 |
+
# - "javascript"
|
| 27 |
+
# - "programming"
|
| 28 |
+
# - "algorithm"
|
| 29 |
+
# - "reason"
|
| 30 |
+
# - "math"
|
| 31 |
+
# - "mathematics"
|
| 32 |
+
# - "solve"
|
| 33 |
+
# - "count"
|
| 34 |
+
|
| 35 |
+
base_model: examples/TroyDoesAI_BlackSheep-24B
|
| 36 |
+
dtype: float16
|
| 37 |
+
gate_mode: hidden
|
| 38 |
+
experts:
|
| 39 |
+
- source_model: examples/TroyDoesAI_BlackSheep-24B
|
| 40 |
+
positive_prompts:
|
| 41 |
+
- "chat"
|
| 42 |
+
- "assistant"
|
| 43 |
+
- "tell me"
|
| 44 |
+
- "explain"
|
| 45 |
+
- "I want"
|
| 46 |
+
- "Hello"
|
| 47 |
+
- "explain"
|
| 48 |
+
- "mermaid"
|
| 49 |
+
- "graph TB;"
|
| 50 |
+
- "sequenceDiagram"
|
| 51 |
+
- source_model: examples/huihui-ai_Mistral-Small-24B-Instruct-2501-abliterated
|
| 52 |
+
positive_prompts:
|
| 53 |
+
- "code"
|
| 54 |
+
- "python"
|
| 55 |
+
- "javascript"
|
| 56 |
+
- "programming"
|
| 57 |
+
- "algorithm"
|
| 58 |
+
- source_model: examples/TroyDoesAI_BlackSheep-24B
|
| 59 |
+
positive_prompts:
|
| 60 |
+
- "storywriting"
|
| 61 |
+
- "write"
|
| 62 |
+
- "scene"
|
| 63 |
+
- "story"
|
| 64 |
+
- "character"
|
| 65 |
+
- "Hero"
|
| 66 |
+
- "Villian"
|
| 67 |
+
- "darkness"
|
| 68 |
+
- "violence"
|
| 69 |
+
- "crime"
|
| 70 |
+
- "conflict"
|
| 71 |
+
- "roleplay"
|
| 72 |
+
- "adventure"
|
| 73 |
+
- "ai"
|
| 74 |
+
- "dungeon"
|
| 75 |
+
- "brutal"
|
| 76 |
+
- "chat"
|
| 77 |
+
- "faction"
|
| 78 |
+
- "location"
|
| 79 |
+
- "emotion"
|
| 80 |
+
- "plot"
|
| 81 |
+
- "armor"
|
| 82 |
+
- "bad"
|
| 83 |
+
- "evil"
|
| 84 |
+
- "blacksheep"
|
| 85 |
+
- "<|BlackSheep|>"
|
| 86 |
+
- source_model: examples/huihui-ai_Mistral-Small-24B-Instruct-2501-abliterated
|
| 87 |
+
positive_prompts:
|
| 88 |
+
- "reason"
|
| 89 |
+
- "math"
|
| 90 |
+
- "mathematics"
|
| 91 |
+
- "solve"
|
| 92 |
+
- "count"
|
| 93 |
+
- "think"
|
| 94 |
+
- "plan"
|
| 95 |
+
- "puzzle"
|
| 96 |
+
- "predict"
|
| 97 |
+
|
| 98 |
+
|
| 99 |
+
|
| 100 |
+
# models:
|
| 101 |
+
# - model: Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2+kloodia/lora-8b-bio
|
| 102 |
+
# - model: arcee-ai/Llama-3.1-SuperNova-Lite+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
|
| 103 |
+
# - model: mlabonne/Hermes-3-Llama-3.1-8B-lorablated+kloodia/lora-8b-physic
|
| 104 |
+
# - model: aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored+kloodia/lora-8b-medic
|
| 105 |
+
# - model: ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.1+Blackroot/Llama-3-8B-Abomination-LORA
|
| 106 |
+
# merge_method: model_stock
|
| 107 |
+
# base_model: Sao10K/L3-8B-Stheno-v3.2+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
|
| 108 |
+
# normalize: false
|
| 109 |
+
# int8_mask: true
|
| 110 |
+
# dtype: bfloat16
|