LLAMA3 70B: LoRa enabled in all modules instead of only LinearQKV

scripts/performance/configs/llama/llama3_llm_finetune.py

@@ -210,6 +210,9 @@ def llama3_70b_lora_config_gb300(precision: str = "bf16", config_variant: str =
     set_llama3_common_peft_configs(cfg)
     set_workload_base_configs(cfg, base_cfg)
 
+    # Override target_modules to only apply LoRA to QKV
+    cfg.peft.target_modules = ["linear_qkv"]
+
     # Enable pad_cu_seqlens for CUDA graphs compatibility with packed sequences.
     # This ensures consistent cu_seqlens tensor shapes across batches, which is required
     # for CUDA graphs and avoids NaN issues in attention kernels.
@@ -240,6 +243,9 @@ def llama3_70b_lora_config_gb200(precision: str = "bf16", config_variant: str =
     set_llama3_common_peft_configs(cfg)
     set_workload_base_configs(cfg, base_cfg)
 
+    # Override target_modules to only apply LoRA to QKV
+    cfg.peft.target_modules = ["linear_qkv"]
+
     return cfg
 
 
@@ -264,4 +270,7 @@ def llama3_70b_lora_config_h100(precision: str = "bf16", config_variant: str = "
     set_llama3_common_peft_configs(cfg)
     set_workload_base_configs(cfg, base_cfg)
 
+    # Override target_modules to only apply LoRA to QKV
+    cfg.peft.target_modules = ["linear_qkv"]
+
     return cfg