instruct model setup

unsloth_compiled_cache/UnslothDDPOTrainer.py

@@ -1,7 +1,7 @@
 """
-2025.8.4
-2025.8.5
-4.55.1
+2025.8.9
+2025.8.10
+4.55.4
 0.21.0
 __UNSLOTH_VERSIONING__
 """
@@ -152,9 +152,10 @@ class UnslothDDPOConfig(DDPOConfig):
         default = -1,
         metadata = {'help': 'Chunk size to reduce memory usage. -1 is most efficient.'},
     )
+    
     def __init__(
         self,
-        exp_name = 'inference',
+        exp_name = 'train',
         run_name = '',
         seed = 3407,
         log_with = None,
@@ -194,6 +195,7 @@ class UnslothDDPOConfig(DDPOConfig):
         push_to_hub = False,
         vllm_sampling_params = None,
         unsloth_num_chunks = -1,
+        
         **kwargs,
     ):
         
@@ -238,6 +240,7 @@ class UnslothDDPOConfig(DDPOConfig):
             push_to_hub = push_to_hub,**kwargs)
         self.vllm_sampling_params = vllm_sampling_params
         self.unsloth_num_chunks = unsloth_num_chunks
+        
 pass
 
 class _UnslothDDPOTrainer(PyTorchModelHubMixin):
@@ -899,3 +902,13 @@ class UnslothDDPOTrainer(_UnslothDDPOTrainer):
             image_samples_hook = image_samples_hook,**kwargs)
         
 pass
+
+
+if hasattr(logger, "addFilter"):
+    import logging
+    class HideLoggingMessage(logging.Filter):
+        def __init__(self, text): self.text = text
+        def filter(self, x): return not (self.text in x.getMessage())
+    pass
+    logger.addFilter(HideLoggingMessage("`use_cache=True`"))
+