Fix: Mixtral docstring inconsistency for top_k_index/weights

Author: ShivanshTiwari1

src/transformers/models/mixtral/modular_mixtral.py

@@ -83,7 +83,9 @@ def load_balancing_loss_func(
 
     if isinstance(gate_logits, tuple):
         compute_device = gate_logits[0].device
-        concatenated_gate_logits = torch.cat([layer_gate.to(compute_device) for layer_gate in gate_logits], dim=0)
+        concatenated_gate_logits = torch.cat(
+            [layer_gate.to(compute_device) for layer_gate in gate_logits], dim=0
+        )
 
     routing_weights = torch.nn.functional.softmax(concatenated_gate_logits, dim=-1)
 
@@ -99,20 +101,24 @@ def load_balancing_loss_func(
         router_prob_per_expert = torch.mean(routing_weights, dim=0)
     else:
         batch_size, sequence_length = attention_mask.shape
-        num_hidden_layers = concatenated_gate_logits.shape[0] // (batch_size * sequence_length)
+        num_hidden_layers = concatenated_gate_logits.shape[0] // (
+            batch_size * sequence_length
+        )
 
         # Compute the mask that masks all padding tokens as 0 with the same shape of expert_mask
         expert_attention_mask = (
             attention_mask[None, :, :, None, None]
-            .expand((num_hidden_layers, batch_size, sequence_length, top_k, num_experts))
+            .expand(
+                (num_hidden_layers, batch_size, sequence_length, top_k, num_experts)
+            )
             .reshape(-1, top_k, num_experts)
             .to(compute_device)
         )
 
         # Compute the percentage of tokens routed to each experts
-        tokens_per_expert = torch.sum(expert_mask.float() * expert_attention_mask, dim=0) / torch.sum(
-            expert_attention_mask, dim=0
-        )
+        tokens_per_expert = torch.sum(
+            expert_mask.float() * expert_attention_mask, dim=0
+        ) / torch.sum(expert_attention_mask, dim=0)
 
         # Compute the mask that masks all padding tokens as 0 with the same shape of tokens_per_expert
         router_per_expert_attention_mask = (
@@ -123,9 +129,9 @@ def load_balancing_loss_func(
         )
 
         # Compute the average probability of routing to these experts
-        router_prob_per_expert = torch.sum(routing_weights * router_per_expert_attention_mask, dim=0) / torch.sum(
-            router_per_expert_attention_mask, dim=0
-        )
+        router_prob_per_expert = torch.sum(
+            routing_weights * router_per_expert_attention_mask, dim=0
+        ) / torch.sum(router_per_expert_attention_mask, dim=0)
 
     overall_loss = torch.sum(tokens_per_expert * router_prob_per_expert.unsqueeze(0))
     return overall_loss * num_experts
@@ -144,7 +150,9 @@ def __init__(self, config: MixtralConfig):
         self.act_fn = ACT2FN[config.hidden_act]
 
     def forward(self, hidden_states):
-        current_hidden_states = self.act_fn(self.w1(hidden_states)) * self.w3(hidden_states)
+        current_hidden_states = self.act_fn(self.w1(hidden_states)) * self.w3(
+            hidden_states
+        )
         current_hidden_states = self.w2(current_hidden_states)
         return current_hidden_states
 
@@ -162,25 +170,38 @@ def __init__(self, config: MixtralConfig):
             self.append(MixtralMLP(config))
 
     def forward(
-        self, hidden_states: torch.Tensor, top_k_index: torch.Tensor, top_k_weights: torch.Tensor
+        self,
+        hidden_states: torch.Tensor,
+        top_k_index: torch.Tensor,
+        top_k_weights: torch.Tensor,
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states: (batch_size * sequence_length, hidden_dim)
-            selected_experts: (batch_size * sequence_length, top_k)
-            routing_weights: (batch_size * sequence_length, top_k)
-        Returns:
-            (batch_size * sequence_length, hidden_dim)
+            hidden_states (`torch.Tensor`):
+                Input tensor.
+            top_k_index (`torch.Tensor`):
+                The indices of the top-k selected experts.
+            top_k_weights (`torch.Tensor`):
+                The weights corresponding to the top-k selected experts.
         """
+
         final_hidden_states = torch.zeros_like(hidden_states)
-        expert_mask = torch.nn.functional.one_hot(top_k_index, num_classes=self.num_experts).permute(2, 1, 0)
+        expert_mask = torch.nn.functional.one_hot(
+            top_k_index, num_classes=self.num_experts
+        ).permute(2, 1, 0)
 
         expert_hit = torch.greater(expert_mask.sum(dim=(-1, -2)), 0).nonzero()
         for expert_idx in expert_hit:
             idx, top_x = torch.where(expert_mask[expert_idx].squeeze(0))
-            current_state = hidden_states[None, top_x].reshape(-1, hidden_states.shape[-1])
-            current_hidden_states = self[expert_idx](current_state) * top_k_weights[top_x, idx, None]
-            final_hidden_states.index_add_(0, top_x, current_hidden_states.to(hidden_states.dtype))
+            current_state = hidden_states[None, top_x].reshape(
+                -1, hidden_states.shape[-1]
+            )
+            current_hidden_states = (
+                self[expert_idx](current_state) * top_k_weights[top_x, idx, None]
+            )
+            final_hidden_states.index_add_(
+                0, top_x, current_hidden_states.to(hidden_states.dtype)
+            )
         return final_hidden_states
 
 
@@ -201,11 +222,15 @@ def route_tokens_to_experts(self, router_logits):
     def forward(self, hidden_states: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
         batch_size, sequence_length, hidden_dim = hidden_states.shape
         if self.training and self.jitter_noise > 0:
-            hidden_states *= torch.empty_like(hidden_states).uniform_(1.0 - self.jitter_noise, 1.0 + self.jitter_noise)
+            hidden_states *= torch.empty_like(hidden_states).uniform_(
+                1.0 - self.jitter_noise, 1.0 + self.jitter_noise
+            )
         hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
         router_logits = self.gate(hidden_states)
         top_k_index, top_k_weights = self.route_tokens_to_experts(router_logits)
-        hidden_states = self.experts(hidden_states, top_k_index, top_k_weights.to(hidden_states.dtype))
+        hidden_states = self.experts(
+            hidden_states, top_k_index, top_k_weights.to(hidden_states.dtype)
+        )
         hidden_states = hidden_states.reshape(batch_size, sequence_length, hidden_dim)
         return hidden_states
 
@@ -230,8 +255,12 @@ def __init__(self, config: MixtralConfig, layer_idx: int):
         self.self_attn = MixtralAttention(config, layer_idx)
 
         self.block_sparse_moe = MixtralSparseMoeBlock(config)
-        self.input_layernorm = MixtralRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
-        self.post_attention_layernorm = MixtralRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.input_layernorm = MixtralRMSNorm(
+            config.hidden_size, eps=config.rms_norm_eps
+        )
+        self.post_attention_layernorm = MixtralRMSNorm(
+            config.hidden_size, eps=config.rms_norm_eps
+        )
 
     def forward(
         self,
@@ -265,7 +294,9 @@ def forward(
 class MixtralPreTrainedModel(MistralPreTrainedModel):
     _can_compile_fullgraph = False  # MoE models don't work with torch.compile (`torch.where(condition)` not supported)
     _can_record_outputs = {
-        "router_logits": OutputRecorder(nn.Linear, layer_name="block_sparse_moe.gate", index=0),
+        "router_logits": OutputRecorder(
+            nn.Linear, layer_name="block_sparse_moe.gate", index=0
+        ),
         "hidden_states": MixtralDecoderLayer,
         "attentions": MixtralAttention,
     }
@@ -284,7 +315,9 @@ def forward(
         **kwargs: Unpack[TransformersKwargs],
     ) -> MoeModelOutputWithPast:
         if (input_ids is None) ^ (inputs_embeds is not None):
-            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+            raise ValueError(
+                "You must specify exactly one of input_ids or inputs_embeds"
+            )
 
         if use_cache and past_key_values is None:
             past_key_values = DynamicCache(config=self.config)
@@ -293,14 +326,22 @@ def forward(
             inputs_embeds = self.embed_tokens(input_ids)
 
         if cache_position is None:
-            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            past_seen_tokens = (
+                past_key_values.get_seq_length() if past_key_values is not None else 0
+            )
             cache_position = torch.arange(
-                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+                past_seen_tokens,
+                past_seen_tokens + inputs_embeds.shape[1],
+                device=inputs_embeds.device,
             )
         if position_ids is None:
             position_ids = cache_position.unsqueeze(0)
 
-        mask_function = create_causal_mask if self.config.sliding_window is None else create_sliding_window_causal_mask
+        mask_function = (
+            create_causal_mask
+            if self.config.sliding_window is None
+            else create_sliding_window_causal_mask
+        )
         causal_mask = mask_function(
             config=self.config,
             input_embeds=inputs_embeds,
@@ -381,7 +422,9 @@ def forward(
         ```"""
 
         output_router_logits = (
-            output_router_logits if output_router_logits is not None else self.config.output_router_logits
+            output_router_logits
+            if output_router_logits is not None
+            else self.config.output_router_logits
         )
 
         # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
@@ -399,7 +442,11 @@ def forward(
 
         hidden_states = outputs.last_hidden_state
         # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
-        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
+        slice_indices = (
+            slice(-logits_to_keep, None)
+            if isinstance(logits_to_keep, int)
+            else logits_to_keep
+        )
         logits = self.lm_head(hidden_states[:, slice_indices, :])
 
         loss = None
@@ -415,7 +462,9 @@ def forward(
                 attention_mask,
             )
             if labels is not None:
-                loss += self.router_aux_loss_coef * aux_loss.to(loss.device)  # make sure to reside in the same device
+                loss += self.router_aux_loss_coef * aux_loss.to(
+                    loss.device
+                )  # make sure to reside in the same device
 
         return MoeCausalLMOutputWithPast(
             loss=loss,