From 2b3c49c9cb387883749459b75f89e8d070af8f2e Mon Sep 17 00:00:00 2001 From: jayeshthk Date: Thu, 19 Jun 2025 17:46:50 +0530 Subject: [PATCH] Add commented options for output attentions and hidden states in DeepSeekMathConfig --- train/model.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/train/model.py b/train/model.py index 15b2f1b..509ac75 100644 --- a/train/model.py +++ b/train/model.py @@ -21,6 +21,8 @@ class DeepSeekMathConfig: use_cache: bool = True rope_scaling: Optional[dict] = None tie_word_embeddings: bool = False + # output_attentions:bool=True + # output_hidden_states:int=12 class RMSNorm(nn.Module): def __init__(self, hidden_size, eps=1e-6):