@ -359,7 +359,7 @@ To prepare your training data, you need to put all the samples into a list and s
"conversations": [
{
"from": "user",
"value": "你好",
"value": "你好"
},
"from": "assistant",