From ed82181e09cc121e67d8dfe085c47e72575ad1cf Mon Sep 17 00:00:00 2001 From: Niklas Muennighoff Date: Tue, 20 Aug 2024 20:27:39 -0700 Subject: [PATCH] Fix typo Co-authored-by: Shane A --- olmo/model.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/olmo/model.py b/olmo/model.py index 44970b9d5..2b827cc34 100644 --- a/olmo/model.py +++ b/olmo/model.py @@ -670,7 +670,7 @@ def build(cls, layer_id: int, config: ModelConfig, cache: BufferCache) -> OLMoBl class OLMoEBlock(OLMoBlock): """ - This is a a transformer MoE block where the output is computed as ``MoE(LN(x + Attention(LN(x))))`` + This is a transformer MoE block where the output is computed as ``MoE(LN(x + Attention(LN(x))))`` (plus another skip connection). """