ZachW commited on
Commit
a2d1b30
·
verified ·
1 Parent(s): 9f113f4

Upload extrapolation artifact: 050302 RL OP13-20 step200 metrics

Browse files
050302/evals/rl050302_op13_20_step200_merged_hf/rl050302_op13_20_step200_merged_hf_metrics.json ADDED
@@ -0,0 +1,1897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "rl050302_op13_20_step200_merged_hf",
3
+ "id": {
4
+ "answer_accuracy": 0.7096996527777778,
5
+ "avg_response_len": 206.4914982638889,
6
+ "avg_loss": 0.0,
7
+ "count": 1152000,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {
10
+ "pass@1": 0.7096996527777778,
11
+ "pass@2": 0.7628915545712934,
12
+ "pass@4": 0.805451991355243,
13
+ "pass@8": 0.8398434241626531,
14
+ "pass@16": 0.8676994968313153,
15
+ "pass@32": 0.8901977893513012,
16
+ "pass@64": 0.90809973689971,
17
+ "pass@128": 0.9224444444444444
18
+ },
19
+ "per_op_pass_at_k": {
20
+ "10": {
21
+ "pass@1": 0.6372265625,
22
+ "pass@2": 0.7061861466535435,
23
+ "pass@4": 0.7634870098425202,
24
+ "pass@8": 0.8101290041388484,
25
+ "pass@16": 0.8469615673260908,
26
+ "pass@32": 0.8759298968033682,
27
+ "pass@64": 0.8982367360000915,
28
+ "pass@128": 0.915
29
+ },
30
+ "2": {
31
+ "pass@1": 0.9284296875,
32
+ "pass@2": 0.9552456938976382,
33
+ "pass@4": 0.9715452533745776,
34
+ "pass@8": 0.9801859779187886,
35
+ "pass@16": 0.9846789731901037,
36
+ "pass@32": 0.9876162438902186,
37
+ "pass@64": 0.9896641079037857,
38
+ "pass@128": 0.991
39
+ },
40
+ "3": {
41
+ "pass@1": 0.881828125,
42
+ "pass@2": 0.9067679625984255,
43
+ "pass@4": 0.9261030650543686,
44
+ "pass@8": 0.9414119676703517,
45
+ "pass@16": 0.9534031663493782,
46
+ "pass@32": 0.9632878571158628,
47
+ "pass@64": 0.9719825705210999,
48
+ "pass@128": 0.979
49
+ },
50
+ "4": {
51
+ "pass@1": 0.77528125,
52
+ "pass@2": 0.8108379675196851,
53
+ "pass@4": 0.8371120714285711,
54
+ "pass@8": 0.859004820532304,
55
+ "pass@16": 0.8781944005645403,
56
+ "pass@32": 0.8952841256653082,
57
+ "pass@64": 0.9091673998059235,
58
+ "pass@128": 0.92
59
+ },
60
+ "5": {
61
+ "pass@1": 0.6686484375,
62
+ "pass@2": 0.7283939468503943,
63
+ "pass@4": 0.7806958164604425,
64
+ "pass@8": 0.8230992439889415,
65
+ "pass@16": 0.8552869515145666,
66
+ "pass@32": 0.8789656965059433,
67
+ "pass@64": 0.896464829124427,
68
+ "pass@128": 0.91
69
+ },
70
+ "6": {
71
+ "pass@1": 0.6456640625,
72
+ "pass@2": 0.6965431840551182,
73
+ "pass@4": 0.7367055818335206,
74
+ "pass@8": 0.7715006117837285,
75
+ "pass@16": 0.8038770259652023,
76
+ "pass@32": 0.833591385701847,
77
+ "pass@64": 0.8590963035813479,
78
+ "pass@128": 0.88
79
+ },
80
+ "7": {
81
+ "pass@1": 0.63528125,
82
+ "pass@2": 0.6974590305118106,
83
+ "pass@4": 0.7489790861454824,
84
+ "pass@8": 0.7920296413695324,
85
+ "pass@16": 0.8276452842242492,
86
+ "pass@32": 0.856389168515698,
87
+ "pass@64": 0.8784913178107205,
88
+ "pass@128": 0.896
89
+ },
90
+ "8": {
91
+ "pass@1": 0.608125,
92
+ "pass@2": 0.6794880659448826,
93
+ "pass@4": 0.738280693194601,
94
+ "pass@8": 0.787894891228803,
95
+ "pass@16": 0.8287240309358392,
96
+ "pass@32": 0.8602222897163663,
97
+ "pass@64": 0.8840336104013778,
98
+ "pass@128": 0.902
99
+ },
100
+ "9": {
101
+ "pass@1": 0.6068125,
102
+ "pass@2": 0.6851019931102368,
103
+ "pass@4": 0.746159344863142,
104
+ "pass@8": 0.7933346588325872,
105
+ "pass@16": 0.830524071411799,
106
+ "pass@32": 0.8604934402470777,
107
+ "pass@64": 0.8857607569485678,
108
+ "pass@128": 0.909
109
+ }
110
+ },
111
+ "per_op_accuracy": {
112
+ "10": 0.6372265625,
113
+ "2": 0.9284296875,
114
+ "3": 0.881828125,
115
+ "4": 0.77528125,
116
+ "5": 0.6686484375,
117
+ "6": 0.6456640625,
118
+ "7": 0.63528125,
119
+ "8": 0.608125,
120
+ "9": 0.6068125
121
+ },
122
+ "per_op_avg_response_len": {
123
+ "10": 279.7796171875,
124
+ "2": 133.83121875,
125
+ "3": 153.0110625,
126
+ "4": 163.8181328125,
127
+ "5": 182.535953125,
128
+ "6": 207.005859375,
129
+ "7": 227.3503203125,
130
+ "8": 247.654734375,
131
+ "9": 263.4365859375
132
+ },
133
+ "per_op_avg_loss": {},
134
+ "per_op_length": {
135
+ "10": 279.7796171875,
136
+ "2": 133.83121875,
137
+ "3": 153.0110625,
138
+ "4": 163.8181328125,
139
+ "5": 182.535953125,
140
+ "6": 207.005859375,
141
+ "7": 227.3503203125,
142
+ "8": 247.654734375,
143
+ "9": 263.4365859375
144
+ },
145
+ "per_op_loss": {},
146
+ "per_template": {
147
+ "crazy_zootopia": {
148
+ "count": 379008,
149
+ "correct": 270717,
150
+ "answer_accuracy": 0.7142777988855117,
151
+ "avg_response_len": 197.0716765873016,
152
+ "resp_tokens_sum": 74691742,
153
+ "pass_at_k": {
154
+ "pass@1": 0.7142777988855117,
155
+ "pass@2": 0.7655705270883723,
156
+ "pass@4": 0.8061099160983823,
157
+ "pass@8": 0.8400332102538691,
158
+ "pass@16": 0.8686058261945824,
159
+ "pass@32": 0.8920989514120751,
160
+ "pass@64": 0.9104397793771627,
161
+ "pass@128": 0.9250253292806484
162
+ },
163
+ "per_op_pass_at_k": {
164
+ "10": {
165
+ "pass@1": 0.6432291666666666,
166
+ "pass@2": 0.7086545217522591,
167
+ "pass@4": 0.7614043908228281,
168
+ "pass@8": 0.8048343235263006,
169
+ "pass@16": 0.8406351544122402,
170
+ "pass@32": 0.8704496732524306,
171
+ "pass@64": 0.8947665326243408,
172
+ "pass@128": 0.9144542772861357
173
+ },
174
+ "2": {
175
+ "pass@1": 0.9332408489096573,
176
+ "pass@2": 0.9600331610125838,
177
+ "pass@4": 0.9771553707655704,
178
+ "pass@8": 0.9854254731466165,
179
+ "pass@16": 0.988133712621007,
180
+ "pass@32": 0.9893193713664016,
181
+ "pass@64": 0.9902739585246382,
182
+ "pass@128": 0.9906542056074766
183
+ },
184
+ "3": {
185
+ "pass@1": 0.8757645259938838,
186
+ "pass@2": 0.9027283645163623,
187
+ "pass@4": 0.9244454405928614,
188
+ "pass@8": 0.9431411848874264,
189
+ "pass@16": 0.9582971577237228,
190
+ "pass@32": 0.9709724633793967,
191
+ "pass@64": 0.9820870411980793,
192
+ "pass@128": 0.9908256880733946
193
+ },
194
+ "4": {
195
+ "pass@1": 0.786328125,
196
+ "pass@2": 0.811225086122047,
197
+ "pass@4": 0.8296586022450316,
198
+ "pass@8": 0.8480276124139998,
199
+ "pass@16": 0.8670405071229179,
200
+ "pass@32": 0.886580837586342,
201
+ "pass@64": 0.9031976667014938,
202
+ "pass@128": 0.915625
203
+ },
204
+ "5": {
205
+ "pass@1": 0.6481150793650794,
206
+ "pass@2": 0.7089566929133857,
207
+ "pass@4": 0.7631540902625267,
208
+ "pass@8": 0.8083630689659272,
209
+ "pass@16": 0.8431150398714341,
210
+ "pass@32": 0.8683170445108382,
211
+ "pass@64": 0.8869570995314593,
212
+ "pass@128": 0.9015873015873016
213
+ },
214
+ "6": {
215
+ "pass@1": 0.6656296781437125,
216
+ "pass@2": 0.7128238601537084,
217
+ "pass@4": 0.7472944426108409,
218
+ "pass@8": 0.777625115201791,
219
+ "pass@16": 0.8070151119011811,
220
+ "pass@32": 0.8349043396850152,
221
+ "pass@64": 0.8587159112471352,
222
+ "pass@128": 0.8772455089820359
223
+ },
224
+ "7": {
225
+ "pass@1": 0.6652875369822485,
226
+ "pass@2": 0.7235667922704184,
227
+ "pass@4": 0.77103220131877,
228
+ "pass@8": 0.8127494418381218,
229
+ "pass@16": 0.8505656587294267,
230
+ "pass@32": 0.8821587788982282,
231
+ "pass@64": 0.9041908255934645,
232
+ "pass@128": 0.9201183431952663
233
+ },
234
+ "8": {
235
+ "pass@1": 0.6062199519230769,
236
+ "pass@2": 0.6758885082273367,
237
+ "pass@4": 0.7324884194163228,
238
+ "pass@8": 0.7816419557403208,
239
+ "pass@16": 0.8223828369246351,
240
+ "pass@32": 0.852237918950051,
241
+ "pass@64": 0.8736887652975474,
242
+ "pass@128": 0.8910256410256411
243
+ },
244
+ "9": {
245
+ "pass@1": 0.6165272887323944,
246
+ "pass@2": 0.6952627675501827,
247
+ "pass@4": 0.7554617862027808,
248
+ "pass@8": 0.8041103178637404,
249
+ "pass@16": 0.8443959343789783,
250
+ "pass@32": 0.8765953737508427,
251
+ "pass@64": 0.9015005102566016,
252
+ "pass@128": 0.923943661971831
253
+ }
254
+ }
255
+ },
256
+ "teachers_in_school": {
257
+ "count": 390656,
258
+ "correct": 277017,
259
+ "answer_accuracy": 0.709107245249017,
260
+ "avg_response_len": 203.19174158338794,
261
+ "resp_tokens_sum": 79378073,
262
+ "pass_at_k": {
263
+ "pass@1": 0.709107245249017,
264
+ "pass@2": 0.7605472095876229,
265
+ "pass@4": 0.8022974578005748,
266
+ "pass@8": 0.8364300786615789,
267
+ "pass@16": 0.8646192690065752,
268
+ "pass@32": 0.8880120134325337,
269
+ "pass@64": 0.907141750943048,
270
+ "pass@128": 0.922346002621232
271
+ },
272
+ "per_op_pass_at_k": {
273
+ "10": {
274
+ "pass@1": 0.6664589723926381,
275
+ "pass@2": 0.7259352658084149,
276
+ "pass@4": 0.7729965696389173,
277
+ "pass@8": 0.8135385076355918,
278
+ "pass@16": 0.8485919842569918,
279
+ "pass@32": 0.8769118448464734,
280
+ "pass@64": 0.8981314161379662,
281
+ "pass@128": 0.9141104294478528
282
+ },
283
+ "2": {
284
+ "pass@1": 0.9396846504559271,
285
+ "pass@2": 0.9592216032836322,
286
+ "pass@4": 0.9723210259811746,
287
+ "pass@8": 0.9800715101221468,
288
+ "pass@16": 0.9846143152197563,
289
+ "pass@32": 0.9880899912285871,
290
+ "pass@64": 0.9903714652039789,
291
+ "pass@128": 0.9908814589665653
292
+ },
293
+ "3": {
294
+ "pass@1": 0.8826467803030303,
295
+ "pass@2": 0.9058510796945836,
296
+ "pass@4": 0.9239027323289134,
297
+ "pass@8": 0.9375980631196633,
298
+ "pass@16": 0.9490507863390486,
299
+ "pass@32": 0.959496338875189,
300
+ "pass@64": 0.9681067897370846,
301
+ "pass@128": 0.9727272727272728
302
+ },
303
+ "4": {
304
+ "pass@1": 0.7678439349112426,
305
+ "pass@2": 0.8053503412850018,
306
+ "pass@4": 0.8354628671600961,
307
+ "pass@8": 0.8625236702397913,
308
+ "pass@16": 0.8866633816803322,
309
+ "pass@32": 0.9069734335131655,
310
+ "pass@64": 0.9228128772044121,
311
+ "pass@128": 0.9349112426035503
312
+ },
313
+ "5": {
314
+ "pass@1": 0.6854166666666667,
315
+ "pass@2": 0.741926280954011,
316
+ "pass@4": 0.7908239029903872,
317
+ "pass@8": 0.8307448053320945,
318
+ "pass@16": 0.8603277390818179,
319
+ "pass@32": 0.881465096873712,
320
+ "pass@64": 0.897257332752116,
321
+ "pass@128": 0.9101449275362319
322
+ },
323
+ "6": {
324
+ "pass@1": 0.6161130014326648,
325
+ "pass@2": 0.6660646701486812,
326
+ "pass@4": 0.7085057325391633,
327
+ "pass@8": 0.7456869974381682,
328
+ "pass@16": 0.7804926677467746,
329
+ "pass@32": 0.8129104072803103,
330
+ "pass@64": 0.8413385287286547,
331
+ "pass@128": 0.8653295128939829
332
+ },
333
+ "7": {
334
+ "pass@1": 0.6342124277456648,
335
+ "pass@2": 0.6974431494925126,
336
+ "pass@4": 0.7495586211044426,
337
+ "pass@8": 0.7918144315944411,
338
+ "pass@16": 0.8253342442369711,
339
+ "pass@32": 0.8523804182147894,
340
+ "pass@64": 0.8755104389451583,
341
+ "pass@128": 0.8959537572254336
342
+ },
343
+ "8": {
344
+ "pass@1": 0.6068882042253522,
345
+ "pass@2": 0.678888834978374,
346
+ "pass@4": 0.7377912274169952,
347
+ "pass@8": 0.7862378604682415,
348
+ "pass@16": 0.8278968421686421,
349
+ "pass@32": 0.8634145607393767,
350
+ "pass@64": 0.8914266288458986,
351
+ "pass@128": 0.9098591549295775
352
+ },
353
+ "9": {
354
+ "pass@1": 0.6005800898203593,
355
+ "pass@2": 0.6798503141355087,
356
+ "pass@4": 0.7427577106753871,
357
+ "pass@8": 0.7911772594066232,
358
+ "pass@16": 0.8284008725461969,
359
+ "pass@32": 0.8587425143072703,
360
+ "pass@64": 0.8862981605259711,
361
+ "pass@128": 0.9131736526946108
362
+ }
363
+ }
364
+ },
365
+ "movie_festival_awards": {
366
+ "count": 382336,
367
+ "correct": 269840,
368
+ "answer_accuracy": 0.7057666555071979,
369
+ "avg_response_len": 219.20088874707065,
370
+ "resp_tokens_sum": 83808391,
371
+ "pass_at_k": {
372
+ "pass@1": 0.7057666555071979,
373
+ "pass@2": 0.7626312610551251,
374
+ "pass@4": 0.8080229726891772,
375
+ "pass@8": 0.8431429132330267,
376
+ "pass@16": 0.8699483130604533,
377
+ "pass@32": 0.8905465162485553,
378
+ "pass@64": 0.906758895709193,
379
+ "pass@128": 0.9199866086374289
380
+ },
381
+ "per_op_pass_at_k": {
382
+ "10": {
383
+ "pass@1": 0.602705223880597,
384
+ "pass@2": 0.6844697526148781,
385
+ "pass@4": 0.7563404168695326,
386
+ "pass@8": 0.8121689999230759,
387
+ "pass@16": 0.8517769048136232,
388
+ "pass@32": 0.8805199885099825,
389
+ "pass@64": 0.9018508650133183,
390
+ "pass@128": 0.9164179104477612
391
+ },
392
+ "2": {
393
+ "pass@1": 0.9134375,
394
+ "pass@2": 0.9471175478065238,
395
+ "pass@4": 0.9656707480314957,
396
+ "pass@8": 0.9754882120243954,
397
+ "pass@16": 0.9815712620898884,
398
+ "pass@32": 0.985608910192567,
399
+ "pass@64": 0.9884398719007649,
400
+ "pass@128": 0.9914285714285714
401
+ },
402
+ "3": {
403
+ "pass@1": 0.8868212463556852,
404
+ "pass@2": 0.9115012568581995,
405
+ "pass@4": 0.9298003041165043,
406
+ "pass@8": 0.9434327678795174,
407
+ "pass@16": 0.9529248871190527,
408
+ "pass@32": 0.9596095328337845,
409
+ "pass@64": 0.966078330717463,
410
+ "pass@128": 0.9737609329446064
411
+ },
412
+ "4": {
413
+ "pass@1": 0.7722953216374269,
414
+ "pass@2": 0.8158991947552606,
415
+ "pass@4": 0.8457159930118383,
416
+ "pass@8": 0.8657981988853063,
417
+ "pass@16": 0.8802608633837845,
418
+ "pass@32": 0.8918749623106107,
419
+ "pass@64": 0.901267233819749,
420
+ "pass@128": 0.9093567251461988
421
+ },
422
+ "5": {
423
+ "pass@1": 0.6706571691176471,
424
+ "pass@2": 0.7326706519221858,
425
+ "pass@4": 0.7866706808707734,
426
+ "pass@8": 0.8289938806620649,
427
+ "pass@16": 0.8614489528583463,
428
+ "pass@32": 0.8862952031282322,
429
+ "pass@64": 0.9044693028604022,
430
+ "pass@128": 0.9176470588235294
431
+ },
432
+ "6": {
433
+ "pass@1": 0.6571618690851735,
434
+ "pass@2": 0.7129446210785165,
435
+ "pass@4": 0.7565953859474185,
436
+ "pass@8": 0.7934670700328363,
437
+ "pass@16": 0.8263155758567313,
438
+ "pass@32": 0.8549766691048055,
439
+ "pass@64": 0.8790474533075844,
440
+ "pass@128": 0.8990536277602523
441
+ },
442
+ "7": {
443
+ "pass@1": 0.604356210443038,
444
+ "pass@2": 0.6695510284810123,
445
+ "pass@4": 0.724756073410127,
446
+ "pass@8": 0.7701029642296547,
447
+ "pass@16": 0.8056596299611132,
448
+ "pass@32": 0.8332147991955038,
449
+ "pass@64": 0.8542664774845095,
450
+ "pass@128": 0.870253164556962
451
+ },
452
+ "8": {
453
+ "pass@1": 0.6112284159159159,
454
+ "pass@2": 0.6834994443262159,
455
+ "pass@4": 0.7442294913023253,
456
+ "pass@8": 0.7955200017165065,
457
+ "pass@16": 0.835547167704164,
458
+ "pass@32": 0.8642999697942094,
459
+ "pass@64": 0.885844631796542,
460
+ "pass@128": 0.9039039039039038
461
+ },
462
+ "9": {
463
+ "pass@1": 0.6024165996784566,
464
+ "pass@2": 0.6791437482593614,
465
+ "pass@4": 0.7391940044873812,
466
+ "pass@8": 0.7833514191290923,
467
+ "pass@16": 0.8169698497647327,
468
+ "pass@32": 0.8439938353276553,
469
+ "pass@64": 0.8672170102630214,
470
+ "pass@128": 0.887459807073955
471
+ }
472
+ }
473
+ }
474
+ }
475
+ },
476
+ "ood": {
477
+ "answer_accuracy": 0.34355625,
478
+ "avg_response_len": 346.27614765625,
479
+ "avg_loss": 0.0,
480
+ "count": 1280000,
481
+ "validation_metrics": {},
482
+ "pass_at_k": {
483
+ "pass@1": 0.34355625,
484
+ "pass@2": 0.4494201648622108,
485
+ "pass@4": 0.550740652034111,
486
+ "pass@8": 0.6372336338476207,
487
+ "pass@16": 0.7054931516572045,
488
+ "pass@32": 0.7585784915695057,
489
+ "pass@64": 0.8008290139926026,
490
+ "pass@128": 0.835
491
+ },
492
+ "per_op_pass_at_k": {
493
+ "11": {
494
+ "pass@1": 0.5650390625,
495
+ "pass@2": 0.6540514271653557,
496
+ "pass@4": 0.7280725492125987,
497
+ "pass@8": 0.7854012029674802,
498
+ "pass@16": 0.8276654184151161,
499
+ "pass@32": 0.8613735335387038,
500
+ "pass@64": 0.8894017990938388,
501
+ "pass@128": 0.909
502
+ },
503
+ "12": {
504
+ "pass@1": 0.56540625,
505
+ "pass@2": 0.6588655265748027,
506
+ "pass@4": 0.7355278310836144,
507
+ "pass@8": 0.7988205667858062,
508
+ "pass@16": 0.847712530162352,
509
+ "pass@32": 0.8827476609527442,
510
+ "pass@64": 0.9082646513909002,
511
+ "pass@128": 0.928
512
+ },
513
+ "13": {
514
+ "pass@1": 0.46809375,
515
+ "pass@2": 0.5840370324803149,
516
+ "pass@4": 0.6835329081364819,
517
+ "pass@8": 0.7621323021661622,
518
+ "pass@16": 0.8188842777252926,
519
+ "pass@32": 0.8582276277047699,
520
+ "pass@64": 0.8857752572809072,
521
+ "pass@128": 0.906
522
+ },
523
+ "14": {
524
+ "pass@1": 0.421625,
525
+ "pass@2": 0.5463829970472442,
526
+ "pass@4": 0.6571138544244463,
527
+ "pass@8": 0.7455900198919282,
528
+ "pass@16": 0.8114067977817702,
529
+ "pass@32": 0.8578387083464307,
530
+ "pass@64": 0.8882131790112293,
531
+ "pass@128": 0.907
532
+ },
533
+ "15": {
534
+ "pass@1": 0.3358671875,
535
+ "pass@2": 0.46225799704724446,
536
+ "pass@4": 0.5836184609111363,
537
+ "pass@8": 0.6835905005192744,
538
+ "pass@16": 0.7555084305704233,
539
+ "pass@32": 0.8032069738517583,
540
+ "pass@64": 0.8366766426001925,
541
+ "pass@128": 0.865
542
+ },
543
+ "16": {
544
+ "pass@1": 0.320109375,
545
+ "pass@2": 0.44445164862204795,
546
+ "pass@4": 0.5615954588488937,
547
+ "pass@8": 0.6587711388526714,
548
+ "pass@16": 0.7343385762204016,
549
+ "pass@32": 0.7918982958088162,
550
+ "pass@64": 0.83495353589886,
551
+ "pass@128": 0.867
552
+ },
553
+ "17": {
554
+ "pass@1": 0.2281171875,
555
+ "pass@2": 0.33797797736220475,
556
+ "pass@4": 0.454242666854143,
557
+ "pass@8": 0.5582650971317871,
558
+ "pass@16": 0.6427283783542358,
559
+ "pass@32": 0.710422792436585,
560
+ "pass@64": 0.7640788455706569,
561
+ "pass@128": 0.804
562
+ },
563
+ "18": {
564
+ "pass@1": 0.2006484375,
565
+ "pass@2": 0.30093971456692936,
566
+ "pass@4": 0.4072490021559802,
567
+ "pass@8": 0.5040502013444325,
568
+ "pass@16": 0.5859192620834119,
569
+ "pass@32": 0.6560982486797597,
570
+ "pass@64": 0.7177740778270901,
571
+ "pass@128": 0.769
572
+ },
573
+ "19": {
574
+ "pass@1": 0.171234375,
575
+ "pass@2": 0.259291338582677,
576
+ "pass@4": 0.3559120425571802,
577
+ "pass@8": 0.44750835909422065,
578
+ "pass@16": 0.5271232208415118,
579
+ "pass@32": 0.595105338702795,
580
+ "pass@64": 0.6561536010103154,
581
+ "pass@128": 0.717
582
+ },
583
+ "20": {
584
+ "pass@1": 0.159421875,
585
+ "pass@2": 0.24594598917322827,
586
+ "pass@4": 0.34054174615673044,
587
+ "pass@8": 0.42820694972236445,
588
+ "pass@16": 0.503644624417362,
589
+ "pass@32": 0.568865735672675,
590
+ "pass@64": 0.6269985502419505,
591
+ "pass@128": 0.678
592
+ }
593
+ },
594
+ "per_op_accuracy": {
595
+ "11": 0.5650390625,
596
+ "12": 0.56540625,
597
+ "13": 0.46809375,
598
+ "14": 0.421625,
599
+ "15": 0.3358671875,
600
+ "16": 0.320109375,
601
+ "17": 0.2281171875,
602
+ "18": 0.2006484375,
603
+ "19": 0.171234375,
604
+ "20": 0.159421875
605
+ },
606
+ "per_op_avg_response_len": {
607
+ "11": 297.00659375,
608
+ "12": 302.433140625,
609
+ "13": 318.675875,
610
+ "14": 342.1790859375,
611
+ "15": 353.8288359375,
612
+ "16": 356.4609140625,
613
+ "17": 367.413265625,
614
+ "18": 372.1854140625,
615
+ "19": 371.895203125,
616
+ "20": 380.6831484375
617
+ },
618
+ "per_op_avg_loss": {},
619
+ "per_op_length": {
620
+ "11": 297.00659375,
621
+ "12": 302.433140625,
622
+ "13": 318.675875,
623
+ "14": 342.1790859375,
624
+ "15": 353.8288359375,
625
+ "16": 356.4609140625,
626
+ "17": 367.413265625,
627
+ "18": 372.1854140625,
628
+ "19": 371.895203125,
629
+ "20": 380.6831484375
630
+ },
631
+ "per_op_loss": {},
632
+ "per_template": {
633
+ "crazy_zootopia": {
634
+ "count": 430720,
635
+ "correct": 149425,
636
+ "answer_accuracy": 0.34691911218424965,
637
+ "avg_response_len": 330.69783385958397,
638
+ "resp_tokens_sum": 142438171,
639
+ "pass_at_k": {
640
+ "pass@1": 0.34691911218424965,
641
+ "pass@2": 0.4529183509611454,
642
+ "pass@4": 0.5544624632080759,
643
+ "pass@8": 0.6416435266881714,
644
+ "pass@16": 0.7107898965756626,
645
+ "pass@32": 0.7635782858532801,
646
+ "pass@64": 0.804925744320323,
647
+ "pass@128": 0.8392273402674592
648
+ },
649
+ "per_op_pass_at_k": {
650
+ "11": {
651
+ "pass@1": 0.5303868626062322,
652
+ "pass@2": 0.6172469245611297,
653
+ "pass@4": 0.6922284186856246,
654
+ "pass@8": 0.7539144484826834,
655
+ "pass@16": 0.8025265039258181,
656
+ "pass@32": 0.8409687637230883,
657
+ "pass@64": 0.8707484010776608,
658
+ "pass@128": 0.8895184135977338
659
+ },
660
+ "12": {
661
+ "pass@1": 0.5872175141242938,
662
+ "pass@2": 0.6783650851906224,
663
+ "pass@4": 0.753178664107664,
664
+ "pass@8": 0.8143028042975027,
665
+ "pass@16": 0.859033496675691,
666
+ "pass@32": 0.8894844562598956,
667
+ "pass@64": 0.9126887445809961,
668
+ "pass@128": 0.9322033898305084
669
+ },
670
+ "13": {
671
+ "pass@1": 0.48278311128526646,
672
+ "pass@2": 0.6048583170834043,
673
+ "pass@4": 0.7072133163017632,
674
+ "pass@8": 0.7851313248049716,
675
+ "pass@16": 0.8411216588483615,
676
+ "pass@32": 0.8791363164554562,
677
+ "pass@64": 0.9047157885030569,
678
+ "pass@128": 0.9247648902821317
679
+ },
680
+ "14": {
681
+ "pass@1": 0.43460377725856697,
682
+ "pass@2": 0.555995045011897,
683
+ "pass@4": 0.6646590192581069,
684
+ "pass@8": 0.7541116803627453,
685
+ "pass@16": 0.8232941417018945,
686
+ "pass@32": 0.8719027706379187,
687
+ "pass@64": 0.9013330720969001,
688
+ "pass@128": 0.9190031152647975
689
+ },
690
+ "15": {
691
+ "pass@1": 0.3279895231213873,
692
+ "pass@2": 0.45296072095034384,
693
+ "pass@4": 0.5703828111623328,
694
+ "pass@8": 0.6669801186470863,
695
+ "pass@16": 0.7386815028490659,
696
+ "pass@32": 0.7855314858948964,
697
+ "pass@64": 0.8175070460293162,
698
+ "pass@128": 0.846820809248555
699
+ },
700
+ "16": {
701
+ "pass@1": 0.3288054435483871,
702
+ "pass@2": 0.4507123920497839,
703
+ "pass@4": 0.5656597097741817,
704
+ "pass@8": 0.6606423419172535,
705
+ "pass@16": 0.7316974325599377,
706
+ "pass@32": 0.785344071661488,
707
+ "pass@64": 0.828913082215784,
708
+ "pass@128": 0.864516129032258
709
+ },
710
+ "17": {
711
+ "pass@1": 0.2260883309248555,
712
+ "pass@2": 0.33683285637886257,
713
+ "pass@4": 0.4542962630935586,
714
+ "pass@8": 0.5596214750308827,
715
+ "pass@16": 0.645575443790684,
716
+ "pass@32": 0.7144334545854256,
717
+ "pass@64": 0.768551512230959,
718
+ "pass@128": 0.8063583815028902
719
+ },
720
+ "18": {
721
+ "pass@1": 0.20683035714285714,
722
+ "pass@2": 0.31012197694038246,
723
+ "pass@4": 0.4170737739032621,
724
+ "pass@8": 0.5116615581807312,
725
+ "pass@16": 0.5892902093361223,
726
+ "pass@32": 0.6520943406851805,
727
+ "pass@64": 0.7067899408793984,
728
+ "pass@128": 0.76
729
+ },
730
+ "19": {
731
+ "pass@1": 0.17354072822822822,
732
+ "pass@2": 0.26094906422170194,
733
+ "pass@4": 0.358604284543261,
734
+ "pass@8": 0.456174177410017,
735
+ "pass@16": 0.5451820364907284,
736
+ "pass@32": 0.620654648238403,
737
+ "pass@64": 0.6820607217017262,
738
+ "pass@128": 0.7357357357357357
739
+ },
740
+ "20": {
741
+ "pass@1": 0.16500093843843844,
742
+ "pass@2": 0.2588242770329385,
743
+ "pass@4": 0.36199158973371576,
744
+ "pass@8": 0.4565102354746186,
745
+ "pass@16": 0.5352755572347072,
746
+ "pass@32": 0.6006625757883709,
747
+ "pass@64": 0.6604650436060762,
748
+ "pass@128": 0.7177177177177178
749
+ }
750
+ }
751
+ },
752
+ "teachers_in_school": {
753
+ "count": 430464,
754
+ "correct": 149525,
755
+ "answer_accuracy": 0.34735773490930716,
756
+ "avg_response_len": 341.8155943354148,
757
+ "resp_tokens_sum": 147139308,
758
+ "pass_at_k": {
759
+ "pass@1": 0.34735773490930716,
760
+ "pass@2": 0.4515589916085431,
761
+ "pass@4": 0.5514921892847711,
762
+ "pass@8": 0.6375586473392231,
763
+ "pass@16": 0.7058241555441483,
764
+ "pass@32": 0.7587062678349155,
765
+ "pass@64": 0.7997698004744774,
766
+ "pass@128": 0.8331846565566459
767
+ },
768
+ "per_op_pass_at_k": {
769
+ "11": {
770
+ "pass@1": 0.6095161897590361,
771
+ "pass@2": 0.69583760731904,
772
+ "pass@4": 0.7645579268705863,
773
+ "pass@8": 0.8162260764737186,
774
+ "pass@16": 0.8516898769534356,
775
+ "pass@32": 0.8782974399669164,
776
+ "pass@64": 0.9013430821206966,
777
+ "pass@128": 0.9216867469879518
778
+ },
779
+ "12": {
780
+ "pass@1": 0.5153311128526645,
781
+ "pass@2": 0.6049512656431268,
782
+ "pass@4": 0.6858902189538215,
783
+ "pass@8": 0.7578828775191587,
784
+ "pass@16": 0.8176346019227054,
785
+ "pass@32": 0.8628753895579362,
786
+ "pass@64": 0.8937528903345976,
787
+ "pass@128": 0.9153605015673981
788
+ },
789
+ "13": {
790
+ "pass@1": 0.4445684523809524,
791
+ "pass@2": 0.5486098474822998,
792
+ "pass@4": 0.6393596874445313,
793
+ "pass@8": 0.716317434796044,
794
+ "pass@16": 0.7785482587992962,
795
+ "pass@32": 0.8284156321914673,
796
+ "pass@64": 0.8666372329432644,
797
+ "pass@128": 0.8935574229691877
798
+ },
799
+ "14": {
800
+ "pass@1": 0.43891630116959063,
801
+ "pass@2": 0.5642393487820595,
802
+ "pass@4": 0.6743917816194029,
803
+ "pass@8": 0.7619551785897168,
804
+ "pass@16": 0.8262880821557116,
805
+ "pass@32": 0.8700014401361008,
806
+ "pass@64": 0.8976582359273326,
807
+ "pass@128": 0.9122807017543859
808
+ },
809
+ "15": {
810
+ "pass@1": 0.34739583333333335,
811
+ "pass@2": 0.47785917740396083,
812
+ "pass@4": 0.6037705800979423,
813
+ "pass@8": 0.7069009982792729,
814
+ "pass@16": 0.780389449448767,
815
+ "pass@32": 0.8276691872442753,
816
+ "pass@64": 0.8579498255115579,
817
+ "pass@128": 0.8818181818181818
818
+ },
819
+ "16": {
820
+ "pass@1": 0.32979433760683763,
821
+ "pass@2": 0.4497512730780443,
822
+ "pass@4": 0.5584297905603677,
823
+ "pass@8": 0.6485196575901173,
824
+ "pass@16": 0.7218197491529739,
825
+ "pass@32": 0.7783995418342657,
826
+ "pass@64": 0.8163805628573818,
827
+ "pass@128": 0.8433048433048433
828
+ },
829
+ "17": {
830
+ "pass@1": 0.25642954192546585,
831
+ "pass@2": 0.3702109417029389,
832
+ "pass@4": 0.4881588191421722,
833
+ "pass@8": 0.5911789755276028,
834
+ "pass@16": 0.6696642459629725,
835
+ "pass@32": 0.7277160161846408,
836
+ "pass@64": 0.7728749228117372,
837
+ "pass@128": 0.8105590062111802
838
+ },
839
+ "18": {
840
+ "pass@1": 0.2017233455882353,
841
+ "pass@2": 0.3029668683418246,
842
+ "pass@4": 0.4153720160899007,
843
+ "pass@8": 0.522158553810786,
844
+ "pass@16": 0.6110499248219567,
845
+ "pass@32": 0.6838685275214142,
846
+ "pass@64": 0.7456635518468799,
847
+ "pass@128": 0.7941176470588235
848
+ },
849
+ "19": {
850
+ "pass@1": 0.17510695684523808,
851
+ "pass@2": 0.2658245600159356,
852
+ "pass@4": 0.36414872164416945,
853
+ "pass@8": 0.4539296143435633,
854
+ "pass@16": 0.5288852495897007,
855
+ "pass@32": 0.5929693177682022,
856
+ "pass@64": 0.6520420203037799,
857
+ "pass@128": 0.7142857142857143
858
+ },
859
+ "20": {
860
+ "pass@1": 0.15629678143712575,
861
+ "pass@2": 0.23557437290772787,
862
+ "pass@4": 0.32072419053905704,
863
+ "pass@8": 0.40036653384062204,
864
+ "pass@16": 0.4718154181796683,
865
+ "pass@32": 0.5358254127504247,
866
+ "pass@64": 0.5920170158378041,
867
+ "pass@128": 0.6437125748502994
868
+ }
869
+ }
870
+ },
871
+ "movie_festival_awards": {
872
+ "count": 418816,
873
+ "correct": 140802,
874
+ "answer_accuracy": 0.3361905944376528,
875
+ "avg_response_len": 366.8818526512836,
876
+ "resp_tokens_sum": 153655990,
877
+ "pass_at_k": {
878
+ "pass@1": 0.3361905944376528,
879
+ "pass@2": 0.4436242386486146,
880
+ "pass@4": 0.5461406170786598,
881
+ "pass@8": 0.6323643460173062,
882
+ "pass@16": 0.6997056477688991,
883
+ "pass@32": 0.7533052582731945,
884
+ "pass@64": 0.7977045144536511,
885
+ "pass@128": 0.8325183374083129
886
+ },
887
+ "per_op_pass_at_k": {
888
+ "11": {
889
+ "pass@1": 0.5569940476190476,
890
+ "pass@2": 0.6512545306836639,
891
+ "pass@4": 0.7297863037953585,
892
+ "pass@8": 0.7881979214724417,
893
+ "pass@16": 0.8305159472405158,
894
+ "pass@32": 0.8664025710331352,
895
+ "pass@64": 0.8977197150773083,
896
+ "pass@128": 0.9174603174603174
897
+ },
898
+ "12": {
899
+ "pass@1": 0.5906441131498471,
900
+ "pass@2": 0.6903511702665612,
901
+ "pass@4": 0.7648428261260917,
902
+ "pass@8": 0.8219961349721057,
903
+ "pass@16": 0.8647988817303227,
904
+ "pass@32": 0.8948407161093591,
905
+ "pass@64": 0.9176319993654141,
906
+ "pass@128": 0.9357798165137615
907
+ },
908
+ "13": {
909
+ "pass@1": 0.47955246913580246,
910
+ "pass@2": 0.6025725733936037,
911
+ "pass@4": 0.7088902833905013,
912
+ "pass@8": 0.7899693991703378,
913
+ "pass@16": 0.8414343832139441,
914
+ "pass@32": 0.8704900063676707,
915
+ "pass@64": 0.8882142857644638,
916
+ "pass@128": 0.9012345679012346
917
+ },
918
+ "14": {
919
+ "pass@1": 0.3917145771513353,
920
+ "pass@2": 0.5191060246734736,
921
+ "pass@4": 0.6323926407381574,
922
+ "pass@8": 0.7208649834356172,
923
+ "pass@16": 0.7849817631994386,
924
+ "pass@32": 0.8320991882644259,
925
+ "pass@64": 0.8661309975696629,
926
+ "pass@128": 0.8902077151335311
927
+ },
928
+ "15": {
929
+ "pass@1": 0.33253761574074076,
930
+ "pass@2": 0.4562964785651788,
931
+ "pass@4": 0.5772275210390362,
932
+ "pass@8": 0.677586574182789,
933
+ "pass@16": 0.7481361489708472,
934
+ "pass@32": 0.7971674319182525,
935
+ "pass@64": 0.8354807477013398,
936
+ "pass@128": 0.8672839506172839
937
+ },
938
+ "16": {
939
+ "pass@1": 0.30212942477876104,
940
+ "pass@2": 0.4332392632336884,
941
+ "pass@4": 0.5611566145492871,
942
+ "pass@8": 0.6676743747616286,
943
+ "pass@16": 0.7497157527260427,
944
+ "pass@32": 0.8118684200882799,
945
+ "pass@64": 0.8597076780207225,
946
+ "pass@128": 0.8938053097345132
947
+ },
948
+ "17": {
949
+ "pass@1": 0.20277202560240964,
950
+ "pass@2": 0.30790929465895067,
951
+ "pass@4": 0.42129222909636277,
952
+ "pass@8": 0.5249290260578731,
953
+ "pass@16": 0.6136367096463329,
954
+ "pass@32": 0.6894706624655844,
955
+ "pass@64": 0.750886437329411,
956
+ "pass@128": 0.7951807228915663
957
+ },
958
+ "18": {
959
+ "pass@1": 0.19248991935483872,
960
+ "pass@2": 0.28834931419862814,
961
+ "pass@4": 0.3872474058686214,
962
+ "pass@8": 0.4755959602758333,
963
+ "pass@16": 0.5545506915364673,
964
+ "pass@32": 0.6301610647827953,
965
+ "pass@64": 0.6995870673914846,
966
+ "pass@128": 0.7516129032258064
967
+ },
968
+ "19": {
969
+ "pass@1": 0.16498300604229607,
970
+ "pass@2": 0.25099168589575854,
971
+ "pass@4": 0.3448424330266873,
972
+ "pass@8": 0.4322719262756728,
973
+ "pass@16": 0.5071666429847727,
974
+ "pass@32": 0.5715699397863766,
975
+ "pass@64": 0.634263631001721,
976
+ "pass@128": 0.7009063444108762
977
+ },
978
+ "20": {
979
+ "pass@1": 0.15697728978978978,
980
+ "pass@2": 0.24347046357380994,
981
+ "pass@4": 0.33896897037645063,
982
+ "pass@8": 0.4278276847043507,
983
+ "pass@16": 0.5039384810396262,
984
+ "pass@32": 0.5702084386681836,
985
+ "pass@64": 0.6286186409348357,
986
+ "pass@128": 0.6726726726726727
987
+ }
988
+ }
989
+ }
990
+ }
991
+ },
992
+ "total": {
993
+ "answer_accuracy": 0.5169925986842105,
994
+ "avg_loss": 0.0,
995
+ "count": 2432000,
996
+ "validation_metrics": {
997
+ "id": {},
998
+ "ood": {}
999
+ },
1000
+ "pass_at_k": {
1001
+ "pass@1": 0.5169925986842105,
1002
+ "pass@2": 0.597906612619119,
1003
+ "pass@4": 0.6713933917125223,
1004
+ "pass@8": 0.7332066924179007,
1005
+ "pass@16": 0.7823277362133715,
1006
+ "pass@32": 0.8209244747293224,
1007
+ "pass@64": 0.8516414616854432,
1008
+ "pass@128": 0.876421052631579
1009
+ },
1010
+ "per_op_pass_at_k": {
1011
+ "10": {
1012
+ "pass@1": 0.6372265625,
1013
+ "pass@2": 0.7061861466535435,
1014
+ "pass@4": 0.7634870098425202,
1015
+ "pass@8": 0.8101290041388484,
1016
+ "pass@16": 0.8469615673260908,
1017
+ "pass@32": 0.8759298968033682,
1018
+ "pass@64": 0.8982367360000915,
1019
+ "pass@128": 0.915
1020
+ },
1021
+ "2": {
1022
+ "pass@1": 0.9284296875,
1023
+ "pass@2": 0.9552456938976382,
1024
+ "pass@4": 0.9715452533745776,
1025
+ "pass@8": 0.9801859779187886,
1026
+ "pass@16": 0.9846789731901037,
1027
+ "pass@32": 0.9876162438902186,
1028
+ "pass@64": 0.9896641079037857,
1029
+ "pass@128": 0.991
1030
+ },
1031
+ "3": {
1032
+ "pass@1": 0.881828125,
1033
+ "pass@2": 0.9067679625984255,
1034
+ "pass@4": 0.9261030650543686,
1035
+ "pass@8": 0.9414119676703517,
1036
+ "pass@16": 0.9534031663493782,
1037
+ "pass@32": 0.9632878571158628,
1038
+ "pass@64": 0.9719825705210999,
1039
+ "pass@128": 0.979
1040
+ },
1041
+ "4": {
1042
+ "pass@1": 0.77528125,
1043
+ "pass@2": 0.8108379675196851,
1044
+ "pass@4": 0.8371120714285711,
1045
+ "pass@8": 0.859004820532304,
1046
+ "pass@16": 0.8781944005645403,
1047
+ "pass@32": 0.8952841256653082,
1048
+ "pass@64": 0.9091673998059235,
1049
+ "pass@128": 0.92
1050
+ },
1051
+ "5": {
1052
+ "pass@1": 0.6686484375,
1053
+ "pass@2": 0.7283939468503943,
1054
+ "pass@4": 0.7806958164604425,
1055
+ "pass@8": 0.8230992439889415,
1056
+ "pass@16": 0.8552869515145666,
1057
+ "pass@32": 0.8789656965059433,
1058
+ "pass@64": 0.896464829124427,
1059
+ "pass@128": 0.91
1060
+ },
1061
+ "6": {
1062
+ "pass@1": 0.6456640625,
1063
+ "pass@2": 0.6965431840551182,
1064
+ "pass@4": 0.7367055818335206,
1065
+ "pass@8": 0.7715006117837285,
1066
+ "pass@16": 0.8038770259652023,
1067
+ "pass@32": 0.833591385701847,
1068
+ "pass@64": 0.8590963035813479,
1069
+ "pass@128": 0.88
1070
+ },
1071
+ "7": {
1072
+ "pass@1": 0.63528125,
1073
+ "pass@2": 0.6974590305118106,
1074
+ "pass@4": 0.7489790861454824,
1075
+ "pass@8": 0.7920296413695324,
1076
+ "pass@16": 0.8276452842242492,
1077
+ "pass@32": 0.856389168515698,
1078
+ "pass@64": 0.8784913178107205,
1079
+ "pass@128": 0.896
1080
+ },
1081
+ "8": {
1082
+ "pass@1": 0.608125,
1083
+ "pass@2": 0.6794880659448826,
1084
+ "pass@4": 0.738280693194601,
1085
+ "pass@8": 0.787894891228803,
1086
+ "pass@16": 0.8287240309358392,
1087
+ "pass@32": 0.8602222897163663,
1088
+ "pass@64": 0.8840336104013778,
1089
+ "pass@128": 0.902
1090
+ },
1091
+ "9": {
1092
+ "pass@1": 0.6068125,
1093
+ "pass@2": 0.6851019931102368,
1094
+ "pass@4": 0.746159344863142,
1095
+ "pass@8": 0.7933346588325872,
1096
+ "pass@16": 0.830524071411799,
1097
+ "pass@32": 0.8604934402470777,
1098
+ "pass@64": 0.8857607569485678,
1099
+ "pass@128": 0.909
1100
+ },
1101
+ "11": {
1102
+ "pass@1": 0.5650390625,
1103
+ "pass@2": 0.6540514271653557,
1104
+ "pass@4": 0.7280725492125987,
1105
+ "pass@8": 0.7854012029674802,
1106
+ "pass@16": 0.8276654184151161,
1107
+ "pass@32": 0.8613735335387038,
1108
+ "pass@64": 0.8894017990938388,
1109
+ "pass@128": 0.909
1110
+ },
1111
+ "12": {
1112
+ "pass@1": 0.56540625,
1113
+ "pass@2": 0.6588655265748027,
1114
+ "pass@4": 0.7355278310836144,
1115
+ "pass@8": 0.7988205667858062,
1116
+ "pass@16": 0.847712530162352,
1117
+ "pass@32": 0.8827476609527442,
1118
+ "pass@64": 0.9082646513909002,
1119
+ "pass@128": 0.928
1120
+ },
1121
+ "13": {
1122
+ "pass@1": 0.46809375,
1123
+ "pass@2": 0.5840370324803149,
1124
+ "pass@4": 0.6835329081364819,
1125
+ "pass@8": 0.7621323021661622,
1126
+ "pass@16": 0.8188842777252926,
1127
+ "pass@32": 0.8582276277047699,
1128
+ "pass@64": 0.8857752572809072,
1129
+ "pass@128": 0.906
1130
+ },
1131
+ "14": {
1132
+ "pass@1": 0.421625,
1133
+ "pass@2": 0.5463829970472442,
1134
+ "pass@4": 0.6571138544244463,
1135
+ "pass@8": 0.7455900198919282,
1136
+ "pass@16": 0.8114067977817702,
1137
+ "pass@32": 0.8578387083464307,
1138
+ "pass@64": 0.8882131790112293,
1139
+ "pass@128": 0.907
1140
+ },
1141
+ "15": {
1142
+ "pass@1": 0.3358671875,
1143
+ "pass@2": 0.46225799704724446,
1144
+ "pass@4": 0.5836184609111363,
1145
+ "pass@8": 0.6835905005192744,
1146
+ "pass@16": 0.7555084305704233,
1147
+ "pass@32": 0.8032069738517583,
1148
+ "pass@64": 0.8366766426001925,
1149
+ "pass@128": 0.865
1150
+ },
1151
+ "16": {
1152
+ "pass@1": 0.320109375,
1153
+ "pass@2": 0.44445164862204795,
1154
+ "pass@4": 0.5615954588488937,
1155
+ "pass@8": 0.6587711388526714,
1156
+ "pass@16": 0.7343385762204016,
1157
+ "pass@32": 0.7918982958088162,
1158
+ "pass@64": 0.83495353589886,
1159
+ "pass@128": 0.867
1160
+ },
1161
+ "17": {
1162
+ "pass@1": 0.2281171875,
1163
+ "pass@2": 0.33797797736220475,
1164
+ "pass@4": 0.454242666854143,
1165
+ "pass@8": 0.5582650971317871,
1166
+ "pass@16": 0.6427283783542358,
1167
+ "pass@32": 0.710422792436585,
1168
+ "pass@64": 0.7640788455706569,
1169
+ "pass@128": 0.804
1170
+ },
1171
+ "18": {
1172
+ "pass@1": 0.2006484375,
1173
+ "pass@2": 0.30093971456692936,
1174
+ "pass@4": 0.4072490021559802,
1175
+ "pass@8": 0.5040502013444325,
1176
+ "pass@16": 0.5859192620834119,
1177
+ "pass@32": 0.6560982486797597,
1178
+ "pass@64": 0.7177740778270901,
1179
+ "pass@128": 0.769
1180
+ },
1181
+ "19": {
1182
+ "pass@1": 0.171234375,
1183
+ "pass@2": 0.259291338582677,
1184
+ "pass@4": 0.3559120425571802,
1185
+ "pass@8": 0.44750835909422065,
1186
+ "pass@16": 0.5271232208415118,
1187
+ "pass@32": 0.595105338702795,
1188
+ "pass@64": 0.6561536010103154,
1189
+ "pass@128": 0.717
1190
+ },
1191
+ "20": {
1192
+ "pass@1": 0.159421875,
1193
+ "pass@2": 0.24594598917322827,
1194
+ "pass@4": 0.34054174615673044,
1195
+ "pass@8": 0.42820694972236445,
1196
+ "pass@16": 0.503644624417362,
1197
+ "pass@32": 0.568865735672675,
1198
+ "pass@64": 0.6269985502419505,
1199
+ "pass@128": 0.678
1200
+ }
1201
+ },
1202
+ "per_op_accuracy": {
1203
+ "10": 0.6372265625,
1204
+ "2": 0.9284296875,
1205
+ "3": 0.881828125,
1206
+ "4": 0.77528125,
1207
+ "5": 0.6686484375,
1208
+ "6": 0.6456640625,
1209
+ "7": 0.63528125,
1210
+ "8": 0.608125,
1211
+ "9": 0.6068125,
1212
+ "11": 0.5650390625,
1213
+ "12": 0.56540625,
1214
+ "13": 0.46809375,
1215
+ "14": 0.421625,
1216
+ "15": 0.3358671875,
1217
+ "16": 0.320109375,
1218
+ "17": 0.2281171875,
1219
+ "18": 0.2006484375,
1220
+ "19": 0.171234375,
1221
+ "20": 0.159421875
1222
+ },
1223
+ "per_op_avg_response_len": {
1224
+ "10": 279.7796171875,
1225
+ "2": 133.83121875,
1226
+ "3": 153.0110625,
1227
+ "4": 163.8181328125,
1228
+ "5": 182.535953125,
1229
+ "6": 207.005859375,
1230
+ "7": 227.3503203125,
1231
+ "8": 247.654734375,
1232
+ "9": 263.4365859375,
1233
+ "11": 297.00659375,
1234
+ "12": 302.433140625,
1235
+ "13": 318.675875,
1236
+ "14": 342.1790859375,
1237
+ "15": 353.8288359375,
1238
+ "16": 356.4609140625,
1239
+ "17": 367.413265625,
1240
+ "18": 372.1854140625,
1241
+ "19": 371.895203125,
1242
+ "20": 380.6831484375
1243
+ },
1244
+ "per_op_avg_loss": {},
1245
+ "per_op_length": {
1246
+ "10": 279.7796171875,
1247
+ "2": 133.83121875,
1248
+ "3": 153.0110625,
1249
+ "4": 163.8181328125,
1250
+ "5": 182.535953125,
1251
+ "6": 207.005859375,
1252
+ "7": 227.3503203125,
1253
+ "8": 247.654734375,
1254
+ "9": 263.4365859375,
1255
+ "11": 297.00659375,
1256
+ "12": 302.433140625,
1257
+ "13": 318.675875,
1258
+ "14": 342.1790859375,
1259
+ "15": 353.8288359375,
1260
+ "16": 356.4609140625,
1261
+ "17": 367.413265625,
1262
+ "18": 372.1854140625,
1263
+ "19": 371.895203125,
1264
+ "20": 380.6831484375
1265
+ },
1266
+ "per_op_loss": {},
1267
+ "per_template": {
1268
+ "crazy_zootopia": {
1269
+ "count": 809728,
1270
+ "correct": 420142,
1271
+ "answer_accuracy": 0.5188680643376541,
1272
+ "avg_response_len": 268.15166697952895,
1273
+ "resp_tokens_sum": 217129913,
1274
+ "pass_at_k": {
1275
+ "pass@1": 0.5188680643376541,
1276
+ "pass@2": 0.5992609202802675,
1277
+ "pass@4": 0.6722506560642548,
1278
+ "pass@8": 0.7345034465487572,
1279
+ "pass@16": 0.7846585288237924,
1280
+ "pass@32": 0.8237347339594457,
1281
+ "pass@64": 0.8543135182380192,
1282
+ "pass@128": 0.879386658235852
1283
+ },
1284
+ "per_op_pass_at_k": {
1285
+ "10": {
1286
+ "pass@1": 0.6432291666666666,
1287
+ "pass@2": 0.7086545217522591,
1288
+ "pass@4": 0.7614043908228281,
1289
+ "pass@8": 0.8048343235263006,
1290
+ "pass@16": 0.8406351544122402,
1291
+ "pass@32": 0.8704496732524306,
1292
+ "pass@64": 0.8947665326243408,
1293
+ "pass@128": 0.9144542772861357
1294
+ },
1295
+ "2": {
1296
+ "pass@1": 0.9332408489096573,
1297
+ "pass@2": 0.9600331610125838,
1298
+ "pass@4": 0.9771553707655704,
1299
+ "pass@8": 0.9854254731466165,
1300
+ "pass@16": 0.988133712621007,
1301
+ "pass@32": 0.9893193713664016,
1302
+ "pass@64": 0.9902739585246382,
1303
+ "pass@128": 0.9906542056074766
1304
+ },
1305
+ "3": {
1306
+ "pass@1": 0.8757645259938838,
1307
+ "pass@2": 0.9027283645163623,
1308
+ "pass@4": 0.9244454405928614,
1309
+ "pass@8": 0.9431411848874264,
1310
+ "pass@16": 0.9582971577237228,
1311
+ "pass@32": 0.9709724633793967,
1312
+ "pass@64": 0.9820870411980793,
1313
+ "pass@128": 0.9908256880733946
1314
+ },
1315
+ "4": {
1316
+ "pass@1": 0.786328125,
1317
+ "pass@2": 0.811225086122047,
1318
+ "pass@4": 0.8296586022450316,
1319
+ "pass@8": 0.8480276124139998,
1320
+ "pass@16": 0.8670405071229179,
1321
+ "pass@32": 0.886580837586342,
1322
+ "pass@64": 0.9031976667014938,
1323
+ "pass@128": 0.915625
1324
+ },
1325
+ "5": {
1326
+ "pass@1": 0.6481150793650794,
1327
+ "pass@2": 0.7089566929133857,
1328
+ "pass@4": 0.7631540902625267,
1329
+ "pass@8": 0.8083630689659272,
1330
+ "pass@16": 0.8431150398714341,
1331
+ "pass@32": 0.8683170445108382,
1332
+ "pass@64": 0.8869570995314593,
1333
+ "pass@128": 0.9015873015873016
1334
+ },
1335
+ "6": {
1336
+ "pass@1": 0.6656296781437125,
1337
+ "pass@2": 0.7128238601537084,
1338
+ "pass@4": 0.7472944426108409,
1339
+ "pass@8": 0.777625115201791,
1340
+ "pass@16": 0.8070151119011811,
1341
+ "pass@32": 0.8349043396850152,
1342
+ "pass@64": 0.8587159112471352,
1343
+ "pass@128": 0.8772455089820359
1344
+ },
1345
+ "7": {
1346
+ "pass@1": 0.6652875369822485,
1347
+ "pass@2": 0.7235667922704184,
1348
+ "pass@4": 0.77103220131877,
1349
+ "pass@8": 0.8127494418381218,
1350
+ "pass@16": 0.8505656587294267,
1351
+ "pass@32": 0.8821587788982282,
1352
+ "pass@64": 0.9041908255934645,
1353
+ "pass@128": 0.9201183431952663
1354
+ },
1355
+ "8": {
1356
+ "pass@1": 0.6062199519230769,
1357
+ "pass@2": 0.6758885082273367,
1358
+ "pass@4": 0.7324884194163228,
1359
+ "pass@8": 0.7816419557403208,
1360
+ "pass@16": 0.8223828369246351,
1361
+ "pass@32": 0.852237918950051,
1362
+ "pass@64": 0.8736887652975474,
1363
+ "pass@128": 0.8910256410256411
1364
+ },
1365
+ "9": {
1366
+ "pass@1": 0.6165272887323944,
1367
+ "pass@2": 0.6952627675501827,
1368
+ "pass@4": 0.7554617862027808,
1369
+ "pass@8": 0.8041103178637404,
1370
+ "pass@16": 0.8443959343789783,
1371
+ "pass@32": 0.8765953737508427,
1372
+ "pass@64": 0.9015005102566016,
1373
+ "pass@128": 0.923943661971831
1374
+ },
1375
+ "11": {
1376
+ "pass@1": 0.5303868626062322,
1377
+ "pass@2": 0.6172469245611297,
1378
+ "pass@4": 0.6922284186856246,
1379
+ "pass@8": 0.7539144484826834,
1380
+ "pass@16": 0.8025265039258181,
1381
+ "pass@32": 0.8409687637230883,
1382
+ "pass@64": 0.8707484010776608,
1383
+ "pass@128": 0.8895184135977338
1384
+ },
1385
+ "12": {
1386
+ "pass@1": 0.5872175141242938,
1387
+ "pass@2": 0.6783650851906224,
1388
+ "pass@4": 0.753178664107664,
1389
+ "pass@8": 0.8143028042975027,
1390
+ "pass@16": 0.859033496675691,
1391
+ "pass@32": 0.8894844562598956,
1392
+ "pass@64": 0.9126887445809961,
1393
+ "pass@128": 0.9322033898305084
1394
+ },
1395
+ "13": {
1396
+ "pass@1": 0.48278311128526646,
1397
+ "pass@2": 0.6048583170834043,
1398
+ "pass@4": 0.7072133163017632,
1399
+ "pass@8": 0.7851313248049716,
1400
+ "pass@16": 0.8411216588483615,
1401
+ "pass@32": 0.8791363164554562,
1402
+ "pass@64": 0.9047157885030569,
1403
+ "pass@128": 0.9247648902821317
1404
+ },
1405
+ "14": {
1406
+ "pass@1": 0.43460377725856697,
1407
+ "pass@2": 0.555995045011897,
1408
+ "pass@4": 0.6646590192581069,
1409
+ "pass@8": 0.7541116803627453,
1410
+ "pass@16": 0.8232941417018945,
1411
+ "pass@32": 0.8719027706379187,
1412
+ "pass@64": 0.9013330720969001,
1413
+ "pass@128": 0.9190031152647975
1414
+ },
1415
+ "15": {
1416
+ "pass@1": 0.3279895231213873,
1417
+ "pass@2": 0.45296072095034384,
1418
+ "pass@4": 0.5703828111623328,
1419
+ "pass@8": 0.6669801186470863,
1420
+ "pass@16": 0.7386815028490659,
1421
+ "pass@32": 0.7855314858948964,
1422
+ "pass@64": 0.8175070460293162,
1423
+ "pass@128": 0.846820809248555
1424
+ },
1425
+ "16": {
1426
+ "pass@1": 0.3288054435483871,
1427
+ "pass@2": 0.4507123920497839,
1428
+ "pass@4": 0.5656597097741817,
1429
+ "pass@8": 0.6606423419172535,
1430
+ "pass@16": 0.7316974325599377,
1431
+ "pass@32": 0.785344071661488,
1432
+ "pass@64": 0.828913082215784,
1433
+ "pass@128": 0.864516129032258
1434
+ },
1435
+ "17": {
1436
+ "pass@1": 0.2260883309248555,
1437
+ "pass@2": 0.33683285637886257,
1438
+ "pass@4": 0.4542962630935586,
1439
+ "pass@8": 0.5596214750308827,
1440
+ "pass@16": 0.645575443790684,
1441
+ "pass@32": 0.7144334545854256,
1442
+ "pass@64": 0.768551512230959,
1443
+ "pass@128": 0.8063583815028902
1444
+ },
1445
+ "18": {
1446
+ "pass@1": 0.20683035714285714,
1447
+ "pass@2": 0.31012197694038246,
1448
+ "pass@4": 0.4170737739032621,
1449
+ "pass@8": 0.5116615581807312,
1450
+ "pass@16": 0.5892902093361223,
1451
+ "pass@32": 0.6520943406851805,
1452
+ "pass@64": 0.7067899408793984,
1453
+ "pass@128": 0.76
1454
+ },
1455
+ "19": {
1456
+ "pass@1": 0.17354072822822822,
1457
+ "pass@2": 0.26094906422170194,
1458
+ "pass@4": 0.358604284543261,
1459
+ "pass@8": 0.456174177410017,
1460
+ "pass@16": 0.5451820364907284,
1461
+ "pass@32": 0.620654648238403,
1462
+ "pass@64": 0.6820607217017262,
1463
+ "pass@128": 0.7357357357357357
1464
+ },
1465
+ "20": {
1466
+ "pass@1": 0.16500093843843844,
1467
+ "pass@2": 0.2588242770329385,
1468
+ "pass@4": 0.36199158973371576,
1469
+ "pass@8": 0.4565102354746186,
1470
+ "pass@16": 0.5352755572347072,
1471
+ "pass@32": 0.6006625757883709,
1472
+ "pass@64": 0.6604650436060762,
1473
+ "pass@128": 0.7177177177177178
1474
+ }
1475
+ }
1476
+ },
1477
+ "teachers_in_school": {
1478
+ "count": 821120,
1479
+ "correct": 426542,
1480
+ "answer_accuracy": 0.5194636593920499,
1481
+ "avg_response_len": 275.8639187938426,
1482
+ "resp_tokens_sum": 226517381,
1483
+ "pass_at_k": {
1484
+ "pass@1": 0.5194636593920499,
1485
+ "pass@2": 0.5985632069276694,
1486
+ "pass@4": 0.6708152881951757,
1487
+ "pass@8": 0.7321737071047528,
1488
+ "pass@16": 0.7813725088235535,
1489
+ "pass@32": 0.8202247612977291,
1490
+ "pass@64": 0.8508530729343532,
1491
+ "pass@128": 0.8756040530007794
1492
+ },
1493
+ "per_op_pass_at_k": {
1494
+ "10": {
1495
+ "pass@1": 0.6664589723926381,
1496
+ "pass@2": 0.7259352658084149,
1497
+ "pass@4": 0.7729965696389173,
1498
+ "pass@8": 0.8135385076355918,
1499
+ "pass@16": 0.8485919842569918,
1500
+ "pass@32": 0.8769118448464734,
1501
+ "pass@64": 0.8981314161379662,
1502
+ "pass@128": 0.9141104294478528
1503
+ },
1504
+ "2": {
1505
+ "pass@1": 0.9396846504559271,
1506
+ "pass@2": 0.9592216032836322,
1507
+ "pass@4": 0.9723210259811746,
1508
+ "pass@8": 0.9800715101221468,
1509
+ "pass@16": 0.9846143152197563,
1510
+ "pass@32": 0.9880899912285871,
1511
+ "pass@64": 0.9903714652039789,
1512
+ "pass@128": 0.9908814589665653
1513
+ },
1514
+ "3": {
1515
+ "pass@1": 0.8826467803030303,
1516
+ "pass@2": 0.9058510796945836,
1517
+ "pass@4": 0.9239027323289134,
1518
+ "pass@8": 0.9375980631196633,
1519
+ "pass@16": 0.9490507863390486,
1520
+ "pass@32": 0.959496338875189,
1521
+ "pass@64": 0.9681067897370846,
1522
+ "pass@128": 0.9727272727272728
1523
+ },
1524
+ "4": {
1525
+ "pass@1": 0.7678439349112426,
1526
+ "pass@2": 0.8053503412850018,
1527
+ "pass@4": 0.8354628671600961,
1528
+ "pass@8": 0.8625236702397913,
1529
+ "pass@16": 0.8866633816803322,
1530
+ "pass@32": 0.9069734335131655,
1531
+ "pass@64": 0.9228128772044121,
1532
+ "pass@128": 0.9349112426035503
1533
+ },
1534
+ "5": {
1535
+ "pass@1": 0.6854166666666667,
1536
+ "pass@2": 0.741926280954011,
1537
+ "pass@4": 0.7908239029903872,
1538
+ "pass@8": 0.8307448053320945,
1539
+ "pass@16": 0.8603277390818179,
1540
+ "pass@32": 0.881465096873712,
1541
+ "pass@64": 0.897257332752116,
1542
+ "pass@128": 0.9101449275362319
1543
+ },
1544
+ "6": {
1545
+ "pass@1": 0.6161130014326648,
1546
+ "pass@2": 0.6660646701486812,
1547
+ "pass@4": 0.7085057325391633,
1548
+ "pass@8": 0.7456869974381682,
1549
+ "pass@16": 0.7804926677467746,
1550
+ "pass@32": 0.8129104072803103,
1551
+ "pass@64": 0.8413385287286547,
1552
+ "pass@128": 0.8653295128939829
1553
+ },
1554
+ "7": {
1555
+ "pass@1": 0.6342124277456648,
1556
+ "pass@2": 0.6974431494925126,
1557
+ "pass@4": 0.7495586211044426,
1558
+ "pass@8": 0.7918144315944411,
1559
+ "pass@16": 0.8253342442369711,
1560
+ "pass@32": 0.8523804182147894,
1561
+ "pass@64": 0.8755104389451583,
1562
+ "pass@128": 0.8959537572254336
1563
+ },
1564
+ "8": {
1565
+ "pass@1": 0.6068882042253522,
1566
+ "pass@2": 0.678888834978374,
1567
+ "pass@4": 0.7377912274169952,
1568
+ "pass@8": 0.7862378604682415,
1569
+ "pass@16": 0.8278968421686421,
1570
+ "pass@32": 0.8634145607393767,
1571
+ "pass@64": 0.8914266288458986,
1572
+ "pass@128": 0.9098591549295775
1573
+ },
1574
+ "9": {
1575
+ "pass@1": 0.6005800898203593,
1576
+ "pass@2": 0.6798503141355087,
1577
+ "pass@4": 0.7427577106753871,
1578
+ "pass@8": 0.7911772594066232,
1579
+ "pass@16": 0.8284008725461969,
1580
+ "pass@32": 0.8587425143072703,
1581
+ "pass@64": 0.8862981605259711,
1582
+ "pass@128": 0.9131736526946108
1583
+ },
1584
+ "11": {
1585
+ "pass@1": 0.6095161897590361,
1586
+ "pass@2": 0.69583760731904,
1587
+ "pass@4": 0.7645579268705863,
1588
+ "pass@8": 0.8162260764737186,
1589
+ "pass@16": 0.8516898769534356,
1590
+ "pass@32": 0.8782974399669164,
1591
+ "pass@64": 0.9013430821206966,
1592
+ "pass@128": 0.9216867469879518
1593
+ },
1594
+ "12": {
1595
+ "pass@1": 0.5153311128526645,
1596
+ "pass@2": 0.6049512656431268,
1597
+ "pass@4": 0.6858902189538215,
1598
+ "pass@8": 0.7578828775191587,
1599
+ "pass@16": 0.8176346019227054,
1600
+ "pass@32": 0.8628753895579362,
1601
+ "pass@64": 0.8937528903345976,
1602
+ "pass@128": 0.9153605015673981
1603
+ },
1604
+ "13": {
1605
+ "pass@1": 0.4445684523809524,
1606
+ "pass@2": 0.5486098474822998,
1607
+ "pass@4": 0.6393596874445313,
1608
+ "pass@8": 0.716317434796044,
1609
+ "pass@16": 0.7785482587992962,
1610
+ "pass@32": 0.8284156321914673,
1611
+ "pass@64": 0.8666372329432644,
1612
+ "pass@128": 0.8935574229691877
1613
+ },
1614
+ "14": {
1615
+ "pass@1": 0.43891630116959063,
1616
+ "pass@2": 0.5642393487820595,
1617
+ "pass@4": 0.6743917816194029,
1618
+ "pass@8": 0.7619551785897168,
1619
+ "pass@16": 0.8262880821557116,
1620
+ "pass@32": 0.8700014401361008,
1621
+ "pass@64": 0.8976582359273326,
1622
+ "pass@128": 0.9122807017543859
1623
+ },
1624
+ "15": {
1625
+ "pass@1": 0.34739583333333335,
1626
+ "pass@2": 0.47785917740396083,
1627
+ "pass@4": 0.6037705800979423,
1628
+ "pass@8": 0.7069009982792729,
1629
+ "pass@16": 0.780389449448767,
1630
+ "pass@32": 0.8276691872442753,
1631
+ "pass@64": 0.8579498255115579,
1632
+ "pass@128": 0.8818181818181818
1633
+ },
1634
+ "16": {
1635
+ "pass@1": 0.32979433760683763,
1636
+ "pass@2": 0.4497512730780443,
1637
+ "pass@4": 0.5584297905603677,
1638
+ "pass@8": 0.6485196575901173,
1639
+ "pass@16": 0.7218197491529739,
1640
+ "pass@32": 0.7783995418342657,
1641
+ "pass@64": 0.8163805628573818,
1642
+ "pass@128": 0.8433048433048433
1643
+ },
1644
+ "17": {
1645
+ "pass@1": 0.25642954192546585,
1646
+ "pass@2": 0.3702109417029389,
1647
+ "pass@4": 0.4881588191421722,
1648
+ "pass@8": 0.5911789755276028,
1649
+ "pass@16": 0.6696642459629725,
1650
+ "pass@32": 0.7277160161846408,
1651
+ "pass@64": 0.7728749228117372,
1652
+ "pass@128": 0.8105590062111802
1653
+ },
1654
+ "18": {
1655
+ "pass@1": 0.2017233455882353,
1656
+ "pass@2": 0.3029668683418246,
1657
+ "pass@4": 0.4153720160899007,
1658
+ "pass@8": 0.522158553810786,
1659
+ "pass@16": 0.6110499248219567,
1660
+ "pass@32": 0.6838685275214142,
1661
+ "pass@64": 0.7456635518468799,
1662
+ "pass@128": 0.7941176470588235
1663
+ },
1664
+ "19": {
1665
+ "pass@1": 0.17510695684523808,
1666
+ "pass@2": 0.2658245600159356,
1667
+ "pass@4": 0.36414872164416945,
1668
+ "pass@8": 0.4539296143435633,
1669
+ "pass@16": 0.5288852495897007,
1670
+ "pass@32": 0.5929693177682022,
1671
+ "pass@64": 0.6520420203037799,
1672
+ "pass@128": 0.7142857142857143
1673
+ },
1674
+ "20": {
1675
+ "pass@1": 0.15629678143712575,
1676
+ "pass@2": 0.23557437290772787,
1677
+ "pass@4": 0.32072419053905704,
1678
+ "pass@8": 0.40036653384062204,
1679
+ "pass@16": 0.4718154181796683,
1680
+ "pass@32": 0.5358254127504247,
1681
+ "pass@64": 0.5920170158378041,
1682
+ "pass@128": 0.6437125748502994
1683
+ }
1684
+ }
1685
+ },
1686
+ "movie_festival_awards": {
1687
+ "count": 801152,
1688
+ "correct": 410642,
1689
+ "answer_accuracy": 0.5125644072535549,
1690
+ "avg_response_len": 296.403654986819,
1691
+ "resp_tokens_sum": 237464381,
1692
+ "pass_at_k": {
1693
+ "pass@1": 0.5125644072535549,
1694
+ "pass@2": 0.5958648483192146,
1695
+ "pass@4": 0.671119462934007,
1696
+ "pass@8": 0.7329547886236961,
1697
+ "pass@16": 0.7809510290160512,
1698
+ "pass@32": 0.8188012860048482,
1699
+ "pass@64": 0.8497488405137779,
1700
+ "pass@128": 0.8742610640677424
1701
+ },
1702
+ "per_op_pass_at_k": {
1703
+ "10": {
1704
+ "pass@1": 0.602705223880597,
1705
+ "pass@2": 0.6844697526148781,
1706
+ "pass@4": 0.7563404168695326,
1707
+ "pass@8": 0.8121689999230759,
1708
+ "pass@16": 0.8517769048136232,
1709
+ "pass@32": 0.8805199885099825,
1710
+ "pass@64": 0.9018508650133183,
1711
+ "pass@128": 0.9164179104477612
1712
+ },
1713
+ "2": {
1714
+ "pass@1": 0.9134375,
1715
+ "pass@2": 0.9471175478065238,
1716
+ "pass@4": 0.9656707480314957,
1717
+ "pass@8": 0.9754882120243954,
1718
+ "pass@16": 0.9815712620898884,
1719
+ "pass@32": 0.985608910192567,
1720
+ "pass@64": 0.9884398719007649,
1721
+ "pass@128": 0.9914285714285714
1722
+ },
1723
+ "3": {
1724
+ "pass@1": 0.8868212463556852,
1725
+ "pass@2": 0.9115012568581995,
1726
+ "pass@4": 0.9298003041165043,
1727
+ "pass@8": 0.9434327678795174,
1728
+ "pass@16": 0.9529248871190527,
1729
+ "pass@32": 0.9596095328337845,
1730
+ "pass@64": 0.966078330717463,
1731
+ "pass@128": 0.9737609329446064
1732
+ },
1733
+ "4": {
1734
+ "pass@1": 0.7722953216374269,
1735
+ "pass@2": 0.8158991947552606,
1736
+ "pass@4": 0.8457159930118383,
1737
+ "pass@8": 0.8657981988853063,
1738
+ "pass@16": 0.8802608633837845,
1739
+ "pass@32": 0.8918749623106107,
1740
+ "pass@64": 0.901267233819749,
1741
+ "pass@128": 0.9093567251461988
1742
+ },
1743
+ "5": {
1744
+ "pass@1": 0.6706571691176471,
1745
+ "pass@2": 0.7326706519221858,
1746
+ "pass@4": 0.7866706808707734,
1747
+ "pass@8": 0.8289938806620649,
1748
+ "pass@16": 0.8614489528583463,
1749
+ "pass@32": 0.8862952031282322,
1750
+ "pass@64": 0.9044693028604022,
1751
+ "pass@128": 0.9176470588235294
1752
+ },
1753
+ "6": {
1754
+ "pass@1": 0.6571618690851735,
1755
+ "pass@2": 0.7129446210785165,
1756
+ "pass@4": 0.7565953859474185,
1757
+ "pass@8": 0.7934670700328363,
1758
+ "pass@16": 0.8263155758567313,
1759
+ "pass@32": 0.8549766691048055,
1760
+ "pass@64": 0.8790474533075844,
1761
+ "pass@128": 0.8990536277602523
1762
+ },
1763
+ "7": {
1764
+ "pass@1": 0.604356210443038,
1765
+ "pass@2": 0.6695510284810123,
1766
+ "pass@4": 0.724756073410127,
1767
+ "pass@8": 0.7701029642296547,
1768
+ "pass@16": 0.8056596299611132,
1769
+ "pass@32": 0.8332147991955038,
1770
+ "pass@64": 0.8542664774845095,
1771
+ "pass@128": 0.870253164556962
1772
+ },
1773
+ "8": {
1774
+ "pass@1": 0.6112284159159159,
1775
+ "pass@2": 0.6834994443262159,
1776
+ "pass@4": 0.7442294913023253,
1777
+ "pass@8": 0.7955200017165065,
1778
+ "pass@16": 0.835547167704164,
1779
+ "pass@32": 0.8642999697942094,
1780
+ "pass@64": 0.885844631796542,
1781
+ "pass@128": 0.9039039039039038
1782
+ },
1783
+ "9": {
1784
+ "pass@1": 0.6024165996784566,
1785
+ "pass@2": 0.6791437482593614,
1786
+ "pass@4": 0.7391940044873812,
1787
+ "pass@8": 0.7833514191290923,
1788
+ "pass@16": 0.8169698497647327,
1789
+ "pass@32": 0.8439938353276553,
1790
+ "pass@64": 0.8672170102630214,
1791
+ "pass@128": 0.887459807073955
1792
+ },
1793
+ "11": {
1794
+ "pass@1": 0.5569940476190476,
1795
+ "pass@2": 0.6512545306836639,
1796
+ "pass@4": 0.7297863037953585,
1797
+ "pass@8": 0.7881979214724417,
1798
+ "pass@16": 0.8305159472405158,
1799
+ "pass@32": 0.8664025710331352,
1800
+ "pass@64": 0.8977197150773083,
1801
+ "pass@128": 0.9174603174603174
1802
+ },
1803
+ "12": {
1804
+ "pass@1": 0.5906441131498471,
1805
+ "pass@2": 0.6903511702665612,
1806
+ "pass@4": 0.7648428261260917,
1807
+ "pass@8": 0.8219961349721057,
1808
+ "pass@16": 0.8647988817303227,
1809
+ "pass@32": 0.8948407161093591,
1810
+ "pass@64": 0.9176319993654141,
1811
+ "pass@128": 0.9357798165137615
1812
+ },
1813
+ "13": {
1814
+ "pass@1": 0.47955246913580246,
1815
+ "pass@2": 0.6025725733936037,
1816
+ "pass@4": 0.7088902833905013,
1817
+ "pass@8": 0.7899693991703378,
1818
+ "pass@16": 0.8414343832139441,
1819
+ "pass@32": 0.8704900063676707,
1820
+ "pass@64": 0.8882142857644638,
1821
+ "pass@128": 0.9012345679012346
1822
+ },
1823
+ "14": {
1824
+ "pass@1": 0.3917145771513353,
1825
+ "pass@2": 0.5191060246734736,
1826
+ "pass@4": 0.6323926407381574,
1827
+ "pass@8": 0.7208649834356172,
1828
+ "pass@16": 0.7849817631994386,
1829
+ "pass@32": 0.8320991882644259,
1830
+ "pass@64": 0.8661309975696629,
1831
+ "pass@128": 0.8902077151335311
1832
+ },
1833
+ "15": {
1834
+ "pass@1": 0.33253761574074076,
1835
+ "pass@2": 0.4562964785651788,
1836
+ "pass@4": 0.5772275210390362,
1837
+ "pass@8": 0.677586574182789,
1838
+ "pass@16": 0.7481361489708472,
1839
+ "pass@32": 0.7971674319182525,
1840
+ "pass@64": 0.8354807477013398,
1841
+ "pass@128": 0.8672839506172839
1842
+ },
1843
+ "16": {
1844
+ "pass@1": 0.30212942477876104,
1845
+ "pass@2": 0.4332392632336884,
1846
+ "pass@4": 0.5611566145492871,
1847
+ "pass@8": 0.6676743747616286,
1848
+ "pass@16": 0.7497157527260427,
1849
+ "pass@32": 0.8118684200882799,
1850
+ "pass@64": 0.8597076780207225,
1851
+ "pass@128": 0.8938053097345132
1852
+ },
1853
+ "17": {
1854
+ "pass@1": 0.20277202560240964,
1855
+ "pass@2": 0.30790929465895067,
1856
+ "pass@4": 0.42129222909636277,
1857
+ "pass@8": 0.5249290260578731,
1858
+ "pass@16": 0.6136367096463329,
1859
+ "pass@32": 0.6894706624655844,
1860
+ "pass@64": 0.750886437329411,
1861
+ "pass@128": 0.7951807228915663
1862
+ },
1863
+ "18": {
1864
+ "pass@1": 0.19248991935483872,
1865
+ "pass@2": 0.28834931419862814,
1866
+ "pass@4": 0.3872474058686214,
1867
+ "pass@8": 0.4755959602758333,
1868
+ "pass@16": 0.5545506915364673,
1869
+ "pass@32": 0.6301610647827953,
1870
+ "pass@64": 0.6995870673914846,
1871
+ "pass@128": 0.7516129032258064
1872
+ },
1873
+ "19": {
1874
+ "pass@1": 0.16498300604229607,
1875
+ "pass@2": 0.25099168589575854,
1876
+ "pass@4": 0.3448424330266873,
1877
+ "pass@8": 0.4322719262756728,
1878
+ "pass@16": 0.5071666429847727,
1879
+ "pass@32": 0.5715699397863766,
1880
+ "pass@64": 0.634263631001721,
1881
+ "pass@128": 0.7009063444108762
1882
+ },
1883
+ "20": {
1884
+ "pass@1": 0.15697728978978978,
1885
+ "pass@2": 0.24347046357380994,
1886
+ "pass@4": 0.33896897037645063,
1887
+ "pass@8": 0.4278276847043507,
1888
+ "pass@16": 0.5039384810396262,
1889
+ "pass@32": 0.5702084386681836,
1890
+ "pass@64": 0.6286186409348357,
1891
+ "pass@128": 0.6726726726726727
1892
+ }
1893
+ }
1894
+ }
1895
+ }
1896
+ }
1897
+ }