Upload runs/jevbench_nextrun_v10.log with huggingface_hub

runs/jevbench_nextrun_v10.log

@@ -0,0 +1,35 @@
+`torch_dtype` is deprecated! Use `dtype` instead!
+easy: 25/48
+
+== easy: n=48 acc=0.688 cci=56.3 brier=0.5126 ece=0.1202 p50=0.039s p95=0.045s
+   extraction             9/12 = 0.750
+   fact                   10/12 = 0.833
+   intent                 6/12 = 0.500
+   tool_selection         8/12 = 0.667
+hard: 25/111
+hard: 50/111
+hard: 75/111
+hard: 100/111
+
+== hard: n=111 acc=0.360 cci=3.6 brier=0.7929 ece=0.2654 p50=0.040s p95=0.049s
+   adversarial            3/6 = 0.500
+   ambiguous              2/7 = 0.286
+   judge_hard             9/17 = 0.529
+   long_policy            6/19 = 0.316
+   multi_hop              6/18 = 0.333
+   probability            4/10 = 0.400
+   routing_hard           3/5 = 0.600
+   temporal_numeric       3/15 = 0.200
+   tradeoff               2/6 = 0.333
+   trap                   2/8 = 0.250
+original: 25/72
+original: 50/72
+
+== original: n=72 acc=0.389 cci=11.3 brier=0.7829 ece=0.3118 p50=0.039s p95=0.039s
+   adequacy               4/12 = 0.333
+   extraction             5/12 = 0.417
+   intent                 4/12 = 0.333
+   ordinal                5/12 = 0.417
+   policy                 5/12 = 0.417
+   routing                5/12 = 0.417
+wrote /root/alea/jevbench_results.json