ChatGLM support

Signed-off-by: mymusise <mymusise1@gmail.com>

README.md

@@ -226,6 +226,7 @@ An example is provided in `~examples/causal_language_modeling/peft_lora_clm_acce
 | GPT-J        | ✅  | ✅  | ✅  | ✅  |
 | GPT-NeoX-20B | ✅  | ✅  | ✅  | ✅  |
 | LLaMA        | ✅  | ✅  | ✅  | ✅  |
+| ChatGLM      | ✅  | ✅  | ✅  | ✅  |
 
 ### Conditional Generation
 |   Model         | LoRA | Prefix Tuning  | P-Tuning | Prompt Tuning  | 

src/peft/mapping.py

@@ -56,6 +56,7 @@ TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING = {
     "deberta": ["in_proj"],
     "layoutlm": ["query", "value"],
     "llama": ["q_proj", "v_proj"],
+    "chatglm": ["query_key_value"],
 }