Update benchmarks.yml

Author: juliagmt-google

.github/workflows/benchmarks.yml

@@ -69,62 +69,7 @@ jobs:
       - name: Configure XLA
         working-directory: xla
         run: ./configure.py --backend CUDA --nccl
-      
-      - name: Set TF_CPP_MAX_VLOG_LEVEL
-        working-directory: xla
-        run: export TF_CPP_MAX_VLOG_LEVEL=1
-        
-      - name: Check TF_CPP_MAX_VLOG_LEVEL
-        working-directory: xla
-        run: echo "$TF_CPP_MAX_VLOG_LEVEL"
-
-      - name: Build hlo_runner_main
-        working-directory: xla
-        run: bazel build -c opt --config=cuda --dynamic_mode=off //xla/tools/multihost_hlo_runner:hlo_runner_main
-      name: Benchmarks
-
-on:
-  pull_request:
-    branches:
-      - main
-  workflow_dispatch:
-    inputs:
-      halt-for-connection:
-        description: 'Should this workflow run wait for a remote connection?'
-        type: choice
-        required: true
-        default: 'no'
-        options:
-        - 'yes'
-        - 'no'
 
-jobs:
-  build-xla-gpu-and-test:
-    runs-on: linux-x86-g2-48-l4-4gpu
-    container:
-      image: "gcr.io/tensorflow-testing/nosla-cuda12.3-cudnn9.1-ubuntu20.04-manylinux2014-multipython:latest"
-      options: --gpus all --privileged 
-
-    steps:
-      - name: Checkout XLA
-        uses: actions/checkout@v3
-        with:
-          repository: openxla/xla 
-          path: xla
-
-      - name: Create results directory
-        working-directory: xla
-        run: mkdir -p results
-
-      - name: Get GPU spec
-        working-directory: xla
-        continue-on-error: true
-        run: nvidia-smi
-
-      - name: Configure XLA
-        working-directory: xla
-        run: ./configure.py --backend CUDA --nccl
-      
       - name: Set TF_CPP_MAX_VLOG_LEVEL
         working-directory: xla
         run: echo "TF_CPP_MAX_VLOG_LEVEL=1" >> $GITHUB_ENV  # Use GITHUB_ENV to persist across steps