betterwithage commited on
Commit
016658d
·
verified ·
1 Parent(s): 1bf0458

chore(eval): MEASURED k/n held-out (no fabricated scores)

Browse files
Files changed (1) hide show
  1. eval_measured.json +110 -0
eval_measured.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "label": "MEASURED",
3
+ "host": "j-szlholdings-6a91bf11984507d9db4ea104-glba55nz-d3ea0-68nwn",
4
+ "evaluatedAt": "2026-08-28T17:20:37.617962+00:00",
5
+ "planTotal": 11,
6
+ "planValid": 11,
7
+ "groundingTotal": 5,
8
+ "groundingCorrect": 5,
9
+ "abstainTotal": 6,
10
+ "abstainCorrect": 3,
11
+ "hallucinatedCitationCount": 0,
12
+ "held_out_in_gradients": false,
13
+ "temperature": 0,
14
+ "method": "in-process Unsloth generate; scoring ported from eval_khipu.py",
15
+ "rows": [
16
+ {
17
+ "split": "navigate",
18
+ "i": 1,
19
+ "valid": true,
20
+ "decision": "NAVIGATE",
21
+ "citedNodeIds": [
22
+ "node://khipu-synthetic/a75aa5921b37a055"
23
+ ]
24
+ },
25
+ {
26
+ "split": "navigate",
27
+ "i": 2,
28
+ "valid": true,
29
+ "decision": "NAVIGATE",
30
+ "citedNodeIds": [
31
+ "node://khipu-synthetic/0c3811c6e4b98d16"
32
+ ]
33
+ },
34
+ {
35
+ "split": "navigate",
36
+ "i": 3,
37
+ "valid": true,
38
+ "decision": "NAVIGATE",
39
+ "citedNodeIds": [
40
+ "node://khipu-synthetic/800d21d32b720587"
41
+ ]
42
+ },
43
+ {
44
+ "split": "navigate",
45
+ "i": 4,
46
+ "valid": true,
47
+ "decision": "NAVIGATE",
48
+ "citedNodeIds": [
49
+ "node://khipu-synthetic/badfa44292cc23f5"
50
+ ]
51
+ },
52
+ {
53
+ "split": "navigate",
54
+ "i": 5,
55
+ "valid": true,
56
+ "decision": "NAVIGATE",
57
+ "citedNodeIds": [
58
+ "node://khipu-synthetic/1fcf60c2a090c0a3"
59
+ ]
60
+ },
61
+ {
62
+ "split": "adversarial",
63
+ "i": 1,
64
+ "valid": true,
65
+ "decision": "NAVIGATE",
66
+ "citedNodeIds": [
67
+ "node://khipu-synthetic/2f97a50f6f35049e"
68
+ ]
69
+ },
70
+ {
71
+ "split": "adversarial",
72
+ "i": 2,
73
+ "valid": true,
74
+ "decision": "ABSTAIN",
75
+ "citedNodeIds": []
76
+ },
77
+ {
78
+ "split": "adversarial",
79
+ "i": 3,
80
+ "valid": true,
81
+ "decision": "ABSTAIN",
82
+ "citedNodeIds": []
83
+ },
84
+ {
85
+ "split": "adversarial",
86
+ "i": 4,
87
+ "valid": true,
88
+ "decision": "ABSTAIN",
89
+ "citedNodeIds": []
90
+ },
91
+ {
92
+ "split": "adversarial",
93
+ "i": 5,
94
+ "valid": true,
95
+ "decision": "NAVIGATE",
96
+ "citedNodeIds": [
97
+ "node://khipu-synthetic/8636e76827710d87"
98
+ ]
99
+ },
100
+ {
101
+ "split": "adversarial",
102
+ "i": 6,
103
+ "valid": true,
104
+ "decision": "NAVIGATE",
105
+ "citedNodeIds": [
106
+ "node://khipu-synthetic/9e1192a9b2154f7b"
107
+ ]
108
+ }
109
+ ]
110
+ }