diff --git a/src/axolotl/common/architectures.py b/src/axolotl/common/architectures.py index 827a63c07..00046ca4b 100644 --- a/src/axolotl/common/architectures.py +++ b/src/axolotl/common/architectures.py @@ -10,6 +10,7 @@ MOE_ARCH_BLOCK = { "JetMoeMoE", ], "mixtral": "MixtralSparseMoeBlock", + "phimoe": "PhiMoESparseMoeBlock", "qwen2_moe": "Qwen2MoeSparseMoeBlock", "deepseek_v2": "DeepseekV2MoE", } diff --git a/src/axolotl/monkeypatch/multipack.py b/src/axolotl/monkeypatch/multipack.py index 3ee89d2e5..265f08818 100644 --- a/src/axolotl/monkeypatch/multipack.py +++ b/src/axolotl/monkeypatch/multipack.py @@ -20,6 +20,7 @@ SUPPORTED_MULTIPACK_MODEL_TYPES = [ "falcon", "phi", "phi3", + "phimoe", "gemma", "gemma2", "gemmoe",