Replace slashes with double dashes

pegasus/nlp-eval.yaml

@@ -1,5 +1,5 @@
 - command:
-    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-causal-experimental --model_args pretrained={{model}},trust_remote_code=True,use_accelerate=True --tasks arc_challenge --num_fewshot 25
+    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-causal-experimental --model_args pretrained={{ model }},trust_remote_code=True,use_accelerate=True --tasks arc_challenge --num_fewshot 25 --output_path /data/leaderboard/benchmark/nlp/{{ replace model "/" "--" }}.json
   model:
     - /data/leaderboard/weights/metaai/llama-7B
     - /data/leaderboard/weights/metaai/llama-13B
@@ -21,7 +21,7 @@
     - togethercomputer/RedPajama-INCITE-7B-Chat
 
 - command:
-    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-causal-experimental --model_args pretrained={{model}},trust_remote_code=True,use_accelerate=True --tasks hellaswag --num_fewshot 10
+    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-causal-experimental --model_args pretrained={{ model }},trust_remote_code=True,use_accelerate=True --tasks hellaswag --num_fewshot 10 --output_path /data/leaderboard/benchmark/nlp/{{ replace model "/" "--" }}.json
   model:
     - /data/leaderboard/weights/metaai/llama-7B
     - /data/leaderboard/weights/metaai/llama-13B
@@ -43,7 +43,7 @@
     - togethercomputer/RedPajama-INCITE-7B-Chat
 
 - command:
-    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-causal-experimental --model_args pretrained={{model}},trust_remote_code=True,use_accelerate=True --tasks truthfulqa_mc --num_fewshot 0
+    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-causal-experimental --model_args pretrained={{ model }},trust_remote_code=True,use_accelerate=True --tasks truthfulqa_mc --num_fewshot 0 --output_path /data/leaderboard/benchmark/nlp/{{ replace model "/" "--" }}.json
   model:
     - /data/leaderboard/weights/metaai/llama-7B
     - /data/leaderboard/weights/metaai/llama-13B
@@ -65,6 +65,8 @@
     - togethercomputer/RedPajama-INCITE-7B-Chat
 
 - command:
-    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-seq2seq --model_args pretrained=lmsys/fastchat-t5-3b-v1.0,trust_remote_code=True,use_accelerate=True --tasks arc_challenge --num_fewshot 25
-    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-seq2seq --model_args pretrained=lmsys/fastchat-t5-3b-v1.0,trust_remote_code=True,use_accelerate=True --tasks hellaswag --num_fewshot 10
-    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-seq2seq --model_args pretrained=lmsys/fastchat-t5-3b-v1.0,trust_remote_code=True,use_accelerate=True --tasks truthfulqa_mc --num_fewshot 0
+    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-seq2seq --model_args pretrained={{ model }},trust_remote_code=True,use_accelerate=True --tasks arc_challenge --num_fewshot 25 --output_path /data/leaderboard/benchmark/nlp/{{ replace model "/" "--" }}.json
+    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-seq2seq --model_args pretrained={{ model }},trust_remote_code=True,use_accelerate=True --tasks hellaswag --num_fewshot 10 --output_path /data/leaderboard/benchmark/nlp/{{ replace model "/" "--" }}.json
+    - docker exec leaderboard{{ gpu }} python lm-evaluation-harness/main.py --device cuda --no_cache --model hf-seq2seq --model_args pretrained={{ model }},trust_remote_code=True,use_accelerate=True --tasks truthfulqa_mc --num_fewshot 0 --output_path /data/leaderboard/benchmark/nlp/{{ replace model "/" "--" }}.json
+  model:
+    - lmsys/fastchat-t5-3b-v1.0