add support for num_generations

src/axolotl/core/trainers/grpo/__init__.py

@@ -34,6 +34,8 @@ class GRPOStrategy:
                 grpo_args_kwargs[
                     "vllm_gpu_memory_utilization"
                 ] = cfg.grpo_vllm_gpu_memory_utilization
+        if cfg.grpo_num_generations:
+            grpo_args_kwargs["num_generations"] = cfg.grpo_num_generations
         return grpo_args_kwargs
 
     @classmethod

src/axolotl/utils/config/models/input/v0_4_1/grpo.py

@@ -15,3 +15,4 @@ class GRPOConfig(BaseModel):
     grpo_vllm_device: Optional[str] = "auto"
     grpo_vllm_gpu_memory_utilization: Optional[float] = 0.9
     grpo_reward_funcs: Optional[List[str]] = None
+    grpo_num_generations: Optional[int] = None