Update training_infos.md

training_infos.md

@@ -68,4 +68,22 @@ model, tokenizer = FastLanguageModel.from_pretrained(
     dtype = dtype,
     load_in_4bit = load_in_4bit,
 )
+```
+
+# Peft model args
+
+```
+model = FastLanguageModel.get_peft_model(
+    model,
+    r = 8,
+    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
+                      "gate_proj", "up_proj", "down_proj",],
+    lora_alpha = 16,
+    lora_dropout = 0,
+    bias = "none",
+    use_gradient_checkpointing = True
+    random_state = 3407,
+    use_rslora = False,
+    loftq_config = None,
+)
 ```