[docs] Fix FlashAttention link (#35171)

fix link

docs/source/en/model_doc/idefics2.md

@@ -141,7 +141,7 @@ Do note that when training Idefics2 on multi-turn conversations between a user a
 
 ## Model optimizations: Flash Attention
 
-The code snippets above showcase inference without any optimization tricks. However, one can drastically speed up the model by leveraging [Flash Attention](../perf_train_gpu_one.md#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
+The code snippets above showcase inference without any optimization tricks. However, one can drastically speed up the model by leveraging [Flash Attention](../perf_train_gpu_one#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
 
 First, make sure to install the latest version of Flash Attention 2 to include the sliding window attention feature.
 

docs/source/en/model_doc/llava_next_video.md

@@ -240,7 +240,7 @@ model = LlavaNextVideoForConditionalGeneration.from_pretrained("llava-hf/LLaVA-N
 
 ### Flash-Attention 2 to speed-up generation
 
-Additionally, we can greatly speed-up model inference by using [Flash Attention](../perf_train_gpu_one.md#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
+Additionally, we can greatly speed-up model inference by using [Flash Attention](../perf_train_gpu_one#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
 
 First, make sure to install the latest version of Flash Attention 2:
 

docs/source/en/model_doc/mistral.md

@@ -91,7 +91,7 @@ As can be seen, the instruction-tuned model requires a [chat template](../chat_t
 
 ## Speeding up Mistral by using Flash Attention
 
-The code snippets above showcase inference without any optimization tricks. However, one can drastically speed up the model by leveraging [Flash Attention](../perf_train_gpu_one.md#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
+The code snippets above showcase inference without any optimization tricks. However, one can drastically speed up the model by leveraging [Flash Attention](../perf_train_gpu_one#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
 
 First, make sure to install the latest version of Flash Attention 2 to include the sliding window attention feature.
 

docs/source/en/model_doc/mixtral.md

@@ -93,7 +93,7 @@ As can be seen, the instruction-tuned model requires a [chat template](../chat_t
 
 ## Speeding up Mixtral by using Flash Attention
 
-The code snippets above showcase inference without any optimization tricks. However, one can drastically speed up the model by leveraging [Flash Attention](../perf_train_gpu_one.md#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
+The code snippets above showcase inference without any optimization tricks. However, one can drastically speed up the model by leveraging [Flash Attention](../perf_train_gpu_one#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
 
 First, make sure to install the latest version of Flash Attention 2 to include the sliding window attention feature.
 

docs/source/en/model_doc/video_llava.md

@@ -174,7 +174,7 @@ model = VideoLlavaForConditionalGeneration.from_pretrained("LanguageBind/Video-L
 
 ### Flash-Attention 2 to speed-up generation
 
-Additionally, we can greatly speed-up model inference by using [Flash Attention](../perf_train_gpu_one.md#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
+Additionally, we can greatly speed-up model inference by using [Flash Attention](../perf_train_gpu_one#flash-attention-2), which is a faster implementation of the attention mechanism used inside the model.
 
 First, make sure to install the latest version of Flash Attention 2: