pushpam14 commited on
Commit
9c794e4
·
verified ·
1 Parent(s): 276b91a

Add baseline_72b_v2_scores.json for judges to inspect from Space repo root

Browse files
Files changed (1) hide show
  1. baseline_72b_v2_scores.json +160 -0
baseline_72b_v2_scores.json ADDED
@@ -0,0 +1,160 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": "Qwen/Qwen2.5-72B-Instruct",
3
+ "benchmark": "api_contract_validator",
4
+ "date": "2026-04-26",
5
+ "scores": {
6
+ "find_type_mismatches": 0.75,
7
+ "validate_nested_objects": 0.99,
8
+ "detect_breaking_changes": 0.01,
9
+ "validate_response_schema": 0.99,
10
+ "validate_cross_field_constraints": 0.99,
11
+ "validate_auth_request": 0.99,
12
+ "trace_downstream_blast_radius": 0.6667,
13
+ "propose_backward_compat_fix": 0.99,
14
+ "multi_service_cascade_fix": 0.99
15
+ },
16
+ "details": [
17
+ {
18
+ "task": "find_type_mismatches",
19
+ "score": 0.75,
20
+ "steps": 5,
21
+ "success": true,
22
+ "rewards": [
23
+ 1.0,
24
+ 1.0,
25
+ 1.0,
26
+ -0.1,
27
+ 0.375
28
+ ]
29
+ },
30
+ {
31
+ "task": "validate_nested_objects",
32
+ "score": 0.99,
33
+ "steps": 12,
34
+ "success": true,
35
+ "rewards": [
36
+ 1.0,
37
+ 1.0,
38
+ 1.0,
39
+ 1.0,
40
+ 1.0,
41
+ -0.3,
42
+ 1.0,
43
+ -0.3,
44
+ -0.3,
45
+ -0.3,
46
+ -0.5,
47
+ 1.0
48
+ ]
49
+ },
50
+ {
51
+ "task": "detect_breaking_changes",
52
+ "score": 0.01,
53
+ "steps": 20,
54
+ "success": false,
55
+ "rewards": [
56
+ 0.3,
57
+ 0.3,
58
+ 0.3,
59
+ 0.3,
60
+ 0.3,
61
+ 0.3,
62
+ -0.3,
63
+ -0.3,
64
+ -0.3,
65
+ -0.3,
66
+ -0.3,
67
+ -0.3,
68
+ -0.3,
69
+ -0.3,
70
+ -0.3,
71
+ -0.3,
72
+ -0.3,
73
+ -0.3,
74
+ -0.3,
75
+ -0.3
76
+ ]
77
+ },
78
+ {
79
+ "task": "validate_response_schema",
80
+ "score": 0.99,
81
+ "steps": 10,
82
+ "success": true,
83
+ "rewards": [
84
+ 1.0,
85
+ 1.0,
86
+ 1.0,
87
+ 1.0,
88
+ 1.0,
89
+ 1.0,
90
+ 1.0,
91
+ 1.0,
92
+ 1.0,
93
+ 1.0
94
+ ]
95
+ },
96
+ {
97
+ "task": "validate_cross_field_constraints",
98
+ "score": 0.99,
99
+ "steps": 8,
100
+ "success": true,
101
+ "rewards": [
102
+ 1.0,
103
+ 1.0,
104
+ -0.3,
105
+ 1.0,
106
+ 1.0,
107
+ 1.0,
108
+ 1.0,
109
+ 1.0
110
+ ]
111
+ },
112
+ {
113
+ "task": "validate_auth_request",
114
+ "score": 0.99,
115
+ "steps": 10,
116
+ "success": true,
117
+ "rewards": [
118
+ 1.0,
119
+ 1.0,
120
+ 1.0,
121
+ 1.0,
122
+ 1.0,
123
+ -0.3,
124
+ -0.3,
125
+ -0.3,
126
+ -0.5,
127
+ 1.0
128
+ ]
129
+ },
130
+ {
131
+ "task": "trace_downstream_blast_radius",
132
+ "score": 0.6667,
133
+ "steps": 1,
134
+ "success": true,
135
+ "rewards": [
136
+ 0.1
137
+ ]
138
+ },
139
+ {
140
+ "task": "propose_backward_compat_fix",
141
+ "score": 0.99,
142
+ "steps": 1,
143
+ "success": true,
144
+ "rewards": [
145
+ 2.0
146
+ ]
147
+ },
148
+ {
149
+ "task": "multi_service_cascade_fix",
150
+ "score": 0.99,
151
+ "steps": 3,
152
+ "success": true,
153
+ "rewards": [
154
+ 1.6,
155
+ -2.0,
156
+ 2.0
157
+ ]
158
+ }
159
+ ]
160
+ }