[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

applications/ColossalChat/examples/README.md

@@ -761,8 +761,8 @@ For SFT, we recommend using zero2 or zero2-cpu for 7B model and tp is your model
   - zero2_cpu, micro batch size=8, VRAM Usage=19412.77 MB
   - zero2, micro batch size=8, VRAM Usage=43446.31 MB
   - zero2, micro batch size=16, VRAM Usage=58082.30 MB
-  - zero2, micro batch size=8, lora_rank=8, VRAM Usage=21167.73 MB  
-  - zero2, micro batch size=8, lora_rank=32, VRAM Usage=21344.17 MB  
+  - zero2, micro batch size=8, lora_rank=8, VRAM Usage=21167.73 MB
+  - zero2, micro batch size=8, lora_rank=32, VRAM Usage=21344.17 MB
 
 For PPO, we suggest using Tensor Parallelism. The following table shows the VRAM consumption of training a 7B model (llama2-7B-hf) on a dummy dataset with a sequence length of 2048 and a layout length of 512 with different tp_size (equal to the number of GPUs).
 | PPO   | tp=8          | tp=4          |

applications/ColossalChat/examples/training_scripts/train_dpo.py

@@ -135,7 +135,7 @@ def train(args):
         # Note, for some models, lora may not be compatible with gradient checkpointing
         model.gradient_checkpointing_enable()
         coordinator.print_on_master(msg="Gradient checkpointing enabled successfully")
-        
+
     # configure tokenizer
     tokenizer_dir = args.tokenizer_dir if args.tokenizer_dir is not None else args.pretrain
     tokenizer = AutoTokenizer.from_pretrained(tokenizer_dir, use_fast=False, trust_remote_code=True)

applications/ColossalChat/examples/training_scripts/train_orpo.py

@@ -122,7 +122,7 @@ def train(args):
         # Note, for some models, lora may not be compatible with gradient checkpointing
         model.gradient_checkpointing_enable()
         coordinator.print_on_master(msg="Gradient checkpointing enabled successfully")
-        
+
     # configure tokenizer
     tokenizer_dir = args.tokenizer_dir if args.tokenizer_dir is not None else args.pretrain
     tokenizer = AutoTokenizer.from_pretrained(tokenizer_dir, use_fast=False, trust_remote_code=True)

applications/ColossalChat/examples/training_scripts/train_sft.py

@@ -126,7 +126,7 @@ def train(args):
         # Note, for some models, lora may not be compatible with gradient checkpointing
         model.gradient_checkpointing_enable()
         coordinator.print_on_master(msg="Gradient checkpointing enabled successfully")
-        
+
     # configure tokenizer
     tokenizer = AutoTokenizer.from_pretrained(
         args.tokenizer_dir or args.pretrain, use_fast=False, trust_remote_code=True