Mixtral: More correct MoE, lower loss (#932)

* More correct MoE

* Fix formatting

src/axolotl/models/mixtral/modeling_moe_mistral.py

@@ -215,23 +215,22 @@ class MoE(nn.Module):
     ):
         super().__init__()
         self.config = config
-        num_experts = config.num_experts
-        self.experts = nn.ModuleList([FeedForward(config) for i in range(num_experts)])
-        self.gate = nn.Linear(config.hidden_size, num_experts, bias=False)
-        self.num_experts_per_token = config.num_experts_per_token
+        self.gate = nn.Linear(config.hidden_size, config.num_experts, bias=False)
+        self.experts = nn.ModuleList(
+            [FeedForward(config) for i in range(config.num_experts)]
+        )
 
     def forward(self, x):
         orig_shape = x.shape
         x = x.view(-1, x.shape[-1])
 
-        scores = self.gate(x)
+        scores = self.gate(x).softmax(dim=-1)
         expert_weights, expert_indices = torch.topk(
-            scores, self.num_experts_per_token, dim=-1
+            scores, self.config.num_experts_per_token, dim=-1
         )
-        expert_weights = expert_weights.softmax(dim=-1)
         flat_expert_indices = expert_indices.view(-1)
 
-        x = x.repeat_interleave(self.num_experts_per_token, dim=0)
+        x = x.repeat_interleave(self.config.num_experts_per_token, dim=0)
         y = torch.empty_like(x)
         for i, expert in enumerate(self.experts):
             y[flat_expert_indices == i] = expert(x[flat_expert_indices == i])