Upload folder using huggingface_hub
Browse files
.eval_results/multi_agent_evaluation.yaml
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
- dataset:
|
| 2 |
+
id: TIGER-Lab/MMLU-Pro
|
| 3 |
+
task_id: default
|
| 4 |
+
value: 82.35
|
| 5 |
+
notes: multi-agent-consensus-dynamic-gating
|
| 6 |
+
date: '2025-01-26'
|
| 7 |
+
- dataset:
|
| 8 |
+
id: princeton-nlp/SWE-bench
|
| 9 |
+
task_id: swe_bench_pro
|
| 10 |
+
value: 64.0
|
| 11 |
+
notes: metacognitive-reflector-arbitration
|
| 12 |
+
date: '2025-01-26'
|
| 13 |
+
- dataset:
|
| 14 |
+
id: evaluation/Long-Horizon-Terminal-Bench
|
| 15 |
+
task_id: terminal_cli_trajectory
|
| 16 |
+
value: 91.52
|
| 17 |
+
notes: context-retention-94.20-command-accuracy-87.50
|
| 18 |
+
date: '2025-01-26'
|