langmoe

Author: haeggee

src/nanotron/config/models_config.py

@@ -262,4 +262,97 @@ def hidden_act(self):
         return self.activation_function
 
 
-NanotronConfigs = LlamaConfig | Starcoder2Config | GPT3Config | GPT3MoEConfig
+@dataclass
+class GPT3LangMoEConfig:
+    """Configuration for a GPT3 __MoE__ model with language aware gating"""
+
+    activation_function: str = "gelu"
+    attn_pdrop: float = 0.1
+    embd_pdrop: float = 0.1
+    eos_token_id: int = 49152
+    hidden_size: int = 2048
+    intermediate_size: Optional[int] = None
+    layer_norm_epsilon: float = 1e-05
+    max_position_embeddings: int = 4096
+    num_attention_heads: int = 16
+    num_hidden_layers: int = 24
+    resid_pdrop: float = 0.1
+    scale_attention_softmax_in_fp32: bool = True
+    scale_attn_weights: bool = True
+    vocab_size: int = 49280
+    sinusoidal_position_embedding: bool = True
+    position_embedding_offset: int = 2
+    use_spda: bool = False
+    act_pdrop: float = 0.0
+    scale_embedding: bool = True
+    # MoE specific
+    is_moe: bool = True
+    moe_num_experts: int = 1
+    num_experts_per_tok: int = 1
+    moe_loss_weight: float = 0.01
+    moe_z_loss_weight: float = 0.001
+    moe_glu: bool = False
+
+    # Language aware gating
+    num_languages: int = 100
+    language_embedding_size: int = 128
+
+    def as_gpt3(self) -> GPT3Config:
+        config = dict(**vars(self))
+
+        # Moe
+        del config["is_moe"]
+        del config["moe_num_experts"]
+        del config["num_experts_per_tok"]
+        del config["moe_loss_weight"]
+        del config["moe_z_loss_weight"]
+        del config["moe_glu"]
+
+        # language aware gating
+        del config["num_languages"]
+        del config["language_embedding_size"]
+
+        if "_is_using_mup" in config:
+            del config["_is_using_mup"]
+        return GPT3Config(**config)
+
+    def as_starcoder2(self) -> Starcoder2Config:
+        # same as gpt3 conversion above
+        config = dict(**vars(self))
+        del config["sinusoidal_position_embedding"]
+        del config["use_spda"]
+        del config["position_embedding_offset"]
+        del config["act_pdrop"]
+        del config["scale_embedding"]
+
+        # Moe
+        del config["is_moe"]
+        del config["moe_num_experts"]
+        del config["num_experts_per_tok"]
+        del config["moe_loss_weight"]
+        del config["moe_z_loss_weight"]
+        del config["moe_glu"]
+
+        # language aware gating
+        del config["num_languages"]
+        del config["language_embedding_size"]
+
+        if "_is_using_mup" in config:
+            del config["_is_using_mup"]
+        return Starcoder2Config(
+            grouped_query=True,
+            num_kv_heads=self.num_attention_heads,
+            use_rotary_embeddings=False,
+            **config,
+        )
+
+    @property
+    def n_inner(self):
+        return self.intermediate_size
+
+    @property
+    def hidden_act(self):
+        return self.activation_function
+
+
+NanotronConfigs = LlamaConfig | Starcoder2Config | GPT3Config | GPT3MoEConfig | GPT3LangMoEConfig