fix: merge benchmarks

Signed-off-by: Mehant Kammakomati <mehant.kammakomati2@ibm.com>

Author: kmehant

scripts/benchmarks/scenarios-moe.yaml

@@ -38,30 +38,6 @@
 scenarios:
     -   name: accelerated-moe-full
         framework_config: 
-            - # without acceleration
-            - moe-scattermoe-granite-ep1
-            - moe-scattermoe-granite-ep2
-            # - moe-scattermoe-granite-ep4
-            # - moe-scattermoe-granite-ep1-padding-free
-            # - moe-scattermoe-granite-ep1-padding-free-foak
-            # - moe-scattermoe-granite-ep2-padding-free
-            # - moe-scattermoe-granite-ep2-padding-free-foak
-            # - moe-scattermoe-granite-ep4-padding-free
-            # - moe-scattermoe-granite-ep4-padding-free-foak
-        arguments:
-            learning_rate: 5e-5
-            torch_dtype: bfloat16
-            gradient_accumulation_steps: null
-            per_device_train_batch_size: 8
-            logging_steps: 1
-            packing: False
-            adam_epsilon: 1e-8
-            model_name_or_path: 
-                - 'ibm-granite/granite-3.0-3b-a800m-instruct'
-                - 'ibm-research/moe-7b-1b-active-shared-experts'
-
-    -   name: accelerated-moe-granite-4
-        framework_config:
             - # without acceleration
             - moe-scattermoe-granite-ep1
             - moe-scattermoe-granite-ep2
@@ -81,6 +57,8 @@ scenarios:
             packing: False
             adam_epsilon: 1e-8
             model_name_or_path: 
+                - 'ibm-granite/granite-3.0-3b-a800m-instruct'
+                - 'ibm-research/moe-7b-1b-active-shared-experts'
                 - 'ibm-granite/granite-4.0-tiny-preview'
 
 
@@ -100,4 +78,4 @@ scenarios:
             packing: False
             adam_epsilon: 1e-8
             model_name_or_path: 
-                - 'mistralai/Mixtral-8x7B-Instruct-v0.1'
+                - 'mistralai/Mixtral-8x7B-Instruct-v0.1'