Update 2024-08-07-flexattention.md (#1707)

* Update 2024-08-07-flexattention.md

* Update 2024-08-07-flexattention.md

---------

Co-authored-by: Chris Abraham <cjyabraham@gmail.com>

_posts/2024-08-07-flexattention.md

@@ -1,7 +1,7 @@
 ---
 layout: blog_detail
 title: "FlexAttention: The Flexibility of PyTorch with the Performance of FlashAttention"
-author: "Team PyTorch: Horace He, Driss Guessous, Yanbo Liang, Joy Dong"
+author: "Team PyTorch: Driss Guessous, Yanbo Liang, Joy Dong, Horace He"
 ---
 
 ![a cartoon chart flexing his muscles](/assets/images/flexattention/fg1.jpg){:style="width:100%"}
@@ -131,7 +131,7 @@ Alibi is similar to relative positional encodings with one exception \- it has a
 alibi_bias = generate_alibi_bias() # [num_heads]
 
 def alibi(score, b, h, q_idx, kv_idx):
-    bias = alibi_bias[h] * (q_idx - kv_idx)
+    bias = alibi_bias[h] * (kv_idx - q_idx)
     return score + bias
 ```
 
@@ -218,12 +218,12 @@ def sliding_window_causal(b, h, q_idx, kv_idx):
     return causal_mask & window_mask
 
 # If you want to be cute...
-from torch.nn.attention import or_masks
+from torch.nn.attention import and_masks
 
 def sliding_window(b, h, q_idx, kv_idx)
     return q_idx - kv_idx <= SLIDING_WINDOW
 
-sliding_window_causal = or_masks(causal_mask, sliding_window)
+sliding_window_causal = and_masks(causal_mask, sliding_window)
 ```
 
 We benchmark it against `F.scaled_dot_product_attention` with a sliding window mask as well as FA2 with a causal mask (as a reference point for performance). Not only are we significantly faster than `F.scaled_dot_product_attention`, we’re *also* significantly faster than FA2 with a causal mask as this mask has significantly more sparsity.
@@ -479,4 +479,4 @@ We want to highlight some prior work (and people) that have inspired FlexAttenti
 - The Jax team's work on SplashAttention
 - Philippe Tillet and Keren Zhou for helping us with Triton 
 - Ali Hassani for discussions on neighborhood attention
-- Everybody who's complained about attention kernels not supporting their favorite attention variant :)
+- Everybody who's complained about attention kernels not supporting their favorite attention variant :)