Update configuration_phi.py

configuration_phi.py

@@ -21,11 +21,11 @@ class PhiConfig(PretrainedConfig):
     def __init__(
         self,
         vocab_size: int = 50295,
-        n_positions: int = 2048,
+        n_positions: int = 1024,
         n_embd: int = 1024,
-        n_layer: int = 20,
+        n_layer: int = 24,
         n_inner: Optional[int] = None,
-        n_head: int = 16,
+        n_head: int = 32,
         n_head_kv: Optional[int] = None,
         rotary_dim: Optional[int] = 32,
         activation_function: Optional[str] = "gelu_new",
@@ -41,7 +41,7 @@ class PhiConfig(PretrainedConfig):
         pad_vocab_size_multiple: int = 64,
         **kwargs
     ) -> None:
-        self.vocab_size = int(math.ceil(vocab_size / pad_vocab_size_multiple) * pad_vocab_size_multiple)
+        self.vocab_size = vocab_size
         self.n_positions = n_positions
         self.n_embd = n_embd
         self.n_layer = n_layer