no max batch tokens

ServiceNow · Dec 22, 2024 · 210a933 · 210a933
1 parent 06876e2
commit 210a933
Showing 1 changed file with 0 additions and 1 deletion.
diff --git a/conf/rl_gsm8k.yaml b/conf/rl_gsm8k.yaml
@@ -41,7 +41,6 @@ vllm_config:
     --gpu-memory-utilization: 0.9
     # VLLM get log probs OOM https://github.com/vllm-project/vllm/issues/5907
     --enable-chunked-prefill: ""
-    --max-num-batched-tokens: 256
     --enable-prefix-caching: ""
   ref_vllm_kwargs:
     --download-dir: /mnt/llmd/base_models/