Edit model card
base_model: rinna/llama-3-youko-8b
gate_mode: random # "hidden", "cheap_embed", "random" の中から選択
dtype: bfloat16
experts_per_token: 2 # 各トークンに対してアクティブなエキスパートの数
experts:
  - source_model: aixsatoshi/Llama-3-youko-8b-instruct-chatvector
  - source_model: elyza/Llama-3-ELYZA-JP-8B
Downloads last month
36
Safetensors
Model size
13.7B params
Tensor type
BF16
·
Inference API
Input a message to start chatting with aixsatoshi/Llama-3-Elyza-Youko-moe-2x8B.
Model is too large to load in Inference API (serverless). To try the model, launch it on Inference Endpoints (dedicated) instead.

Space using aixsatoshi/Llama-3-Elyza-Youko-moe-2x8B 1