set fp8 option as true

anandhu-eng · Jul 8, 2024 · 0247b0a · 0247b0a
1 parent 0b80cbc
commit 0247b0a
Showing 1 changed file with 4 additions and 0 deletions.
diff --git a/script/app-mlperf-inference-nvidia/_cm.yaml b/script/app-mlperf-inference-nvidia/_cm.yaml
@@ -1025,6 +1025,10 @@ variations:
     default_variations:
       batch-size: batch_size.1
 
+  llama2-70b_,run_harness:
+    env:
+      CM_MLPERF_NVIDIA_HARNESS_USE_FP8: 'True'
+
   gptj_,run_harness:
     deps:
     - tags: install,pytorch,from.src