upgrade trl to 0.24.0 and liger to 0.6.3 (#3230)

* upgrade trl to 0.24.0

* fix reward collator init

* use newer DataCollatorForPreference instead

* DataCollatorForPreference doesn't use padding kwarg

* fix input id labels

* fix fbgemm-gpu version for pytorch versions

* tweak pinned deps

* transformers doesn't support hub 1.0 yet

* upgrade liger dep to 0.6.3

* set TORCH_CUDA_ARCH_LIST correctly

requirements.txt

@@ -5,27 +5,27 @@ bitsandbytes==0.47.0
 triton>=3.0.0
 mamba-ssm==1.2.0.post1
 xformers>=0.0.23.post1
-liger-kernel==0.6.1
+liger-kernel==0.6.3
 # END section
 
 packaging==23.2
 
-huggingface_hub>=0.33.0
+huggingface_hub>=0.36.0
 peft>=0.17.1
 tokenizers>=0.21.1
 transformers==4.57.1
 accelerate==1.10.1
 datasets==4.0.0
 deepspeed>=0.17.0
-trl==0.23.1
-hf_xet==1.1.5
-kernels==0.9.0
+trl==0.24.0
+hf_xet==1.2.0
+kernels>=0.9.0
 trackio
 
 optimum==1.16.2
 hf_transfer
 sentencepiece
-gradio==5.41.1
+gradio==5.49.1
 
 modal==1.0.2
 pydantic==2.10.6