Remove Agent's Last Exam entry
#5
by SaylorTwift HF Staff - opened
.eval_results/Step-5-Preview.yaml
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
- dataset:
|
| 2 |
+
id: datacurve/deep-swe
|
| 3 |
+
task_id: deep_swe
|
| 4 |
+
value: 67.7
|
| 5 |
+
date: "2026-09-20"
|
| 6 |
+
source:
|
| 7 |
+
url: https://huggingface.co/stepfun-ai/Step-5-Preview
|
| 8 |
+
name: "Step 5 Preview model card"
|
| 9 |
+
notes: "DeepSWE 1.1, run with the SWE-agent harness."
|
| 10 |
+
- dataset:
|
| 11 |
+
id: harborframework/terminal-bench
|
| 12 |
+
task_id: terminalbench_4
|
| 13 |
+
value: 33.3
|
| 14 |
+
date: "2026-09-20"
|
| 15 |
+
source:
|
| 16 |
+
url: https://huggingface.co/stepfun-ai/Step-5-Preview
|
| 17 |
+
name: "Step 5 Preview model card"
|