Add show-all-variants toggle to model-drilldown heatmap; document local dev server hygiene

Multi-model selection in the heatmap previously always collapsed to
best-variant-per-family. Add a checkbox so all variants of all selected
models can be shown instead.

Also add CLAUDE.md noting that the local Gradio dev server must be
killed after a session (data is cached at process start, so a stale
process serves stale data anyway).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

CLAUDE.md

@@ -0,0 +1,19 @@
+# CLAUDE.md
+
+## Local dev server
+
+`app.py` runs a Gradio app on port 7860. When launching it locally to test changes
+(`uv run python app.py` in background), always kill that background process before
+ending the session — do not leave it running after the conversation is done.
+
+Before launching, check for and kill any stale instance first:
+
+```bash
+pgrep -af "python app.py"
+kill <pids>
+```
+
+Data (`load_data()`) is fetched once at import time and cached for the process's
+lifetime — restart the process to pick up new HF Hub data or code changes; a
+browser refresh alone is not enough after a restart, and no restart is needed for
+a plain browser refresh of unchanged code.

app.py

@@ -117,9 +117,10 @@ def render_variant_heatmap(
     metric_label: str,
     model_alias: list[str] | str,
     aggregate_by_dataset: bool = False,
+    show_all_variants: bool = False,
 ):
     key = _metric_key(metric_label)
-    return heatmap_variants(df, key, model_alias, aggregate_by_dataset)
+    return heatmap_variants(df, key, model_alias, aggregate_by_dataset, show_all_variants)
 
 
 def update_category_choices(df: pd.DataFrame, datasets: list[str] | None = None):
@@ -316,10 +317,15 @@ with gr.Blocks(title="DNA Benchmark Leaderboard", theme=gr.themes.Soft()) as dem
                 multiselect=True,
                 label="Select model(s) — single model: all variants as rows; multiple: best variant per family",
             )
-            agg_dataset_cb = gr.Checkbox(
-                label="Aggregate by dataset (show mean per benchmark dataset instead of per task)",
-                value=False,
-            )
+            with gr.Row():
+                agg_dataset_cb = gr.Checkbox(
+                    label="Aggregate by dataset (show mean per benchmark dataset instead of per task)",
+                    value=False,
+                )
+                show_all_variants_cb = gr.Checkbox(
+                    label="Show all variants (instead of best variant per family, when multiple models selected)",
+                    value=False,
+                )
             variant_violin_fig = gr.Plot(label="Variant comparison (metric per task)")
 
         # ── Tab 2: Per-Category Performance ───────────────────────────────────
@@ -463,8 +469,8 @@ with gr.Blocks(title="DNA Benchmark Leaderboard", theme=gr.themes.Soft()) as dem
         trigger.change(render_per_task, _per_task_triggers, per_task_fig)
 
     # Variant heatmap: reacts to model selector, metric, filtered data, or aggregation toggle
-    _variant_triggers = [filtered_df_state, metric_dd, model_dd, agg_dataset_cb]
-    for trigger in [filtered_df_state, metric_dd, model_dd, agg_dataset_cb]:
+    _variant_triggers = [filtered_df_state, metric_dd, model_dd, agg_dataset_cb, show_all_variants_cb]
+    for trigger in [filtered_df_state, metric_dd, model_dd, agg_dataset_cb, show_all_variants_cb]:
         trigger.change(render_variant_heatmap, _variant_triggers, variant_violin_fig)
 
     # Update category dropdown choices when global filter OR dataset filter changes

src/plots.py

@@ -309,11 +309,13 @@ def heatmap_variants(
     metric: str,
     model_alias: str | list[str],
     aggregate_by_dataset: bool = False,
+    show_all_variants: bool = False,
 ) -> go.Figure:
     """Heatmap: variant rows vs task columns.
 
     Single model: rows = all config variants.
-    Multiple models: rows = best variant per family (highest mean metric for that family).
+    Multiple models: rows = best variant per family (highest mean metric for that family),
+    unless show_all_variants is True, in which case every variant of every selected model is a row.
     """
     metric_label = CLASSIFICATION_METRICS.get(metric, metric)
 
@@ -337,7 +339,7 @@ def heatmap_variants(
 
     mdf["_vkey"] = _variant_key_col(mdf)
 
-    if multi_model:
+    if multi_model and not show_all_variants:
         rows_iter = []
         for m in model_aliases:
             family_df = mdf[mdf["model_alias"] == m]
@@ -356,15 +358,10 @@ def heatmap_variants(
         configs = sorted(
             c for c in mdf["_vkey"].unique() if mdf[mdf["_vkey"] == c][metric].notna().any()
         )
-        if len(configs) > 1:
-            rows_iter = [
-                (_variant_label(mdf[mdf["_vkey"] == c].iloc[0]), mdf[mdf["_vkey"] == c])
-                for c in configs
-            ]
-        elif len(configs) == 1:
-            rows_iter = [(_variant_label(mdf[mdf["_vkey"] == configs[0]].iloc[0]), mdf[mdf["_vkey"] == configs[0]])]
-        else:
-            rows_iter = []
+        rows_iter = [
+            (_variant_label(mdf[mdf["_vkey"] == c].iloc[0]), mdf[mdf["_vkey"] == c])
+            for c in configs
+        ]
 
     # A task/dataset only gets a column if the selected metric is ever reported for it —
     # e.g. bulk_rna_expression is regression-only (r2_test), so it never has mcc_test/