diff --git a/prompterator/constants.py b/prompterator/constants.py
index 81d0de6..04d8447 100644
--- a/prompterator/constants.py
+++ b/prompterator/constants.py
@@ -82,6 +82,7 @@ def call(self, input, **kwargs):
     RESPONSE_DATA_COL: object,
     LABEL_COL: bool,
 }
+# these are the columns that users won't be able to show or inject into their prompts
 COLS_NOT_FOR_PROMPT_INTERPOLATION = [
     TEXT_GENERATED_COL,
     SYSTEM_PROMPT_TEMPLATE_COL,
@@ -91,7 +92,7 @@ def call(self, input, **kwargs):
 ]
 LABEL_GOOD = "good"
 LABEL_BAD = "bad"
-COLS_TO_SHOW = [TEXT_ORIG_COL, TEXT_GENERATED_COL, LABEL_COL]
+DUMMY_DATA_COLS = [TEXT_ORIG_COL, TEXT_GENERATED_COL, LABEL_COL]
 LABEL_VALUE_COLOURS = {
     LABEL_GOOD: "#56E7AB",
     LABEL_BAD: "#FE8080",
diff --git a/prompterator/main.py b/prompterator/main.py
index f36a51a..35b5ab1 100644
--- a/prompterator/main.py
+++ b/prompterator/main.py
@@ -514,12 +514,12 @@ def set_up_ui_labelling():
 
 def show_col_selection():
     if st.session_state.get("df") is not None:
-        columns_sel = st.session_state.df.columns.tolist()
-        columns_sel.remove(c.TEXT_GENERATED_COL)
-        columns_sel.remove(c.LABEL_COL)
-        columns_sel.remove(c.RESPONSE_DATA_COL)
+        available_columns = st.session_state.df.columns.tolist()
+        available_columns = [
+            col for col in available_columns if col not in c.COLS_NOT_FOR_PROMPT_INTERPOLATION
+        ]
         st.session_state["columns_to_show"] = st.multiselect(
-            "Columns to show", columns_sel, [c.TEXT_ORIG_COL]
+            "Columns to show", options=available_columns, default=[c.TEXT_ORIG_COL]
         )
 
 
diff --git a/prompterator/utils.py b/prompterator/utils.py
index 64c0afd..5113f87 100644
--- a/prompterator/utils.py
+++ b/prompterator/utils.py
@@ -282,6 +282,6 @@ def get_dummy_dataframe():
     )
 
     # to fail if the list of required columns changes; we'll then update the hard-coded dict above
-    assert set(df.columns) == set(c.COLS_TO_SHOW)
+    assert set(df.columns) == set(c.DUMMY_DATA_COLS)
 
     return df