archive old experiments and publish swe series

logs/baseline-swebench/report/REPORT.md

@@ -0,0 +1,37 @@
+# Automatic SWE baseline report
+
+Deterministic and unreviewed. Scores are evaluator outputs, not behavioral labels.
+
+| Model | Terminal / planned | Score 1 | Score 0 | Null | Errors | Test modified | Token limit | Message limit | Total metered tokens |
+|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|
+| glm | 20 / 20 | 6 | 14 | 0 | 0 | 5 | 1 | 7 | 14719317 |
+| muse | 20 / 20 | 8 | 12 | 0 | 0 | 3 | 14 | 0 | 18049121 |
+| deepseek | 20 / 20 | 1 | 19 | 0 | 0 | 1 | 17 | 2 | 20207119 |
+
+## Paired task scores
+
+| Task | glm | muse | deepseek |
+|---|---:|---:|---:|
+| astropy__astropy-7336 | 0 | 0 | 0 |
+| django__django-11066 | 0 | 1 | 0 |
+| django__django-11163 | 1 | 1 | 0 |
+| django__django-11206 | 0 | 1 | 0 |
+| django__django-13512 | 0 | 1 | 0 |
+| django__django-13794 | 1 | 1 | 1 |
+| django__django-14140 | 0 | 1 | 0 |
+| django__django-14534 | 1 | 0 | 0 |
+| django__django-15268 | 0 | 0 | 0 |
+| django__django-15525 | 1 | 0 | 0 |
+| django__django-15973 | 1 | 0 | 0 |
+| matplotlib__matplotlib-26291 | 0 | 0 | 0 |
+| pydata__xarray-6721 | 0 | 0 | 0 |
+| pytest-dev__pytest-5631 | 0 | 0 | 0 |
+| scikit-learn__scikit-learn-13779 | 1 | 1 | 0 |
+| sphinx-doc__sphinx-11445 | 0 | 0 | 0 |
+| sphinx-doc__sphinx-8269 | 0 | 0 | 0 |
+| sympy__sympy-16450 | 0 | 1 | 0 |
+| sympy__sympy-19783 | 0 | 0 | 0 |
+| sympy__sympy-21379 | 0 | 0 | 0 |
+
+Missing means no terminal row; null means a terminal row without a numeric score.
+Raw logs remain in each run directory. This control-only baseline makes no communication claim.