Bump to v2.2.3

flash_attn/__init__.py

-__version__ = "2.2.2"
+__version__ = "2.2.3"
 
 from flash_attn.flash_attn_interface import (
     flash_attn_func,

training/Dockerfile

@@ -85,14 +85,13 @@ RUN pip install transformers==4.25.1 datasets==2.8.0 pytorch-lightning==1.8.6 tr
 RUN pip install git+https://github.com/mlcommons/logging.git@2.1.0
 
 # Install FlashAttention
-RUN pip install flash-attn==2.2.2
+RUN pip install flash-attn==2.2.3
 
 # Install CUDA extensions for cross-entropy, fused dense, layer norm
 RUN git clone https://github.com/HazyResearch/flash-attention \
-    && cd flash-attention && git checkout v2.2.2 \
+    && cd flash-attention && git checkout v2.2.3 \
     && cd csrc/fused_softmax && pip install . && cd ../../ \
     && cd csrc/rotary && pip install . && cd ../../ \
-    && cd csrc/xentropy && pip install . && cd ../../ \
     && cd csrc/layer_norm && pip install . && cd ../../ \
     && cd csrc/fused_dense_lib && pip install . && cd ../../ \
     && cd csrc/ft_attention && pip install . && cd ../../ \