Score tab: clearer length prompt + friendly too-short error message

app.py

@@ -133,7 +133,7 @@ def draw(dataset, which):
 def predict(text, clf_name):
     text = (text or "").strip()
     if not text:
-        return "Paste a passage first.", None
+        return "Please enter some input text first.", None
     if EXTRACTOR is None:
         return f"⚠️ Base model unavailable: {_LOAD_ERR}", None
     try:
@@ -176,10 +176,11 @@ with gr.Blocks(title="Predicting Memorization Before Fine-Tuning") as demo:
         feat = gr.Plot(label="Where this example's features fall (memorized vs not-memorized)")
         go.click(draw, [d_in, f_in], [pref, gen, verdict, feat])
     with gr.Tab("Score your own text"):
-        gr.Markdown("Paste a passage (at least ~100 tokens). Using its base-model features from "
-                    "Pythia-1.4B, our learned classifiers (one per fine-tuning dataset) predict how likely "
-                    "Pythia-1.4B would be to memorize it if you fine-tuned on it. The plots below show where "
-                    "your text's features fall relative to memorized and not-memorized examples.")
+        gr.Markdown("Please enter an input text that is at least 100 tokens (or 75 words). Using its "
+                    "base-model features from Pythia-1.4B, our learned classifiers (one per fine-tuning "
+                    "dataset) predict how likely Pythia-1.4B would be to memorize it if you fine-tuned on "
+                    "it. The plots below show where your text's features fall relative to memorized and "
+                    "not-memorized examples.")
         txt = gr.Textbox(label="Your text", lines=6)
         clf_in = gr.Dropdown(sorted(CLFS.keys()), value=sorted(CLFS.keys())[0],
                              label="Choose a classifier")

byo_features.py

@@ -33,7 +33,8 @@ class BYOExtractor:
         ids = enc["input_ids"]
         if ids.shape[1] < WINDOW_LEN:
             raise ValueError(
-                f"Need at least {WINDOW_LEN} tokens (got {ids.shape[1]}). Paste a longer passage.")
+                f"The input should be at least {WINDOW_LEN} tokens (or about 75 words); "
+                f"you entered {ids.shape[1]} tokens.")
         return self.features_from_ids(ids[:, :WINDOW_LEN])
 
     def features_from_ids(self, ids):