Update configuration_internlm.py

configuration_internlm.py

@@ -19,9 +19,8 @@
 # limitations under the License.
 """ InternLM model configuration"""
 
-from transformers.utils import logging
 from transformers.configuration_utils import PretrainedConfig
-
+from transformers.utils import logging
 
 logger = logging.get_logger(__name__)
 
@@ -30,14 +29,11 @@ INTERNLM_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
 
 class InternLMConfig(PretrainedConfig):
     r"""
-    This is the configuration class to store the configuration of a [`InternLMModel`]. It is used to instantiate an InternLM
-    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
-    defaults will yield a similar configuration to that of the InternLM-7B.
-
+    This is the configuration class to store the configuration of a [`InternLMModel`]. It is used to instantiate
+    an InternLM model according to the specified arguments, defining the model architecture. Instantiating a
+    configuration with the defaults will yield a similar configuration to that of the InternLM-7B.
     Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
     documentation from [`PretrainedConfig`] for more information.
-
-
     Args:
         vocab_size (`int`, *optional*, defaults to 32000):
             Vocabulary size of the InternLM model. Defines the number of different tokens that can be represented by the
@@ -50,19 +46,6 @@ class InternLMConfig(PretrainedConfig):
             Number of hidden layers in the Transformer encoder.
         num_attention_heads (`int`, *optional*, defaults to 32):
             Number of attention heads for each attention layer in the Transformer encoder.
-        num_key_value_heads (`int`, *optional*):
-            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
-            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
-            `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
-            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
-            by meanpooling all the original heads within that group. For more details checkout [this
-            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
-            `num_attention_heads`.
-        pretraining_tp (`int`, *optional*, defaults to `1`):
-            Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
-            document](https://huggingface.co/docs/transformers/parallelism) to understand more about it. This value is
-            necessary to ensure exact reproducibility of the pretraining results. Please refer to [this
-            issue](https://github.com/pytorch/pytorch/issues/76232).
         hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
             The non-linear activation function (function or string) in the decoder.
         max_position_embeddings (`int`, *optional*, defaults to 2048):
@@ -78,30 +61,25 @@ class InternLMConfig(PretrainedConfig):
         tie_word_embeddings(`bool`, *optional*, defaults to `False`):
             Whether to tie weight embeddings
         Example:
-
     ```python
     >>> from transformers import InternLMModel, InternLMConfig
-
     >>> # Initializing a InternLM internlm-7b style configuration
     >>> configuration = InternLMConfig()
-
     >>> # Initializing a model from the internlm-7b style configuration
     >>> model = InternLMModel(configuration)
-
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
     model_type = "internlm"
     _auto_class = "AutoConfig"
 
-    def __init__(
+    def __init__(  # pylint: disable=W0102
         self,
         vocab_size=103168,
         hidden_size=4096,
         intermediate_size=11008,
         num_hidden_layers=32,
         num_attention_heads=32,
-        num_key_value_heads=None,
         hidden_act="silu",
         max_position_embeddings=2048,
         initializer_range=0.02,
@@ -112,6 +90,7 @@ class InternLMConfig(PretrainedConfig):
         eos_token_id=2,
         tie_word_embeddings=False,
         bias=True,
+        rotary={"base": 10000, "type": "dynamic"},  # pylint: disable=W0102
         **kwargs,
     ):
         self.vocab_size = vocab_size
@@ -120,20 +99,16 @@ class InternLMConfig(PretrainedConfig):
         self.intermediate_size = intermediate_size
         self.num_hidden_layers = num_hidden_layers
         self.num_attention_heads = num_attention_heads
-    
-        if num_key_value_heads is None:
-            num_key_value_heads = num_attention_heads
-        self.num_key_value_heads = num_key_value_heads
-
         self.hidden_act = hidden_act
         self.initializer_range = initializer_range
         self.rms_norm_eps = rms_norm_eps
         self.use_cache = use_cache
         self.bias = bias
+        self.rotary = rotary
         super().__init__(
             pad_token_id=pad_token_id,
             bos_token_id=bos_token_id,
             eos_token_id=eos_token_id,
             tie_word_embeddings=tie_word_embeddings,
             **kwargs,
-        )
+        )