[XPU] set appropriate mask value for xpu (#9495)

Author: runzhech

paddlenlp/transformers/llama/modeling.py

@@ -1546,8 +1546,9 @@ def _prepare_decoder_attention_mask(attention_mask, input_shape, past_key_values
             expanded_attn_mask = expanded_attn_mask.astype("float32")
             expanded_attn_mask = paddle.where(expanded_attn_mask, x, y).astype(dtype)
         elif get_env_device() in ["xpu", "gcu"]:
+            min_val = paddle.finfo(dtype).min if get_env_device() == "gcu" else -1e37  # mask value for xpu
             x = paddle.to_tensor(0.0, dtype=dtype)
-            y = paddle.to_tensor(paddle.finfo(dtype).min, dtype=dtype)
+            y = paddle.to_tensor(min_val, dtype=dtype)
             expanded_attn_mask = expanded_attn_mask.astype(dtype)
             expanded_attn_mask = paddle.where(expanded_attn_mask, x, y).astype(dtype)
         else: