ZachW commited on
Commit
610329e
·
verified ·
1 Parent(s): b53fab2

Upload extrapolation artifact: 050302 OPD Ref-KL OP13-20 step200 metrics

Browse files
050302/evals/opd050302_refkl_s200_merged_hf/opd050302_refkl_s200_merged_hf_metrics.json ADDED
@@ -0,0 +1,1897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "opd050302_refkl_s200_merged_hf",
3
+ "id": {
4
+ "answer_accuracy": 0.7093480902777778,
5
+ "avg_response_len": 206.4830920138889,
6
+ "avg_loss": 0.0,
7
+ "count": 1152000,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {
10
+ "pass@1": 0.7093480902777778,
11
+ "pass@2": 0.7624876284995514,
12
+ "pass@4": 0.8052020920614034,
13
+ "pass@8": 0.8398279391270769,
14
+ "pass@16": 0.8679966339442259,
15
+ "pass@32": 0.8909792672260827,
16
+ "pass@64": 0.9095811254482696,
17
+ "pass@128": 0.925
18
+ },
19
+ "per_op_pass_at_k": {
20
+ "10": {
21
+ "pass@1": 0.6380078125,
22
+ "pass@2": 0.7071900836614177,
23
+ "pass@4": 0.7646811167979007,
24
+ "pass@8": 0.8108247991830362,
25
+ "pass@16": 0.8470848940620732,
26
+ "pass@32": 0.8769506701709128,
27
+ "pass@64": 0.901155896698448,
28
+ "pass@128": 0.919
29
+ },
30
+ "2": {
31
+ "pass@1": 0.928890625,
32
+ "pass@2": 0.9555078740157483,
33
+ "pass@4": 0.9714628627671541,
34
+ "pass@8": 0.9799992080126586,
35
+ "pass@16": 0.9846548615385947,
36
+ "pass@32": 0.9879458575670311,
37
+ "pass@64": 0.9906994511169885,
38
+ "pass@128": 0.993
39
+ },
40
+ "3": {
41
+ "pass@1": 0.881953125,
42
+ "pass@2": 0.9067508612204728,
43
+ "pass@4": 0.9258298205849268,
44
+ "pass@8": 0.9409859160859129,
45
+ "pass@16": 0.9533466751882298,
46
+ "pass@32": 0.9633973208452283,
47
+ "pass@64": 0.971630684690094,
48
+ "pass@128": 0.979
49
+ },
50
+ "4": {
51
+ "pass@1": 0.7756484375,
52
+ "pass@2": 0.811566067913386,
53
+ "pass@4": 0.8379122627484064,
54
+ "pass@8": 0.8596563517143947,
55
+ "pass@16": 0.8789873759555055,
56
+ "pass@32": 0.8969012408477046,
57
+ "pass@64": 0.9125268085444034,
58
+ "pass@128": 0.927
59
+ },
60
+ "5": {
61
+ "pass@1": 0.667046875,
62
+ "pass@2": 0.726266732283465,
63
+ "pass@4": 0.778575683633296,
64
+ "pass@8": 0.8219898978294717,
65
+ "pass@16": 0.8555899647715695,
66
+ "pass@32": 0.8806531787516202,
67
+ "pass@64": 0.8989304221905279,
68
+ "pass@128": 0.912
69
+ },
70
+ "6": {
71
+ "pass@1": 0.644078125,
72
+ "pass@2": 0.6951700295275594,
73
+ "pass@4": 0.7363283152418446,
74
+ "pass@8": 0.772404418920689,
75
+ "pass@16": 0.8058868562027328,
76
+ "pass@32": 0.8366017115916559,
77
+ "pass@64": 0.863674735933478,
78
+ "pass@128": 0.888
79
+ },
80
+ "7": {
81
+ "pass@1": 0.6341484375,
82
+ "pass@2": 0.695350270669291,
83
+ "pass@4": 0.7460603857330329,
84
+ "pass@8": 0.7880873637047299,
85
+ "pass@16": 0.8229143209349995,
86
+ "pass@32": 0.8514165197895321,
87
+ "pass@64": 0.8754540936668659,
88
+ "pass@128": 0.899
89
+ },
90
+ "8": {
91
+ "pass@1": 0.607140625,
92
+ "pass@2": 0.6786822096456687,
93
+ "pass@4": 0.7383630651481062,
94
+ "pass@8": 0.7885355189101158,
95
+ "pass@16": 0.8293463979004868,
96
+ "pass@32": 0.8604975652604689,
97
+ "pass@64": 0.8824759274923865,
98
+ "pass@128": 0.897
99
+ },
100
+ "9": {
101
+ "pass@1": 0.60721875,
102
+ "pass@2": 0.6859045275590553,
103
+ "pass@4": 0.7476053158980122,
104
+ "pass@8": 0.7959679777827171,
105
+ "pass@16": 0.8341583589437844,
106
+ "pass@32": 0.864449340210571,
107
+ "pass@64": 0.889682108701199,
108
+ "pass@128": 0.911
109
+ }
110
+ },
111
+ "per_op_accuracy": {
112
+ "10": 0.6380078125,
113
+ "2": 0.928890625,
114
+ "3": 0.881953125,
115
+ "4": 0.7756484375,
116
+ "5": 0.667046875,
117
+ "6": 0.644078125,
118
+ "7": 0.6341484375,
119
+ "8": 0.607140625,
120
+ "9": 0.60721875
121
+ },
122
+ "per_op_avg_response_len": {
123
+ "10": 279.7841328125,
124
+ "2": 133.7599609375,
125
+ "3": 152.97484375,
126
+ "4": 163.8752578125,
127
+ "5": 182.675140625,
128
+ "6": 207.0015078125,
129
+ "7": 227.182375,
130
+ "8": 247.6363125,
131
+ "9": 263.458296875
132
+ },
133
+ "per_op_avg_loss": {},
134
+ "per_op_length": {
135
+ "10": 279.7841328125,
136
+ "2": 133.7599609375,
137
+ "3": 152.97484375,
138
+ "4": 163.8752578125,
139
+ "5": 182.675140625,
140
+ "6": 207.0015078125,
141
+ "7": 227.182375,
142
+ "8": 247.6363125,
143
+ "9": 263.458296875
144
+ },
145
+ "per_op_loss": {},
146
+ "per_template": {
147
+ "crazy_zootopia": {
148
+ "count": 379008,
149
+ "correct": 270533,
150
+ "answer_accuracy": 0.7137923210064168,
151
+ "avg_response_len": 197.06139975937182,
152
+ "resp_tokens_sum": 74687847,
153
+ "pass_at_k": {
154
+ "pass@1": 0.7137923210064168,
155
+ "pass@2": 0.7649061320792365,
156
+ "pass@4": 0.8056184499543949,
157
+ "pass@8": 0.8399269260118238,
158
+ "pass@16": 0.8687010580139358,
159
+ "pass@32": 0.8919768507604368,
160
+ "pass@64": 0.9102056120327188,
161
+ "pass@128": 0.9240121580547113
162
+ },
163
+ "per_op_pass_at_k": {
164
+ "10": {
165
+ "pass@1": 0.6446580014749262,
166
+ "pass@2": 0.7086265765451887,
167
+ "pass@4": 0.7599441355781849,
168
+ "pass@8": 0.8023810909443855,
169
+ "pass@16": 0.8380110251063062,
170
+ "pass@32": 0.8698539957395686,
171
+ "pass@64": 0.8976561781654913,
172
+ "pass@128": 0.9174041297935103
173
+ },
174
+ "2": {
175
+ "pass@1": 0.9338979750778816,
176
+ "pass@2": 0.9595586657100107,
177
+ "pass@4": 0.9756525451841886,
178
+ "pass@8": 0.98409817917281,
179
+ "pass@16": 0.9875394159838597,
180
+ "pass@32": 0.9894012652917533,
181
+ "pass@64": 0.9905213626273273,
182
+ "pass@128": 0.9906542056074766
183
+ },
184
+ "3": {
185
+ "pass@1": 0.8746655198776758,
186
+ "pass@2": 0.9018562329938111,
187
+ "pass@4": 0.9235414171279048,
188
+ "pass@8": 0.9425918675855139,
189
+ "pass@16": 0.9583857140671359,
190
+ "pass@32": 0.9698585704998334,
191
+ "pass@64": 0.9771319454203935,
192
+ "pass@128": 0.981651376146789
193
+ },
194
+ "4": {
195
+ "pass@1": 0.78740234375,
196
+ "pass@2": 0.8135100116879921,
197
+ "pass@4": 0.8321275546611361,
198
+ "pass@8": 0.8498828584765968,
199
+ "pass@16": 0.8688294182277513,
200
+ "pass@32": 0.8893387219985044,
201
+ "pass@64": 0.9084337390426462,
202
+ "pass@128": 0.921875
203
+ },
204
+ "5": {
205
+ "pass@1": 0.6442956349206349,
206
+ "pass@2": 0.7044240563679539,
207
+ "pass@4": 0.7589214889805439,
208
+ "pass@8": 0.8062138260832564,
209
+ "pass@16": 0.8448474665392458,
210
+ "pass@32": 0.8742810691280744,
211
+ "pass@64": 0.8951867234279609,
212
+ "pass@128": 0.9079365079365079
213
+ },
214
+ "6": {
215
+ "pass@1": 0.6633607784431138,
216
+ "pass@2": 0.7106940155122822,
217
+ "pass@4": 0.7466236113262785,
218
+ "pass@8": 0.7790429849284155,
219
+ "pass@16": 0.8107574745740735,
220
+ "pass@32": 0.8402023234939167,
221
+ "pass@64": 0.8645513213029393,
222
+ "pass@128": 0.8832335329341318
223
+ },
224
+ "7": {
225
+ "pass@1": 0.6632766272189349,
226
+ "pass@2": 0.7206016604155985,
227
+ "pass@4": 0.7671970370382695,
228
+ "pass@8": 0.8074138976675181,
229
+ "pass@16": 0.843155207773576,
230
+ "pass@32": 0.8729361699244267,
231
+ "pass@64": 0.8974884273605426,
232
+ "pass@128": 0.9230769230769231
233
+ },
234
+ "8": {
235
+ "pass@1": 0.6051933092948718,
236
+ "pass@2": 0.6755312436906926,
237
+ "pass@4": 0.7341284034207262,
238
+ "pass@8": 0.7835400933327527,
239
+ "pass@16": 0.8211524415475958,
240
+ "pass@32": 0.8458170234934912,
241
+ "pass@64": 0.862760090779185,
242
+ "pass@128": 0.875
243
+ },
244
+ "9": {
245
+ "pass@1": 0.6189040492957747,
246
+ "pass@2": 0.6980834812021737,
247
+ "pass@4": 0.7593200700264576,
248
+ "pass@8": 0.8095852069004352,
249
+ "pass@16": 0.8496181125257684,
250
+ "pass@32": 0.8787592991762989,
251
+ "pass@64": 0.8996024699023663,
252
+ "pass@128": 0.9154929577464789
253
+ }
254
+ }
255
+ },
256
+ "teachers_in_school": {
257
+ "count": 390656,
258
+ "correct": 276629,
259
+ "answer_accuracy": 0.7081140440694627,
260
+ "avg_response_len": 203.20306100507864,
261
+ "resp_tokens_sum": 79382495,
262
+ "pass_at_k": {
263
+ "pass@1": 0.7081140440694627,
264
+ "pass@2": 0.7598788808939049,
265
+ "pass@4": 0.8019528938101302,
266
+ "pass@8": 0.8362082935748024,
267
+ "pass@16": 0.8646761278706295,
268
+ "pass@32": 0.8887579086075584,
269
+ "pass@64": 0.9083049990221742,
270
+ "pass@128": 0.9246395806028833
271
+ },
272
+ "per_op_pass_at_k": {
273
+ "10": {
274
+ "pass@1": 0.6655483128834356,
275
+ "pass@2": 0.7261790644171779,
276
+ "pass@4": 0.774273106486689,
277
+ "pass@8": 0.8154503077702322,
278
+ "pass@16": 0.8513827000228652,
279
+ "pass@32": 0.880742119528748,
280
+ "pass@64": 0.9012886048245673,
281
+ "pass@128": 0.9171779141104295
282
+ },
283
+ "2": {
284
+ "pass@1": 0.9393522036474165,
285
+ "pass@2": 0.9591322278199266,
286
+ "pass@4": 0.9724622240419033,
287
+ "pass@8": 0.9801611460870774,
288
+ "pass@16": 0.9848945784068139,
289
+ "pass@32": 0.9885009927054951,
290
+ "pass@64": 0.9904609452424755,
291
+ "pass@128": 0.9908814589665653
292
+ },
293
+ "3": {
294
+ "pass@1": 0.8837121212121212,
295
+ "pass@2": 0.9071820567883556,
296
+ "pass@4": 0.9250117525082093,
297
+ "pass@8": 0.938517803565236,
298
+ "pass@16": 0.9501655320873388,
299
+ "pass@32": 0.9610128503681503,
300
+ "pass@64": 0.9710034013222898,
301
+ "pass@128": 0.9787878787878788
302
+ },
303
+ "4": {
304
+ "pass@1": 0.7681213017751479,
305
+ "pass@2": 0.806347333201323,
306
+ "pass@4": 0.8370125978150657,
307
+ "pass@8": 0.8643633805767534,
308
+ "pass@16": 0.8894955108387014,
309
+ "pass@32": 0.9112200498217551,
310
+ "pass@64": 0.9273172393177025,
311
+ "pass@128": 0.9408284023668639
312
+ },
313
+ "5": {
314
+ "pass@1": 0.6835597826086957,
315
+ "pass@2": 0.7398622047244096,
316
+ "pass@4": 0.7888190215353517,
317
+ "pass@8": 0.8288249541909402,
318
+ "pass@16": 0.8586746625294572,
319
+ "pass@32": 0.8804319932824916,
320
+ "pass@64": 0.8952530838618767,
321
+ "pass@128": 0.9043478260869565
322
+ },
323
+ "6": {
324
+ "pass@1": 0.6133372134670487,
325
+ "pass@2": 0.6639956963653184,
326
+ "pass@4": 0.7073447105823805,
327
+ "pass@8": 0.7443909543160218,
328
+ "pass@16": 0.7771691127933986,
329
+ "pass@32": 0.8081173959255104,
330
+ "pass@64": 0.8379673172871838,
331
+ "pass@128": 0.8681948424068768
332
+ },
333
+ "7": {
334
+ "pass@1": 0.6320899566473989,
335
+ "pass@2": 0.6949434055118112,
336
+ "pass@4": 0.7472537736431784,
337
+ "pass@8": 0.7896054515944385,
338
+ "pass@16": 0.8240666243617005,
339
+ "pass@32": 0.8523702333364069,
340
+ "pass@64": 0.8770839741824703,
341
+ "pass@128": 0.9046242774566474
342
+ },
343
+ "8": {
344
+ "pass@1": 0.6070422535211267,
345
+ "pass@2": 0.680067511367417,
346
+ "pass@4": 0.7394342472155769,
347
+ "pass@8": 0.7874884126649276,
348
+ "pass@16": 0.8286021843757088,
349
+ "pass@32": 0.864914001710332,
350
+ "pass@64": 0.8928018134050257,
351
+ "pass@128": 0.9098591549295775
352
+ },
353
+ "9": {
354
+ "pass@1": 0.5982410179640718,
355
+ "pass@2": 0.6769001367344052,
356
+ "pass@4": 0.7394855126642103,
357
+ "pass@8": 0.788722370608343,
358
+ "pass@16": 0.8276968595373282,
359
+ "pass@32": 0.8601113228758155,
360
+ "pass@64": 0.8888388174876141,
361
+ "pass@128": 0.9131736526946108
362
+ }
363
+ }
364
+ },
365
+ "movie_festival_awards": {
366
+ "count": 382336,
367
+ "correct": 270007,
368
+ "answer_accuracy": 0.7062034440910613,
369
+ "avg_response_len": 219.17418187144293,
370
+ "resp_tokens_sum": 83798180,
371
+ "pass_at_k": {
372
+ "pass@1": 0.7062034440910613,
373
+ "pass@2": 0.7627556929766539,
374
+ "pass@4": 0.8081092622461369,
375
+ "pass@8": 0.8434282263918436,
376
+ "pass@16": 0.8706911049406025,
377
+ "pass@32": 0.8922600645673889,
378
+ "pass@64": 0.9102659708034204,
379
+ "pass@128": 0.9263475058587212
380
+ },
381
+ "per_op_pass_at_k": {
382
+ "10": {
383
+ "pass@1": 0.6044776119402985,
384
+ "pass@2": 0.6872576095898459,
385
+ "pass@4": 0.7601403645439836,
386
+ "pass@8": 0.8148680866262519,
387
+ "pass@16": 0.8520847652047296,
388
+ "pass@32": 0.880442491489034,
389
+ "pass@64": 0.9045682600821996,
390
+ "pass@128": 0.9223880597014925
391
+ },
392
+ "2": {
393
+ "pass@1": 0.9144642857142857,
394
+ "pass@2": 0.9483858267716534,
395
+ "pass@4": 0.9666809258664094,
396
+ "pass@8": 0.9760876441015367,
397
+ "pass@16": 0.9817839791769549,
398
+ "pass@32": 0.9860892137379141,
399
+ "pass@64": 0.9910869792252629,
400
+ "pass@128": 0.9971428571428571
401
+ },
402
+ "3": {
403
+ "pass@1": 0.8872084548104956,
404
+ "pass@2": 0.911002315718188,
405
+ "pass@4": 0.9287985389399213,
406
+ "pass@8": 0.9418294466732434,
407
+ "pass@16": 0.9516032685115303,
408
+ "pass@32": 0.959531567551875,
409
+ "pass@64": 0.9669895513156562,
410
+ "pass@128": 0.9766763848396501
411
+ },
412
+ "4": {
413
+ "pass@1": 0.7720897295321637,
414
+ "pass@2": 0.8149048700326933,
415
+ "pass@4": 0.8442139976472239,
416
+ "pass@8": 0.8641491648156164,
417
+ "pass@16": 0.8781066650852173,
418
+ "pass@32": 0.8898259443521337,
419
+ "pass@64": 0.9017391378987524,
420
+ "pass@128": 0.9181286549707602
421
+ },
422
+ "5": {
423
+ "pass@1": 0.6713694852941177,
424
+ "pass@2": 0.7327079232283464,
425
+ "pass@4": 0.7863907416903769,
426
+ "pass@8": 0.8296703924040345,
427
+ "pass@16": 0.8624125124677811,
428
+ "pass@32": 0.8867811890994626,
429
+ "pass@64": 0.9061302658187409,
430
+ "pass@128": 0.9235294117647059
431
+ },
432
+ "6": {
433
+ "pass@1": 0.6576054810725552,
434
+ "pass@2": 0.7131347959462478,
435
+ "pass@4": 0.7573902998915356,
436
+ "pass@8": 0.7962511637170556,
437
+ "pass@16": 0.8323717329025123,
438
+ "pass@32": 0.8641677109359124,
439
+ "pass@64": 0.8910536305522675,
440
+ "pass@128": 0.9148264984227129
441
+ },
442
+ "7": {
443
+ "pass@1": 0.6052462420886076,
444
+ "pass@2": 0.6687863643725713,
445
+ "pass@4": 0.7221455111821459,
446
+ "pass@8": 0.7657531013969413,
447
+ "pass@16": 0.8000025591088098,
448
+ "pass@32": 0.8273544101920233,
449
+ "pass@64": 0.8501011080755299,
450
+ "pass@128": 0.8670886075949367
451
+ },
452
+ "8": {
453
+ "pass@1": 0.6090700075075075,
454
+ "pass@2": 0.6801576428790994,
455
+ "pass@4": 0.7411887252832131,
456
+ "pass@8": 0.7943322020848278,
457
+ "pass@16": 0.837816999051833,
458
+ "pass@32": 0.8695440940640597,
459
+ "pass@64": 0.889940346578068,
460
+ "pass@128": 0.9039039039039038
461
+ },
462
+ "9": {
463
+ "pass@1": 0.6035219051446945,
464
+ "pass@2": 0.6816728169228042,
465
+ "pass@4": 0.7429534720539352,
466
+ "pass@8": 0.788205651285775,
467
+ "pass@16": 0.8234507328349505,
468
+ "pass@32": 0.8527736564709415,
469
+ "pass@64": 0.8792638805305321,
470
+ "pass@128": 0.9035369774919614
471
+ }
472
+ }
473
+ }
474
+ }
475
+ },
476
+ "ood": {
477
+ "answer_accuracy": 0.34299453125,
478
+ "avg_response_len": 346.3708234375,
479
+ "avg_loss": 0.0,
480
+ "count": 1280000,
481
+ "validation_metrics": {},
482
+ "pass_at_k": {
483
+ "pass@1": 0.34299453125,
484
+ "pass@2": 0.44908989911417846,
485
+ "pass@4": 0.5506458316741525,
486
+ "pass@8": 0.6372070029423487,
487
+ "pass@16": 0.7053478390494748,
488
+ "pass@32": 0.7584346518548263,
489
+ "pass@64": 0.8010283467006837,
490
+ "pass@128": 0.8353
491
+ },
492
+ "per_op_pass_at_k": {
493
+ "11": {
494
+ "pass@1": 0.56490625,
495
+ "pass@2": 0.6544270423228341,
496
+ "pass@4": 0.7290522653730787,
497
+ "pass@8": 0.7873574960028927,
498
+ "pass@16": 0.830889302439461,
499
+ "pass@32": 0.8643758843809567,
500
+ "pass@64": 0.8904049212950518,
501
+ "pass@128": 0.909
502
+ },
503
+ "12": {
504
+ "pass@1": 0.5658828125,
505
+ "pass@2": 0.6598180364173227,
506
+ "pass@4": 0.7372140781777282,
507
+ "pass@8": 0.800772293658024,
508
+ "pass@16": 0.8489480396300843,
509
+ "pass@32": 0.8836785553231575,
510
+ "pass@64": 0.9081395251588857,
511
+ "pass@128": 0.923
512
+ },
513
+ "13": {
514
+ "pass@1": 0.465703125,
515
+ "pass@2": 0.5817186269685033,
516
+ "pass@4": 0.6812352809336336,
517
+ "pass@8": 0.7602656348471736,
518
+ "pass@16": 0.8173740323094875,
519
+ "pass@32": 0.8578417542771846,
520
+ "pass@64": 0.8884755418122985,
521
+ "pass@128": 0.912
522
+ },
523
+ "14": {
524
+ "pass@1": 0.420703125,
525
+ "pass@2": 0.545852854330709,
526
+ "pass@4": 0.6570371287026623,
527
+ "pass@8": 0.7456495211988309,
528
+ "pass@16": 0.8108015406911748,
529
+ "pass@32": 0.8564432616819933,
530
+ "pass@64": 0.8873730721805125,
531
+ "pass@128": 0.91
532
+ },
533
+ "15": {
534
+ "pass@1": 0.333390625,
535
+ "pass@2": 0.4588850885826779,
536
+ "pass@4": 0.5796260408698913,
537
+ "pass@8": 0.6790236952819334,
538
+ "pass@16": 0.7498576660395769,
539
+ "pass@32": 0.7973665148017777,
540
+ "pass@64": 0.8344988374302111,
541
+ "pass@128": 0.869
542
+ },
543
+ "16": {
544
+ "pass@1": 0.31715625,
545
+ "pass@2": 0.44179687500000053,
546
+ "pass@4": 0.5593755952380945,
547
+ "pass@8": 0.6559526315293914,
548
+ "pass@16": 0.7302236740331446,
549
+ "pass@32": 0.78675594542704,
550
+ "pass@64": 0.8297895182154855,
551
+ "pass@128": 0.863
552
+ },
553
+ "17": {
554
+ "pass@1": 0.2290625,
555
+ "pass@2": 0.3393805364173231,
556
+ "pass@4": 0.4559913035245587,
557
+ "pass@8": 0.5601181595897614,
558
+ "pass@16": 0.6445439935836196,
559
+ "pass@32": 0.7123795630488152,
560
+ "pass@64": 0.7662916779603728,
561
+ "pass@128": 0.807
562
+ },
563
+ "18": {
564
+ "pass@1": 0.2013359375,
565
+ "pass@2": 0.3025404773622051,
566
+ "pass@4": 0.40948668794525667,
567
+ "pass@8": 0.5058880822099092,
568
+ "pass@16": 0.5862113526890013,
569
+ "pass@32": 0.6532699393602096,
570
+ "pass@64": 0.710317758712882,
571
+ "pass@128": 0.756
572
+ },
573
+ "19": {
574
+ "pass@1": 0.1714140625,
575
+ "pass@2": 0.25968442421259846,
576
+ "pass@4": 0.3559171721972258,
577
+ "pass@8": 0.44719671330719474,
578
+ "pass@16": 0.5275662750104985,
579
+ "pass@32": 0.5974458157029271,
580
+ "pass@64": 0.6591500641761043,
581
+ "pass@128": 0.71
582
+ },
583
+ "20": {
584
+ "pass@1": 0.160390625,
585
+ "pass@2": 0.24679502952755894,
586
+ "pass@4": 0.341522763779527,
587
+ "pass@8": 0.4298458017983287,
588
+ "pass@16": 0.5070625140685019,
589
+ "pass@32": 0.5747892845441949,
590
+ "pass@64": 0.6358425500649428,
591
+ "pass@128": 0.694
592
+ }
593
+ },
594
+ "per_op_accuracy": {
595
+ "11": 0.56490625,
596
+ "12": 0.5658828125,
597
+ "13": 0.465703125,
598
+ "14": 0.420703125,
599
+ "15": 0.333390625,
600
+ "16": 0.31715625,
601
+ "17": 0.2290625,
602
+ "18": 0.2013359375,
603
+ "19": 0.1714140625,
604
+ "20": 0.160390625
605
+ },
606
+ "per_op_avg_response_len": {
607
+ "11": 297.0618984375,
608
+ "12": 302.5270703125,
609
+ "13": 318.6869609375,
610
+ "14": 342.1836484375,
611
+ "15": 353.993734375,
612
+ "16": 356.512640625,
613
+ "17": 367.45259375,
614
+ "18": 372.406625,
615
+ "19": 371.91615625,
616
+ "20": 380.96690625
617
+ },
618
+ "per_op_avg_loss": {},
619
+ "per_op_length": {
620
+ "11": 297.0618984375,
621
+ "12": 302.5270703125,
622
+ "13": 318.6869609375,
623
+ "14": 342.1836484375,
624
+ "15": 353.993734375,
625
+ "16": 356.512640625,
626
+ "17": 367.45259375,
627
+ "18": 372.406625,
628
+ "19": 371.91615625,
629
+ "20": 380.96690625
630
+ },
631
+ "per_op_loss": {},
632
+ "per_template": {
633
+ "crazy_zootopia": {
634
+ "count": 430720,
635
+ "correct": 148953,
636
+ "answer_accuracy": 0.34582327265973256,
637
+ "avg_response_len": 330.83627182392274,
638
+ "resp_tokens_sum": 142497799,
639
+ "pass_at_k": {
640
+ "pass@1": 0.34582327265973256,
641
+ "pass@2": 0.45180090323033534,
642
+ "pass@4": 0.5530280495305873,
643
+ "pass@8": 0.6394479136879506,
644
+ "pass@16": 0.7075869829033545,
645
+ "pass@32": 0.7596869174457247,
646
+ "pass@64": 0.8009689692880166,
647
+ "pass@128": 0.8344725111441308
648
+ },
649
+ "per_op_pass_at_k": {
650
+ "11": {
651
+ "pass@1": 0.5309401558073654,
652
+ "pass@2": 0.6182775311726265,
653
+ "pass@4": 0.6944196477033849,
654
+ "pass@8": 0.758216362808835,
655
+ "pass@16": 0.8089765941091365,
656
+ "pass@32": 0.8478680917510193,
657
+ "pass@64": 0.8762482825013892,
658
+ "pass@128": 0.8951841359773371
659
+ },
660
+ "12": {
661
+ "pass@1": 0.587548552259887,
662
+ "pass@2": 0.6789871936029179,
663
+ "pass@4": 0.7542816610317772,
664
+ "pass@8": 0.8155258561346665,
665
+ "pass@16": 0.859817387659267,
666
+ "pass@32": 0.8886653853533955,
667
+ "pass@64": 0.9077477566906305,
668
+ "pass@128": 0.9209039548022598
669
+ },
670
+ "13": {
671
+ "pass@1": 0.4798442398119122,
672
+ "pass@2": 0.6009471496803491,
673
+ "pass@4": 0.7033698418967224,
674
+ "pass@8": 0.7829716616327332,
675
+ "pass@16": 0.8393161383385362,
676
+ "pass@32": 0.8770397099307982,
677
+ "pass@64": 0.9037023197930119,
678
+ "pass@128": 0.9247648902821317
679
+ },
680
+ "14": {
681
+ "pass@1": 0.43294879283489096,
682
+ "pass@2": 0.555698389628866,
683
+ "pass@4": 0.6650006631764483,
684
+ "pass@8": 0.7540915897523429,
685
+ "pass@16": 0.8223832370157309,
686
+ "pass@32": 0.8708863421776203,
687
+ "pass@64": 0.9022030922121614,
688
+ "pass@128": 0.9221183800623053
689
+ },
690
+ "15": {
691
+ "pass@1": 0.3218253251445087,
692
+ "pass@2": 0.4450877719493877,
693
+ "pass@4": 0.5600203219720364,
694
+ "pass@8": 0.6543575898312131,
695
+ "pass@16": 0.7259508733588973,
696
+ "pass@32": 0.7762516558552659,
697
+ "pass@64": 0.815225131835705,
698
+ "pass@128": 0.8497109826589595
699
+ },
700
+ "16": {
701
+ "pass@1": 0.32691532258064515,
702
+ "pass@2": 0.4487609537719074,
703
+ "pass@4": 0.5625330040398175,
704
+ "pass@8": 0.654850007133806,
705
+ "pass@16": 0.7224705015275517,
706
+ "pass@32": 0.7722011271883154,
707
+ "pass@64": 0.8137722171021754,
708
+ "pass@128": 0.8516129032258064
709
+ },
710
+ "17": {
711
+ "pass@1": 0.22821080202312138,
712
+ "pass@2": 0.34035418847571786,
713
+ "pass@4": 0.4588241708225774,
714
+ "pass@8": 0.5640470576174099,
715
+ "pass@16": 0.6489542613213412,
716
+ "pass@32": 0.7180814737698088,
717
+ "pass@64": 0.775694384590296,
718
+ "pass@128": 0.8179190751445087
719
+ },
720
+ "18": {
721
+ "pass@1": 0.20578125,
722
+ "pass@2": 0.309048790776153,
723
+ "pass@4": 0.4157900827575123,
724
+ "pass@8": 0.509222118234702,
725
+ "pass@16": 0.5847525333486995,
726
+ "pass@32": 0.6449860122655471,
727
+ "pass@64": 0.6961719024625291,
728
+ "pass@128": 0.74
729
+ },
730
+ "19": {
731
+ "pass@1": 0.17349380630630631,
732
+ "pass@2": 0.26045693823745014,
733
+ "pass@4": 0.35654742819309765,
734
+ "pass@8": 0.4521122354686449,
735
+ "pass@16": 0.5394334685986726,
736
+ "pass@32": 0.6138803577519422,
737
+ "pass@64": 0.6735901886727701,
738
+ "pass@128": 0.7207207207207207
739
+ },
740
+ "20": {
741
+ "pass@1": 0.16450825825825827,
742
+ "pass@2": 0.25776945744957575,
743
+ "pass@4": 0.35973303984524463,
744
+ "pass@8": 0.45108622829881717,
745
+ "pass@16": 0.5266710916442264,
746
+ "pass@32": 0.590217523698882,
747
+ "pass@64": 0.6487109933142678,
748
+ "pass@128": 0.7057057057057057
749
+ }
750
+ }
751
+ },
752
+ "teachers_in_school": {
753
+ "count": 430464,
754
+ "correct": 149665,
755
+ "answer_accuracy": 0.34768296535831106,
756
+ "avg_response_len": 341.9255593963723,
757
+ "resp_tokens_sum": 147186644,
758
+ "pass_at_k": {
759
+ "pass@1": 0.34768296535831106,
760
+ "pass@2": 0.45276453930100896,
761
+ "pass@4": 0.5535522546746799,
762
+ "pass@8": 0.6399841369968611,
763
+ "pass@16": 0.7081335411553796,
764
+ "pass@32": 0.761430484571531,
765
+ "pass@64": 0.8044172767092296,
766
+ "pass@128": 0.83942908117752
767
+ },
768
+ "per_op_pass_at_k": {
769
+ "11": {
770
+ "pass@1": 0.6104103915662651,
771
+ "pass@2": 0.6973562310264679,
772
+ "pass@4": 0.7667736087770652,
773
+ "pass@8": 0.8195464930321138,
774
+ "pass@16": 0.8576035922567216,
775
+ "pass@32": 0.8854497881318836,
776
+ "pass@64": 0.9075046050572457,
777
+ "pass@128": 0.9246987951807228
778
+ },
779
+ "12": {
780
+ "pass@1": 0.5128085815047022,
781
+ "pass@2": 0.6041405690765925,
782
+ "pass@4": 0.685626737978285,
783
+ "pass@8": 0.7566146072328974,
784
+ "pass@16": 0.8134178038024091,
785
+ "pass@32": 0.858042121796473,
786
+ "pass@64": 0.8908438919484275,
787
+ "pass@128": 0.9090909090909091
788
+ },
789
+ "13": {
790
+ "pass@1": 0.4427521008403361,
791
+ "pass@2": 0.5480322542402787,
792
+ "pass@4": 0.6390175104162397,
793
+ "pass@8": 0.715290051078792,
794
+ "pass@16": 0.7766798970641074,
795
+ "pass@32": 0.8267608641818867,
796
+ "pass@64": 0.8672715367959482,
797
+ "pass@128": 0.896358543417367
798
+ },
799
+ "14": {
800
+ "pass@1": 0.43944170321637427,
801
+ "pass@2": 0.5645552015701978,
802
+ "pass@4": 0.6751130694737714,
803
+ "pass@8": 0.7626177000574363,
804
+ "pass@16": 0.8247123104489158,
805
+ "pass@32": 0.8640348131883745,
806
+ "pass@64": 0.8882515823050453,
807
+ "pass@128": 0.9064327485380117
808
+ },
809
+ "15": {
810
+ "pass@1": 0.34924242424242424,
811
+ "pass@2": 0.48043500954426144,
812
+ "pass@4": 0.6070050053402414,
813
+ "pass@8": 0.7092448683715644,
814
+ "pass@16": 0.7805712407453619,
815
+ "pass@32": 0.8284252331696917,
816
+ "pass@64": 0.8643444343864802,
817
+ "pass@128": 0.896969696969697
818
+ },
819
+ "16": {
820
+ "pass@1": 0.32687856125356124,
821
+ "pass@2": 0.4480165640352647,
822
+ "pass@4": 0.5591375172654703,
823
+ "pass@8": 0.6514324270194297,
824
+ "pass@16": 0.7261167053969767,
825
+ "pass@32": 0.7837995880088544,
826
+ "pass@64": 0.8244406218061658,
827
+ "pass@128": 0.8547008547008547
828
+ },
829
+ "17": {
830
+ "pass@1": 0.25883152173913043,
831
+ "pass@2": 0.3733108004841783,
832
+ "pass@4": 0.49168759417495167,
833
+ "pass@8": 0.5938284001034063,
834
+ "pass@16": 0.6708776227542887,
835
+ "pass@32": 0.7290766474988695,
836
+ "pass@64": 0.775271431861227,
837
+ "pass@128": 0.8136645962732919
838
+ },
839
+ "18": {
840
+ "pass@1": 0.20222886029411766,
841
+ "pass@2": 0.30483839451134787,
842
+ "pass@4": 0.41794727552438293,
843
+ "pass@8": 0.5237158396938546,
844
+ "pass@16": 0.6092212993975471,
845
+ "pass@32": 0.6766702683650773,
846
+ "pass@64": 0.7347166739282228,
847
+ "pass@128": 0.7823529411764706
848
+ },
849
+ "19": {
850
+ "pass@1": 0.17661830357142858,
851
+ "pass@2": 0.267703719652231,
852
+ "pass@4": 0.36584130750397276,
853
+ "pass@8": 0.455991164869063,
854
+ "pass@16": 0.5330632685639742,
855
+ "pass@32": 0.6014918780285793,
856
+ "pass@64": 0.6652790438088099,
857
+ "pass@128": 0.7202380952380952
858
+ },
859
+ "20": {
860
+ "pass@1": 0.15938435628742514,
861
+ "pass@2": 0.23976554764486785,
862
+ "pass@4": 0.3274774410683695,
863
+ "pass@8": 0.41152875898223257,
864
+ "pass@16": 0.48862133484451187,
865
+ "pass@32": 0.5596604592346828,
866
+ "pass@64": 0.6251480265327629,
867
+ "pass@128": 0.688622754491018
868
+ }
869
+ }
870
+ },
871
+ "movie_festival_awards": {
872
+ "count": 418816,
873
+ "correct": 140415,
874
+ "answer_accuracy": 0.3352665609718826,
875
+ "avg_response_len": 366.9158078965465,
876
+ "resp_tokens_sum": 153670211,
877
+ "pass_at_k": {
878
+ "pass@1": 0.3352665609718826,
879
+ "pass@2": 0.4425250018650254,
880
+ "pass@4": 0.5452086484108526,
881
+ "pass@8": 0.6320480278554537,
882
+ "pass@16": 0.7001818747308222,
883
+ "pass@32": 0.7540676411153115,
884
+ "pass@64": 0.7976062297614274,
885
+ "pass@128": 0.8319070904645477
886
+ },
887
+ "per_op_pass_at_k": {
888
+ "11": {
889
+ "pass@1": 0.5550099206349206,
890
+ "pass@2": 0.6496914448193972,
891
+ "pass@4": 0.7281056876818967,
892
+ "pass@8": 0.7860878864911501,
893
+ "pass@16": 0.827289435205416,
894
+ "pass@32": 0.8606638677240043,
895
+ "pass@64": 0.8882468212477946,
896
+ "pass@128": 0.9079365079365079
897
+ },
898
+ "12": {
899
+ "pass@1": 0.5942039373088684,
900
+ "pass@2": 0.6933814322521614,
901
+ "pass@4": 0.7690625099614834,
902
+ "pass@8": 0.827877923177547,
903
+ "pass@16": 0.871842217081767,
904
+ "pass@32": 0.9032892111773124,
905
+ "pass@64": 0.9254361399353297,
906
+ "pass@128": 0.9388379204892966
907
+ },
908
+ "13": {
909
+ "pass@1": 0.47706886574074076,
910
+ "pass@2": 0.599904232891028,
911
+ "pass@4": 0.7059600313271024,
912
+ "pass@8": 0.7874664399728473,
913
+ "pass@16": 0.8406094473074397,
914
+ "pass@32": 0.8731865996800195,
915
+ "pass@64": 0.8968474171671995,
916
+ "pass@128": 0.9166666666666666
917
+ },
918
+ "14": {
919
+ "pass@1": 0.39002225519287836,
920
+ "pass@2": 0.5174949327087078,
921
+ "pass@4": 0.6311075550830626,
922
+ "pass@8": 0.7203883307676154,
923
+ "pass@16": 0.7856525561887167,
924
+ "pass@32": 0.8349817202746366,
925
+ "pass@64": 0.8723556036560305,
926
+ "pass@128": 0.9020771513353115
927
+ },
928
+ "15": {
929
+ "pass@1": 0.32959587191358025,
930
+ "pass@2": 0.45167026647710673,
931
+ "pass@4": 0.5726770299545887,
932
+ "pass@8": 0.6745838353015993,
933
+ "pass@16": 0.7441054148500932,
934
+ "pass@32": 0.7882812189193158,
935
+ "pass@64": 0.8246829582330772,
936
+ "pass@128": 0.8611111111111112
937
+ },
938
+ "16": {
939
+ "pass@1": 0.2981655604719764,
940
+ "pass@2": 0.4289886883608576,
941
+ "pass@4": 0.5567347947656982,
942
+ "pass@8": 0.661641142873429,
943
+ "pass@16": 0.741565943850342,
944
+ "pass@32": 0.8031266684588632,
945
+ "pass@64": 0.8499748456042655,
946
+ "pass@128": 0.8820058997050148
947
+ },
948
+ "17": {
949
+ "pass@1": 0.20107774849397592,
950
+ "pass@2": 0.3054575585807797,
951
+ "pass@4": 0.4184178767940154,
952
+ "pass@8": 0.5233287133157847,
953
+ "pass@16": 0.614407303131194,
954
+ "pass@32": 0.6902430500898341,
955
+ "pass@64": 0.7477831922675113,
956
+ "pass@128": 0.7891566265060241
957
+ },
958
+ "18": {
959
+ "pass@1": 0.19533770161290323,
960
+ "pass@2": 0.29267208534417066,
961
+ "pass@4": 0.3930905977478621,
962
+ "pass@8": 0.4825708236511399,
963
+ "pass@16": 0.5626216910380368,
964
+ "pass@32": 0.6369578833004578,
965
+ "pass@64": 0.6995287861787093,
966
+ "pass@128": 0.7451612903225806
967
+ },
968
+ "19": {
969
+ "pass@1": 0.16403889728096677,
970
+ "pass@2": 0.2507668095963082,
971
+ "pass@4": 0.34520906129180984,
972
+ "pass@8": 0.4333241918432945,
973
+ "pass@16": 0.510047346615244,
974
+ "pass@32": 0.5768047901931338,
975
+ "pass@64": 0.6384011257652897,
976
+ "pass@128": 0.6888217522658611
977
+ },
978
+ "20": {
979
+ "pass@1": 0.15728228228228228,
980
+ "pass@2": 0.24287119304343707,
981
+ "pass@4": 0.33739998857124837,
982
+ "pass@8": 0.42697742424851853,
983
+ "pass@16": 0.5059504946333598,
984
+ "pass@32": 0.5745353026068557,
985
+ "pass@64": 0.6337007460340792,
986
+ "pass@128": 0.6876876876876877
987
+ }
988
+ }
989
+ }
990
+ }
991
+ },
992
+ "total": {
993
+ "answer_accuracy": 0.516530427631579,
994
+ "avg_loss": 0.0,
995
+ "count": 2432000,
996
+ "validation_metrics": {
997
+ "id": {},
998
+ "ood": {}
999
+ },
1000
+ "pass_at_k": {
1001
+ "pass@1": 0.516530427631579,
1002
+ "pass@2": 0.5975414551388062,
1003
+ "pass@4": 0.6712251129102028,
1004
+ "pass@8": 0.7331853411351233,
1005
+ "pass@16": 0.7823920050522624,
1006
+ "pass@32": 0.8212189433464985,
1007
+ "pass@64": 0.8524480840021712,
1008
+ "pass@128": 0.8777894736842106
1009
+ },
1010
+ "per_op_pass_at_k": {
1011
+ "10": {
1012
+ "pass@1": 0.6380078125,
1013
+ "pass@2": 0.7071900836614177,
1014
+ "pass@4": 0.7646811167979007,
1015
+ "pass@8": 0.8108247991830362,
1016
+ "pass@16": 0.8470848940620732,
1017
+ "pass@32": 0.8769506701709128,
1018
+ "pass@64": 0.901155896698448,
1019
+ "pass@128": 0.919
1020
+ },
1021
+ "2": {
1022
+ "pass@1": 0.928890625,
1023
+ "pass@2": 0.9555078740157483,
1024
+ "pass@4": 0.9714628627671541,
1025
+ "pass@8": 0.9799992080126586,
1026
+ "pass@16": 0.9846548615385947,
1027
+ "pass@32": 0.9879458575670311,
1028
+ "pass@64": 0.9906994511169885,
1029
+ "pass@128": 0.993
1030
+ },
1031
+ "3": {
1032
+ "pass@1": 0.881953125,
1033
+ "pass@2": 0.9067508612204728,
1034
+ "pass@4": 0.9258298205849268,
1035
+ "pass@8": 0.9409859160859129,
1036
+ "pass@16": 0.9533466751882298,
1037
+ "pass@32": 0.9633973208452283,
1038
+ "pass@64": 0.971630684690094,
1039
+ "pass@128": 0.979
1040
+ },
1041
+ "4": {
1042
+ "pass@1": 0.7756484375,
1043
+ "pass@2": 0.811566067913386,
1044
+ "pass@4": 0.8379122627484064,
1045
+ "pass@8": 0.8596563517143947,
1046
+ "pass@16": 0.8789873759555055,
1047
+ "pass@32": 0.8969012408477046,
1048
+ "pass@64": 0.9125268085444034,
1049
+ "pass@128": 0.927
1050
+ },
1051
+ "5": {
1052
+ "pass@1": 0.667046875,
1053
+ "pass@2": 0.726266732283465,
1054
+ "pass@4": 0.778575683633296,
1055
+ "pass@8": 0.8219898978294717,
1056
+ "pass@16": 0.8555899647715695,
1057
+ "pass@32": 0.8806531787516202,
1058
+ "pass@64": 0.8989304221905279,
1059
+ "pass@128": 0.912
1060
+ },
1061
+ "6": {
1062
+ "pass@1": 0.644078125,
1063
+ "pass@2": 0.6951700295275594,
1064
+ "pass@4": 0.7363283152418446,
1065
+ "pass@8": 0.772404418920689,
1066
+ "pass@16": 0.8058868562027328,
1067
+ "pass@32": 0.8366017115916559,
1068
+ "pass@64": 0.863674735933478,
1069
+ "pass@128": 0.888
1070
+ },
1071
+ "7": {
1072
+ "pass@1": 0.6341484375,
1073
+ "pass@2": 0.695350270669291,
1074
+ "pass@4": 0.7460603857330329,
1075
+ "pass@8": 0.7880873637047299,
1076
+ "pass@16": 0.8229143209349995,
1077
+ "pass@32": 0.8514165197895321,
1078
+ "pass@64": 0.8754540936668659,
1079
+ "pass@128": 0.899
1080
+ },
1081
+ "8": {
1082
+ "pass@1": 0.607140625,
1083
+ "pass@2": 0.6786822096456687,
1084
+ "pass@4": 0.7383630651481062,
1085
+ "pass@8": 0.7885355189101158,
1086
+ "pass@16": 0.8293463979004868,
1087
+ "pass@32": 0.8604975652604689,
1088
+ "pass@64": 0.8824759274923865,
1089
+ "pass@128": 0.897
1090
+ },
1091
+ "9": {
1092
+ "pass@1": 0.60721875,
1093
+ "pass@2": 0.6859045275590553,
1094
+ "pass@4": 0.7476053158980122,
1095
+ "pass@8": 0.7959679777827171,
1096
+ "pass@16": 0.8341583589437844,
1097
+ "pass@32": 0.864449340210571,
1098
+ "pass@64": 0.889682108701199,
1099
+ "pass@128": 0.911
1100
+ },
1101
+ "11": {
1102
+ "pass@1": 0.56490625,
1103
+ "pass@2": 0.6544270423228341,
1104
+ "pass@4": 0.7290522653730787,
1105
+ "pass@8": 0.7873574960028927,
1106
+ "pass@16": 0.830889302439461,
1107
+ "pass@32": 0.8643758843809567,
1108
+ "pass@64": 0.8904049212950518,
1109
+ "pass@128": 0.909
1110
+ },
1111
+ "12": {
1112
+ "pass@1": 0.5658828125,
1113
+ "pass@2": 0.6598180364173227,
1114
+ "pass@4": 0.7372140781777282,
1115
+ "pass@8": 0.800772293658024,
1116
+ "pass@16": 0.8489480396300843,
1117
+ "pass@32": 0.8836785553231575,
1118
+ "pass@64": 0.9081395251588857,
1119
+ "pass@128": 0.923
1120
+ },
1121
+ "13": {
1122
+ "pass@1": 0.465703125,
1123
+ "pass@2": 0.5817186269685033,
1124
+ "pass@4": 0.6812352809336336,
1125
+ "pass@8": 0.7602656348471736,
1126
+ "pass@16": 0.8173740323094875,
1127
+ "pass@32": 0.8578417542771846,
1128
+ "pass@64": 0.8884755418122985,
1129
+ "pass@128": 0.912
1130
+ },
1131
+ "14": {
1132
+ "pass@1": 0.420703125,
1133
+ "pass@2": 0.545852854330709,
1134
+ "pass@4": 0.6570371287026623,
1135
+ "pass@8": 0.7456495211988309,
1136
+ "pass@16": 0.8108015406911748,
1137
+ "pass@32": 0.8564432616819933,
1138
+ "pass@64": 0.8873730721805125,
1139
+ "pass@128": 0.91
1140
+ },
1141
+ "15": {
1142
+ "pass@1": 0.333390625,
1143
+ "pass@2": 0.4588850885826779,
1144
+ "pass@4": 0.5796260408698913,
1145
+ "pass@8": 0.6790236952819334,
1146
+ "pass@16": 0.7498576660395769,
1147
+ "pass@32": 0.7973665148017777,
1148
+ "pass@64": 0.8344988374302111,
1149
+ "pass@128": 0.869
1150
+ },
1151
+ "16": {
1152
+ "pass@1": 0.31715625,
1153
+ "pass@2": 0.44179687500000053,
1154
+ "pass@4": 0.5593755952380945,
1155
+ "pass@8": 0.6559526315293914,
1156
+ "pass@16": 0.7302236740331446,
1157
+ "pass@32": 0.78675594542704,
1158
+ "pass@64": 0.8297895182154855,
1159
+ "pass@128": 0.863
1160
+ },
1161
+ "17": {
1162
+ "pass@1": 0.2290625,
1163
+ "pass@2": 0.3393805364173231,
1164
+ "pass@4": 0.4559913035245587,
1165
+ "pass@8": 0.5601181595897614,
1166
+ "pass@16": 0.6445439935836196,
1167
+ "pass@32": 0.7123795630488152,
1168
+ "pass@64": 0.7662916779603728,
1169
+ "pass@128": 0.807
1170
+ },
1171
+ "18": {
1172
+ "pass@1": 0.2013359375,
1173
+ "pass@2": 0.3025404773622051,
1174
+ "pass@4": 0.40948668794525667,
1175
+ "pass@8": 0.5058880822099092,
1176
+ "pass@16": 0.5862113526890013,
1177
+ "pass@32": 0.6532699393602096,
1178
+ "pass@64": 0.710317758712882,
1179
+ "pass@128": 0.756
1180
+ },
1181
+ "19": {
1182
+ "pass@1": 0.1714140625,
1183
+ "pass@2": 0.25968442421259846,
1184
+ "pass@4": 0.3559171721972258,
1185
+ "pass@8": 0.44719671330719474,
1186
+ "pass@16": 0.5275662750104985,
1187
+ "pass@32": 0.5974458157029271,
1188
+ "pass@64": 0.6591500641761043,
1189
+ "pass@128": 0.71
1190
+ },
1191
+ "20": {
1192
+ "pass@1": 0.160390625,
1193
+ "pass@2": 0.24679502952755894,
1194
+ "pass@4": 0.341522763779527,
1195
+ "pass@8": 0.4298458017983287,
1196
+ "pass@16": 0.5070625140685019,
1197
+ "pass@32": 0.5747892845441949,
1198
+ "pass@64": 0.6358425500649428,
1199
+ "pass@128": 0.694
1200
+ }
1201
+ },
1202
+ "per_op_accuracy": {
1203
+ "10": 0.6380078125,
1204
+ "2": 0.928890625,
1205
+ "3": 0.881953125,
1206
+ "4": 0.7756484375,
1207
+ "5": 0.667046875,
1208
+ "6": 0.644078125,
1209
+ "7": 0.6341484375,
1210
+ "8": 0.607140625,
1211
+ "9": 0.60721875,
1212
+ "11": 0.56490625,
1213
+ "12": 0.5658828125,
1214
+ "13": 0.465703125,
1215
+ "14": 0.420703125,
1216
+ "15": 0.333390625,
1217
+ "16": 0.31715625,
1218
+ "17": 0.2290625,
1219
+ "18": 0.2013359375,
1220
+ "19": 0.1714140625,
1221
+ "20": 0.160390625
1222
+ },
1223
+ "per_op_avg_response_len": {
1224
+ "10": 279.7841328125,
1225
+ "2": 133.7599609375,
1226
+ "3": 152.97484375,
1227
+ "4": 163.8752578125,
1228
+ "5": 182.675140625,
1229
+ "6": 207.0015078125,
1230
+ "7": 227.182375,
1231
+ "8": 247.6363125,
1232
+ "9": 263.458296875,
1233
+ "11": 297.0618984375,
1234
+ "12": 302.5270703125,
1235
+ "13": 318.6869609375,
1236
+ "14": 342.1836484375,
1237
+ "15": 353.993734375,
1238
+ "16": 356.512640625,
1239
+ "17": 367.45259375,
1240
+ "18": 372.406625,
1241
+ "19": 371.91615625,
1242
+ "20": 380.96690625
1243
+ },
1244
+ "per_op_avg_loss": {},
1245
+ "per_op_length": {
1246
+ "10": 279.7841328125,
1247
+ "2": 133.7599609375,
1248
+ "3": 152.97484375,
1249
+ "4": 163.8752578125,
1250
+ "5": 182.675140625,
1251
+ "6": 207.0015078125,
1252
+ "7": 227.182375,
1253
+ "8": 247.6363125,
1254
+ "9": 263.458296875,
1255
+ "11": 297.0618984375,
1256
+ "12": 302.5270703125,
1257
+ "13": 318.6869609375,
1258
+ "14": 342.1836484375,
1259
+ "15": 353.993734375,
1260
+ "16": 356.512640625,
1261
+ "17": 367.45259375,
1262
+ "18": 372.406625,
1263
+ "19": 371.91615625,
1264
+ "20": 380.96690625
1265
+ },
1266
+ "per_op_loss": {},
1267
+ "per_template": {
1268
+ "crazy_zootopia": {
1269
+ "count": 809728,
1270
+ "correct": 419486,
1271
+ "answer_accuracy": 0.5180579157445463,
1272
+ "avg_response_len": 268.2204962654126,
1273
+ "resp_tokens_sum": 217185646,
1274
+ "pass_at_k": {
1275
+ "pass@1": 0.5180579157445463,
1276
+ "pass@2": 0.598355532161988,
1277
+ "pass@4": 0.6712576062259552,
1278
+ "pass@8": 0.7332857820867824,
1279
+ "pass@16": 0.7829993724706175,
1280
+ "pass@32": 0.8216076402634404,
1281
+ "pass@64": 0.8520991778190156,
1282
+ "pass@128": 0.8763831805248182
1283
+ },
1284
+ "per_op_pass_at_k": {
1285
+ "10": {
1286
+ "pass@1": 0.6446580014749262,
1287
+ "pass@2": 0.7086265765451887,
1288
+ "pass@4": 0.7599441355781849,
1289
+ "pass@8": 0.8023810909443855,
1290
+ "pass@16": 0.8380110251063062,
1291
+ "pass@32": 0.8698539957395686,
1292
+ "pass@64": 0.8976561781654913,
1293
+ "pass@128": 0.9174041297935103
1294
+ },
1295
+ "2": {
1296
+ "pass@1": 0.9338979750778816,
1297
+ "pass@2": 0.9595586657100107,
1298
+ "pass@4": 0.9756525451841886,
1299
+ "pass@8": 0.98409817917281,
1300
+ "pass@16": 0.9875394159838597,
1301
+ "pass@32": 0.9894012652917533,
1302
+ "pass@64": 0.9905213626273273,
1303
+ "pass@128": 0.9906542056074766
1304
+ },
1305
+ "3": {
1306
+ "pass@1": 0.8746655198776758,
1307
+ "pass@2": 0.9018562329938111,
1308
+ "pass@4": 0.9235414171279048,
1309
+ "pass@8": 0.9425918675855139,
1310
+ "pass@16": 0.9583857140671359,
1311
+ "pass@32": 0.9698585704998334,
1312
+ "pass@64": 0.9771319454203935,
1313
+ "pass@128": 0.981651376146789
1314
+ },
1315
+ "4": {
1316
+ "pass@1": 0.78740234375,
1317
+ "pass@2": 0.8135100116879921,
1318
+ "pass@4": 0.8321275546611361,
1319
+ "pass@8": 0.8498828584765968,
1320
+ "pass@16": 0.8688294182277513,
1321
+ "pass@32": 0.8893387219985044,
1322
+ "pass@64": 0.9084337390426462,
1323
+ "pass@128": 0.921875
1324
+ },
1325
+ "5": {
1326
+ "pass@1": 0.6442956349206349,
1327
+ "pass@2": 0.7044240563679539,
1328
+ "pass@4": 0.7589214889805439,
1329
+ "pass@8": 0.8062138260832564,
1330
+ "pass@16": 0.8448474665392458,
1331
+ "pass@32": 0.8742810691280744,
1332
+ "pass@64": 0.8951867234279609,
1333
+ "pass@128": 0.9079365079365079
1334
+ },
1335
+ "6": {
1336
+ "pass@1": 0.6633607784431138,
1337
+ "pass@2": 0.7106940155122822,
1338
+ "pass@4": 0.7466236113262785,
1339
+ "pass@8": 0.7790429849284155,
1340
+ "pass@16": 0.8107574745740735,
1341
+ "pass@32": 0.8402023234939167,
1342
+ "pass@64": 0.8645513213029393,
1343
+ "pass@128": 0.8832335329341318
1344
+ },
1345
+ "7": {
1346
+ "pass@1": 0.6632766272189349,
1347
+ "pass@2": 0.7206016604155985,
1348
+ "pass@4": 0.7671970370382695,
1349
+ "pass@8": 0.8074138976675181,
1350
+ "pass@16": 0.843155207773576,
1351
+ "pass@32": 0.8729361699244267,
1352
+ "pass@64": 0.8974884273605426,
1353
+ "pass@128": 0.9230769230769231
1354
+ },
1355
+ "8": {
1356
+ "pass@1": 0.6051933092948718,
1357
+ "pass@2": 0.6755312436906926,
1358
+ "pass@4": 0.7341284034207262,
1359
+ "pass@8": 0.7835400933327527,
1360
+ "pass@16": 0.8211524415475958,
1361
+ "pass@32": 0.8458170234934912,
1362
+ "pass@64": 0.862760090779185,
1363
+ "pass@128": 0.875
1364
+ },
1365
+ "9": {
1366
+ "pass@1": 0.6189040492957747,
1367
+ "pass@2": 0.6980834812021737,
1368
+ "pass@4": 0.7593200700264576,
1369
+ "pass@8": 0.8095852069004352,
1370
+ "pass@16": 0.8496181125257684,
1371
+ "pass@32": 0.8787592991762989,
1372
+ "pass@64": 0.8996024699023663,
1373
+ "pass@128": 0.9154929577464789
1374
+ },
1375
+ "11": {
1376
+ "pass@1": 0.5309401558073654,
1377
+ "pass@2": 0.6182775311726265,
1378
+ "pass@4": 0.6944196477033849,
1379
+ "pass@8": 0.758216362808835,
1380
+ "pass@16": 0.8089765941091365,
1381
+ "pass@32": 0.8478680917510193,
1382
+ "pass@64": 0.8762482825013892,
1383
+ "pass@128": 0.8951841359773371
1384
+ },
1385
+ "12": {
1386
+ "pass@1": 0.587548552259887,
1387
+ "pass@2": 0.6789871936029179,
1388
+ "pass@4": 0.7542816610317772,
1389
+ "pass@8": 0.8155258561346665,
1390
+ "pass@16": 0.859817387659267,
1391
+ "pass@32": 0.8886653853533955,
1392
+ "pass@64": 0.9077477566906305,
1393
+ "pass@128": 0.9209039548022598
1394
+ },
1395
+ "13": {
1396
+ "pass@1": 0.4798442398119122,
1397
+ "pass@2": 0.6009471496803491,
1398
+ "pass@4": 0.7033698418967224,
1399
+ "pass@8": 0.7829716616327332,
1400
+ "pass@16": 0.8393161383385362,
1401
+ "pass@32": 0.8770397099307982,
1402
+ "pass@64": 0.9037023197930119,
1403
+ "pass@128": 0.9247648902821317
1404
+ },
1405
+ "14": {
1406
+ "pass@1": 0.43294879283489096,
1407
+ "pass@2": 0.555698389628866,
1408
+ "pass@4": 0.6650006631764483,
1409
+ "pass@8": 0.7540915897523429,
1410
+ "pass@16": 0.8223832370157309,
1411
+ "pass@32": 0.8708863421776203,
1412
+ "pass@64": 0.9022030922121614,
1413
+ "pass@128": 0.9221183800623053
1414
+ },
1415
+ "15": {
1416
+ "pass@1": 0.3218253251445087,
1417
+ "pass@2": 0.4450877719493877,
1418
+ "pass@4": 0.5600203219720364,
1419
+ "pass@8": 0.6543575898312131,
1420
+ "pass@16": 0.7259508733588973,
1421
+ "pass@32": 0.7762516558552659,
1422
+ "pass@64": 0.815225131835705,
1423
+ "pass@128": 0.8497109826589595
1424
+ },
1425
+ "16": {
1426
+ "pass@1": 0.32691532258064515,
1427
+ "pass@2": 0.4487609537719074,
1428
+ "pass@4": 0.5625330040398175,
1429
+ "pass@8": 0.654850007133806,
1430
+ "pass@16": 0.7224705015275517,
1431
+ "pass@32": 0.7722011271883154,
1432
+ "pass@64": 0.8137722171021754,
1433
+ "pass@128": 0.8516129032258064
1434
+ },
1435
+ "17": {
1436
+ "pass@1": 0.22821080202312138,
1437
+ "pass@2": 0.34035418847571786,
1438
+ "pass@4": 0.4588241708225774,
1439
+ "pass@8": 0.5640470576174099,
1440
+ "pass@16": 0.6489542613213412,
1441
+ "pass@32": 0.7180814737698088,
1442
+ "pass@64": 0.775694384590296,
1443
+ "pass@128": 0.8179190751445087
1444
+ },
1445
+ "18": {
1446
+ "pass@1": 0.20578125,
1447
+ "pass@2": 0.309048790776153,
1448
+ "pass@4": 0.4157900827575123,
1449
+ "pass@8": 0.509222118234702,
1450
+ "pass@16": 0.5847525333486995,
1451
+ "pass@32": 0.6449860122655471,
1452
+ "pass@64": 0.6961719024625291,
1453
+ "pass@128": 0.74
1454
+ },
1455
+ "19": {
1456
+ "pass@1": 0.17349380630630631,
1457
+ "pass@2": 0.26045693823745014,
1458
+ "pass@4": 0.35654742819309765,
1459
+ "pass@8": 0.4521122354686449,
1460
+ "pass@16": 0.5394334685986726,
1461
+ "pass@32": 0.6138803577519422,
1462
+ "pass@64": 0.6735901886727701,
1463
+ "pass@128": 0.7207207207207207
1464
+ },
1465
+ "20": {
1466
+ "pass@1": 0.16450825825825827,
1467
+ "pass@2": 0.25776945744957575,
1468
+ "pass@4": 0.35973303984524463,
1469
+ "pass@8": 0.45108622829881717,
1470
+ "pass@16": 0.5266710916442264,
1471
+ "pass@32": 0.590217523698882,
1472
+ "pass@64": 0.6487109933142678,
1473
+ "pass@128": 0.7057057057057057
1474
+ }
1475
+ }
1476
+ },
1477
+ "teachers_in_school": {
1478
+ "count": 821120,
1479
+ "correct": 426294,
1480
+ "answer_accuracy": 0.5191616328916602,
1481
+ "avg_response_len": 275.9269522116134,
1482
+ "resp_tokens_sum": 226569139,
1483
+ "pass_at_k": {
1484
+ "pass@1": 0.5191616328916602,
1485
+ "pass@2": 0.5988772393074883,
1486
+ "pass@4": 0.6717313272610219,
1487
+ "pass@8": 0.7333397294950537,
1488
+ "pass@16": 0.7826102324500024,
1489
+ "pass@32": 0.822007771891557,
1490
+ "pass@64": 0.8538428929990424,
1491
+ "pass@128": 0.8799688230709275
1492
+ },
1493
+ "per_op_pass_at_k": {
1494
+ "10": {
1495
+ "pass@1": 0.6655483128834356,
1496
+ "pass@2": 0.7261790644171779,
1497
+ "pass@4": 0.774273106486689,
1498
+ "pass@8": 0.8154503077702322,
1499
+ "pass@16": 0.8513827000228652,
1500
+ "pass@32": 0.880742119528748,
1501
+ "pass@64": 0.9012886048245673,
1502
+ "pass@128": 0.9171779141104295
1503
+ },
1504
+ "2": {
1505
+ "pass@1": 0.9393522036474165,
1506
+ "pass@2": 0.9591322278199266,
1507
+ "pass@4": 0.9724622240419033,
1508
+ "pass@8": 0.9801611460870774,
1509
+ "pass@16": 0.9848945784068139,
1510
+ "pass@32": 0.9885009927054951,
1511
+ "pass@64": 0.9904609452424755,
1512
+ "pass@128": 0.9908814589665653
1513
+ },
1514
+ "3": {
1515
+ "pass@1": 0.8837121212121212,
1516
+ "pass@2": 0.9071820567883556,
1517
+ "pass@4": 0.9250117525082093,
1518
+ "pass@8": 0.938517803565236,
1519
+ "pass@16": 0.9501655320873388,
1520
+ "pass@32": 0.9610128503681503,
1521
+ "pass@64": 0.9710034013222898,
1522
+ "pass@128": 0.9787878787878788
1523
+ },
1524
+ "4": {
1525
+ "pass@1": 0.7681213017751479,
1526
+ "pass@2": 0.806347333201323,
1527
+ "pass@4": 0.8370125978150657,
1528
+ "pass@8": 0.8643633805767534,
1529
+ "pass@16": 0.8894955108387014,
1530
+ "pass@32": 0.9112200498217551,
1531
+ "pass@64": 0.9273172393177025,
1532
+ "pass@128": 0.9408284023668639
1533
+ },
1534
+ "5": {
1535
+ "pass@1": 0.6835597826086957,
1536
+ "pass@2": 0.7398622047244096,
1537
+ "pass@4": 0.7888190215353517,
1538
+ "pass@8": 0.8288249541909402,
1539
+ "pass@16": 0.8586746625294572,
1540
+ "pass@32": 0.8804319932824916,
1541
+ "pass@64": 0.8952530838618767,
1542
+ "pass@128": 0.9043478260869565
1543
+ },
1544
+ "6": {
1545
+ "pass@1": 0.6133372134670487,
1546
+ "pass@2": 0.6639956963653184,
1547
+ "pass@4": 0.7073447105823805,
1548
+ "pass@8": 0.7443909543160218,
1549
+ "pass@16": 0.7771691127933986,
1550
+ "pass@32": 0.8081173959255104,
1551
+ "pass@64": 0.8379673172871838,
1552
+ "pass@128": 0.8681948424068768
1553
+ },
1554
+ "7": {
1555
+ "pass@1": 0.6320899566473989,
1556
+ "pass@2": 0.6949434055118112,
1557
+ "pass@4": 0.7472537736431784,
1558
+ "pass@8": 0.7896054515944385,
1559
+ "pass@16": 0.8240666243617005,
1560
+ "pass@32": 0.8523702333364069,
1561
+ "pass@64": 0.8770839741824703,
1562
+ "pass@128": 0.9046242774566474
1563
+ },
1564
+ "8": {
1565
+ "pass@1": 0.6070422535211267,
1566
+ "pass@2": 0.680067511367417,
1567
+ "pass@4": 0.7394342472155769,
1568
+ "pass@8": 0.7874884126649276,
1569
+ "pass@16": 0.8286021843757088,
1570
+ "pass@32": 0.864914001710332,
1571
+ "pass@64": 0.8928018134050257,
1572
+ "pass@128": 0.9098591549295775
1573
+ },
1574
+ "9": {
1575
+ "pass@1": 0.5982410179640718,
1576
+ "pass@2": 0.6769001367344052,
1577
+ "pass@4": 0.7394855126642103,
1578
+ "pass@8": 0.788722370608343,
1579
+ "pass@16": 0.8276968595373282,
1580
+ "pass@32": 0.8601113228758155,
1581
+ "pass@64": 0.8888388174876141,
1582
+ "pass@128": 0.9131736526946108
1583
+ },
1584
+ "11": {
1585
+ "pass@1": 0.6104103915662651,
1586
+ "pass@2": 0.6973562310264679,
1587
+ "pass@4": 0.7667736087770652,
1588
+ "pass@8": 0.8195464930321138,
1589
+ "pass@16": 0.8576035922567216,
1590
+ "pass@32": 0.8854497881318836,
1591
+ "pass@64": 0.9075046050572457,
1592
+ "pass@128": 0.9246987951807228
1593
+ },
1594
+ "12": {
1595
+ "pass@1": 0.5128085815047022,
1596
+ "pass@2": 0.6041405690765925,
1597
+ "pass@4": 0.685626737978285,
1598
+ "pass@8": 0.7566146072328974,
1599
+ "pass@16": 0.8134178038024091,
1600
+ "pass@32": 0.858042121796473,
1601
+ "pass@64": 0.8908438919484275,
1602
+ "pass@128": 0.9090909090909091
1603
+ },
1604
+ "13": {
1605
+ "pass@1": 0.4427521008403361,
1606
+ "pass@2": 0.5480322542402787,
1607
+ "pass@4": 0.6390175104162397,
1608
+ "pass@8": 0.715290051078792,
1609
+ "pass@16": 0.7766798970641074,
1610
+ "pass@32": 0.8267608641818867,
1611
+ "pass@64": 0.8672715367959482,
1612
+ "pass@128": 0.896358543417367
1613
+ },
1614
+ "14": {
1615
+ "pass@1": 0.43944170321637427,
1616
+ "pass@2": 0.5645552015701978,
1617
+ "pass@4": 0.6751130694737714,
1618
+ "pass@8": 0.7626177000574363,
1619
+ "pass@16": 0.8247123104489158,
1620
+ "pass@32": 0.8640348131883745,
1621
+ "pass@64": 0.8882515823050453,
1622
+ "pass@128": 0.9064327485380117
1623
+ },
1624
+ "15": {
1625
+ "pass@1": 0.34924242424242424,
1626
+ "pass@2": 0.48043500954426144,
1627
+ "pass@4": 0.6070050053402414,
1628
+ "pass@8": 0.7092448683715644,
1629
+ "pass@16": 0.7805712407453619,
1630
+ "pass@32": 0.8284252331696917,
1631
+ "pass@64": 0.8643444343864802,
1632
+ "pass@128": 0.896969696969697
1633
+ },
1634
+ "16": {
1635
+ "pass@1": 0.32687856125356124,
1636
+ "pass@2": 0.4480165640352647,
1637
+ "pass@4": 0.5591375172654703,
1638
+ "pass@8": 0.6514324270194297,
1639
+ "pass@16": 0.7261167053969767,
1640
+ "pass@32": 0.7837995880088544,
1641
+ "pass@64": 0.8244406218061658,
1642
+ "pass@128": 0.8547008547008547
1643
+ },
1644
+ "17": {
1645
+ "pass@1": 0.25883152173913043,
1646
+ "pass@2": 0.3733108004841783,
1647
+ "pass@4": 0.49168759417495167,
1648
+ "pass@8": 0.5938284001034063,
1649
+ "pass@16": 0.6708776227542887,
1650
+ "pass@32": 0.7290766474988695,
1651
+ "pass@64": 0.775271431861227,
1652
+ "pass@128": 0.8136645962732919
1653
+ },
1654
+ "18": {
1655
+ "pass@1": 0.20222886029411766,
1656
+ "pass@2": 0.30483839451134787,
1657
+ "pass@4": 0.41794727552438293,
1658
+ "pass@8": 0.5237158396938546,
1659
+ "pass@16": 0.6092212993975471,
1660
+ "pass@32": 0.6766702683650773,
1661
+ "pass@64": 0.7347166739282228,
1662
+ "pass@128": 0.7823529411764706
1663
+ },
1664
+ "19": {
1665
+ "pass@1": 0.17661830357142858,
1666
+ "pass@2": 0.267703719652231,
1667
+ "pass@4": 0.36584130750397276,
1668
+ "pass@8": 0.455991164869063,
1669
+ "pass@16": 0.5330632685639742,
1670
+ "pass@32": 0.6014918780285793,
1671
+ "pass@64": 0.6652790438088099,
1672
+ "pass@128": 0.7202380952380952
1673
+ },
1674
+ "20": {
1675
+ "pass@1": 0.15938435628742514,
1676
+ "pass@2": 0.23976554764486785,
1677
+ "pass@4": 0.3274774410683695,
1678
+ "pass@8": 0.41152875898223257,
1679
+ "pass@16": 0.48862133484451187,
1680
+ "pass@32": 0.5596604592346828,
1681
+ "pass@64": 0.6251480265327629,
1682
+ "pass@128": 0.688622754491018
1683
+ }
1684
+ }
1685
+ },
1686
+ "movie_festival_awards": {
1687
+ "count": 801152,
1688
+ "correct": 410422,
1689
+ "answer_accuracy": 0.5122898026841348,
1690
+ "avg_response_len": 296.40866027919793,
1691
+ "resp_tokens_sum": 237468391,
1692
+ "pass_at_k": {
1693
+ "pass@1": 0.5122898026841348,
1694
+ "pass@2": 0.5953495863594298,
1695
+ "pass@4": 0.6706734404744416,
1696
+ "pass@8": 0.7329255886524246,
1697
+ "pass@16": 0.7815544694962288,
1698
+ "pass@32": 0.8200175961962155,
1699
+ "pass@64": 0.8513711517126111,
1700
+ "pass@128": 0.8769771528998243
1701
+ },
1702
+ "per_op_pass_at_k": {
1703
+ "10": {
1704
+ "pass@1": 0.6044776119402985,
1705
+ "pass@2": 0.6872576095898459,
1706
+ "pass@4": 0.7601403645439836,
1707
+ "pass@8": 0.8148680866262519,
1708
+ "pass@16": 0.8520847652047296,
1709
+ "pass@32": 0.880442491489034,
1710
+ "pass@64": 0.9045682600821996,
1711
+ "pass@128": 0.9223880597014925
1712
+ },
1713
+ "2": {
1714
+ "pass@1": 0.9144642857142857,
1715
+ "pass@2": 0.9483858267716534,
1716
+ "pass@4": 0.9666809258664094,
1717
+ "pass@8": 0.9760876441015367,
1718
+ "pass@16": 0.9817839791769549,
1719
+ "pass@32": 0.9860892137379141,
1720
+ "pass@64": 0.9910869792252629,
1721
+ "pass@128": 0.9971428571428571
1722
+ },
1723
+ "3": {
1724
+ "pass@1": 0.8872084548104956,
1725
+ "pass@2": 0.911002315718188,
1726
+ "pass@4": 0.9287985389399213,
1727
+ "pass@8": 0.9418294466732434,
1728
+ "pass@16": 0.9516032685115303,
1729
+ "pass@32": 0.959531567551875,
1730
+ "pass@64": 0.9669895513156562,
1731
+ "pass@128": 0.9766763848396501
1732
+ },
1733
+ "4": {
1734
+ "pass@1": 0.7720897295321637,
1735
+ "pass@2": 0.8149048700326933,
1736
+ "pass@4": 0.8442139976472239,
1737
+ "pass@8": 0.8641491648156164,
1738
+ "pass@16": 0.8781066650852173,
1739
+ "pass@32": 0.8898259443521337,
1740
+ "pass@64": 0.9017391378987524,
1741
+ "pass@128": 0.9181286549707602
1742
+ },
1743
+ "5": {
1744
+ "pass@1": 0.6713694852941177,
1745
+ "pass@2": 0.7327079232283464,
1746
+ "pass@4": 0.7863907416903769,
1747
+ "pass@8": 0.8296703924040345,
1748
+ "pass@16": 0.8624125124677811,
1749
+ "pass@32": 0.8867811890994626,
1750
+ "pass@64": 0.9061302658187409,
1751
+ "pass@128": 0.9235294117647059
1752
+ },
1753
+ "6": {
1754
+ "pass@1": 0.6576054810725552,
1755
+ "pass@2": 0.7131347959462478,
1756
+ "pass@4": 0.7573902998915356,
1757
+ "pass@8": 0.7962511637170556,
1758
+ "pass@16": 0.8323717329025123,
1759
+ "pass@32": 0.8641677109359124,
1760
+ "pass@64": 0.8910536305522675,
1761
+ "pass@128": 0.9148264984227129
1762
+ },
1763
+ "7": {
1764
+ "pass@1": 0.6052462420886076,
1765
+ "pass@2": 0.6687863643725713,
1766
+ "pass@4": 0.7221455111821459,
1767
+ "pass@8": 0.7657531013969413,
1768
+ "pass@16": 0.8000025591088098,
1769
+ "pass@32": 0.8273544101920233,
1770
+ "pass@64": 0.8501011080755299,
1771
+ "pass@128": 0.8670886075949367
1772
+ },
1773
+ "8": {
1774
+ "pass@1": 0.6090700075075075,
1775
+ "pass@2": 0.6801576428790994,
1776
+ "pass@4": 0.7411887252832131,
1777
+ "pass@8": 0.7943322020848278,
1778
+ "pass@16": 0.837816999051833,
1779
+ "pass@32": 0.8695440940640597,
1780
+ "pass@64": 0.889940346578068,
1781
+ "pass@128": 0.9039039039039038
1782
+ },
1783
+ "9": {
1784
+ "pass@1": 0.6035219051446945,
1785
+ "pass@2": 0.6816728169228042,
1786
+ "pass@4": 0.7429534720539352,
1787
+ "pass@8": 0.788205651285775,
1788
+ "pass@16": 0.8234507328349505,
1789
+ "pass@32": 0.8527736564709415,
1790
+ "pass@64": 0.8792638805305321,
1791
+ "pass@128": 0.9035369774919614
1792
+ },
1793
+ "11": {
1794
+ "pass@1": 0.5550099206349206,
1795
+ "pass@2": 0.6496914448193972,
1796
+ "pass@4": 0.7281056876818967,
1797
+ "pass@8": 0.7860878864911501,
1798
+ "pass@16": 0.827289435205416,
1799
+ "pass@32": 0.8606638677240043,
1800
+ "pass@64": 0.8882468212477946,
1801
+ "pass@128": 0.9079365079365079
1802
+ },
1803
+ "12": {
1804
+ "pass@1": 0.5942039373088684,
1805
+ "pass@2": 0.6933814322521614,
1806
+ "pass@4": 0.7690625099614834,
1807
+ "pass@8": 0.827877923177547,
1808
+ "pass@16": 0.871842217081767,
1809
+ "pass@32": 0.9032892111773124,
1810
+ "pass@64": 0.9254361399353297,
1811
+ "pass@128": 0.9388379204892966
1812
+ },
1813
+ "13": {
1814
+ "pass@1": 0.47706886574074076,
1815
+ "pass@2": 0.599904232891028,
1816
+ "pass@4": 0.7059600313271024,
1817
+ "pass@8": 0.7874664399728473,
1818
+ "pass@16": 0.8406094473074397,
1819
+ "pass@32": 0.8731865996800195,
1820
+ "pass@64": 0.8968474171671995,
1821
+ "pass@128": 0.9166666666666666
1822
+ },
1823
+ "14": {
1824
+ "pass@1": 0.39002225519287836,
1825
+ "pass@2": 0.5174949327087078,
1826
+ "pass@4": 0.6311075550830626,
1827
+ "pass@8": 0.7203883307676154,
1828
+ "pass@16": 0.7856525561887167,
1829
+ "pass@32": 0.8349817202746366,
1830
+ "pass@64": 0.8723556036560305,
1831
+ "pass@128": 0.9020771513353115
1832
+ },
1833
+ "15": {
1834
+ "pass@1": 0.32959587191358025,
1835
+ "pass@2": 0.45167026647710673,
1836
+ "pass@4": 0.5726770299545887,
1837
+ "pass@8": 0.6745838353015993,
1838
+ "pass@16": 0.7441054148500932,
1839
+ "pass@32": 0.7882812189193158,
1840
+ "pass@64": 0.8246829582330772,
1841
+ "pass@128": 0.8611111111111112
1842
+ },
1843
+ "16": {
1844
+ "pass@1": 0.2981655604719764,
1845
+ "pass@2": 0.4289886883608576,
1846
+ "pass@4": 0.5567347947656982,
1847
+ "pass@8": 0.661641142873429,
1848
+ "pass@16": 0.741565943850342,
1849
+ "pass@32": 0.8031266684588632,
1850
+ "pass@64": 0.8499748456042655,
1851
+ "pass@128": 0.8820058997050148
1852
+ },
1853
+ "17": {
1854
+ "pass@1": 0.20107774849397592,
1855
+ "pass@2": 0.3054575585807797,
1856
+ "pass@4": 0.4184178767940154,
1857
+ "pass@8": 0.5233287133157847,
1858
+ "pass@16": 0.614407303131194,
1859
+ "pass@32": 0.6902430500898341,
1860
+ "pass@64": 0.7477831922675113,
1861
+ "pass@128": 0.7891566265060241
1862
+ },
1863
+ "18": {
1864
+ "pass@1": 0.19533770161290323,
1865
+ "pass@2": 0.29267208534417066,
1866
+ "pass@4": 0.3930905977478621,
1867
+ "pass@8": 0.4825708236511399,
1868
+ "pass@16": 0.5626216910380368,
1869
+ "pass@32": 0.6369578833004578,
1870
+ "pass@64": 0.6995287861787093,
1871
+ "pass@128": 0.7451612903225806
1872
+ },
1873
+ "19": {
1874
+ "pass@1": 0.16403889728096677,
1875
+ "pass@2": 0.2507668095963082,
1876
+ "pass@4": 0.34520906129180984,
1877
+ "pass@8": 0.4333241918432945,
1878
+ "pass@16": 0.510047346615244,
1879
+ "pass@32": 0.5768047901931338,
1880
+ "pass@64": 0.6384011257652897,
1881
+ "pass@128": 0.6888217522658611
1882
+ },
1883
+ "20": {
1884
+ "pass@1": 0.15728228228228228,
1885
+ "pass@2": 0.24287119304343707,
1886
+ "pass@4": 0.33739998857124837,
1887
+ "pass@8": 0.42697742424851853,
1888
+ "pass@16": 0.5059504946333598,
1889
+ "pass@32": 0.5745353026068557,
1890
+ "pass@64": 0.6337007460340792,
1891
+ "pass@128": 0.6876876876876877
1892
+ }
1893
+ }
1894
+ }
1895
+ }
1896
+ }
1897
+ }