From 214db9e660f99ee2d7ed63c5784b829bc59caf0d Mon Sep 17 00:00:00 2001
From: chengchengpei <5881383+chengchengpei@users.noreply.github.com>
Date: Mon, 23 Sep 2024 03:54:58 -0700
Subject: [PATCH] add back self.max_position_embeddings =
 config.max_position_embeddings (#33550)

* add back self.max_position_embeddings = config.max_position_embeddings

* fix-copies
---
 src/transformers/models/qwen2/modeling_qwen2.py         | 1 +
 src/transformers/models/qwen2_moe/modeling_qwen2_moe.py | 1 +
 2 files changed, 2 insertions(+)

diff --git a/src/transformers/models/qwen2/modeling_qwen2.py b/src/transformers/models/qwen2/modeling_qwen2.py
index aafecb95b6a..1e79115d347 100644
--- a/src/transformers/models/qwen2/modeling_qwen2.py
+++ b/src/transformers/models/qwen2/modeling_qwen2.py
@@ -310,6 +310,7 @@ class Qwen2Attention(nn.Module):
         self.head_dim = self.hidden_size // self.num_heads
         self.num_key_value_heads = config.num_key_value_heads
         self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.max_position_embeddings = config.max_position_embeddings
         self.rope_theta = config.rope_theta
         self.is_causal = True
         self.attention_dropout = config.attention_dropout
diff --git a/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py b/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
index bc06b406bf4..c9ee7b5f57a 100644
--- a/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
+++ b/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
@@ -388,6 +388,7 @@ class Qwen2MoeAttention(nn.Module):
         self.head_dim = self.hidden_size // self.num_heads
         self.num_key_value_heads = config.num_key_value_heads
         self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.max_position_embeddings = config.max_position_embeddings
         self.rope_theta = config.rope_theta
         self.is_causal = True
         self.attention_dropout = config.attention_dropout