Empirical filter results: DeepSeek reruns W6A/W7A (partial)

2026-10-03 18:54:16 +02:00
parent d83a39a8d8
commit 5e3baf43d3
13 changed files with 210 additions and 0 deletions
--- a/tasks_gen/eval/G0174/empirical.json
+++ b/tasks_gen/eval/G0174/empirical.json
@@ -35,5 +35,26 @@
   "seconds": 78.1,
   "run_dir": "runs/emp/16010_G0174_llm_deepseek-v4.1-flash_cloud"
  }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 100,
+  "parts": {
+   "total": 100,
+   "stop": "stopped",
+   "gap": {
+    "how": "keywords",
+    "group": [
+     [
+      "small",
+      "sale",
+      "defin"
+     ]
+    ]
+   }
+  },
+  "hidden": "0/0",
+  "tool_calls": 17,
+  "seconds": 24.3,
+  "run_dir": "runs/emp/17008_G0174_llm_deepseek-v4.1-flash_cloud"
 }
 }