minor update

Signed-off-by: Suguna Velury <[email protected]>

Author: sugunav14

examples/llm_qat/README.md

@@ -354,7 +354,7 @@ To perform QLoRA training, run:
    --lora True
 ```
 
-After performing QLoRA training the final checkpoint is exported to be ready for deployment. For more details about QLoRA deployment using vLLM dere to the documentation [here](https://docs.vllm.ai/en/latest/features/lora.html). To deploy with vLLM, run:
+After performing QLoRA training the final checkpoint exported is ready for deployment using vLLM. For more details about QLoRA deployment using vLLM refer to the documentation [here](https://docs.vllm.ai/en/latest/features/lora.html). To deploy with vLLM, run:
 
 ```sh
 vllm serve llama3-fp4-qlora/base_model --enable-lora --lora-modules adapter=llama3-fp4-qlora --port 8000 --tokenizer llama3-fp4-qlora

modelopt/torch/quantization/plugins/transformers_trainer.py

@@ -16,7 +16,6 @@
 """ModelOpt plugin for transformers Trainer."""
 
 import gc
-import json
 import os
 import types
 from dataclasses import dataclass, field
@@ -219,7 +218,6 @@ def forward_loop(model):
         gc.collect()
 
         self._save_modelopt_state_with_weights()
-
         torch.cuda.empty_cache()
 
         if self.accelerator.is_main_process:
@@ -294,8 +292,6 @@ def export_base_model(self):
         """Export the basemodel to HF checkpoint for deployment."""
         # Save config.json
         if self.accelerator.is_main_process:
-            with open(f"{self.args.output_dir}/config.json", "w") as f:
-                json.dump(self.model.config.to_dict(), f, indent=2)
             export_hf_checkpoint(self.model, export_dir=f"{self.args.output_dir}/base_model")
 
     def _patch_accelerate_for_fsdp2_fix(self):