ZachW commited on
Commit
94dd7a4
·
verified ·
1 Parent(s): 0e7c75c

Upload extrapolation artifact: 050302 RL OP11-14 step200 metrics

Browse files
050302/evals/rl050302_op11_14_step200_merged_hf/rl050302_op11_14_step200_merged_hf_metrics.json ADDED
@@ -0,0 +1,1897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "rl050302_op11_14_step200_merged_hf",
3
+ "id": {
4
+ "answer_accuracy": 0.7564635416666666,
5
+ "avg_response_len": 208.0127170138889,
6
+ "avg_loss": 0.0,
7
+ "count": 1152000,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {
10
+ "pass@1": 0.7564635416666666,
11
+ "pass@2": 0.7959106791338477,
12
+ "pass@4": 0.8248964637232805,
13
+ "pass@8": 0.846691977659791,
14
+ "pass@16": 0.863269099732004,
15
+ "pass@32": 0.8767512914305521,
16
+ "pass@64": 0.888669030520124,
17
+ "pass@128": 0.8996666666666666
18
+ },
19
+ "per_op_pass_at_k": {
20
+ "10": {
21
+ "pass@1": 0.6998359375,
22
+ "pass@2": 0.7497196112204725,
23
+ "pass@4": 0.7877674823772035,
24
+ "pass@8": 0.8178490135167437,
25
+ "pass@16": 0.8425426476657046,
26
+ "pass@32": 0.8647230780096713,
27
+ "pass@64": 0.8846169498619155,
28
+ "pass@128": 0.902
29
+ },
30
+ "2": {
31
+ "pass@1": 0.934484375,
32
+ "pass@2": 0.9583288631889764,
33
+ "pass@4": 0.9730320496812903,
34
+ "pass@8": 0.9827350742606419,
35
+ "pass@16": 0.9888035787644653,
36
+ "pass@32": 0.9924179792156923,
37
+ "pass@64": 0.9947161450381122,
38
+ "pass@128": 0.996
39
+ },
40
+ "3": {
41
+ "pass@1": 0.9164453125,
42
+ "pass@2": 0.9411937746062994,
43
+ "pass@4": 0.9579434554743155,
44
+ "pass@8": 0.9699985325178049,
45
+ "pass@16": 0.9781683206395383,
46
+ "pass@32": 0.9838437894282911,
47
+ "pass@64": 0.9884535535906521,
48
+ "pass@128": 0.992
49
+ },
50
+ "4": {
51
+ "pass@1": 0.8259296875,
52
+ "pass@2": 0.8585537647637796,
53
+ "pass@4": 0.8806184625046869,
54
+ "pass@8": 0.8961674154317582,
55
+ "pass@16": 0.9082188111888009,
56
+ "pass@32": 0.9181050438511107,
57
+ "pass@64": 0.9259375783769267,
58
+ "pass@128": 0.933
59
+ },
60
+ "5": {
61
+ "pass@1": 0.698609375,
62
+ "pass@2": 0.7370525344488192,
63
+ "pass@4": 0.7667280059992501,
64
+ "pass@8": 0.7889732628668457,
65
+ "pass@16": 0.8060324371228792,
66
+ "pass@32": 0.8203173429933404,
67
+ "pass@64": 0.8325269722338159,
68
+ "pass@128": 0.843
69
+ },
70
+ "6": {
71
+ "pass@1": 0.6761796875,
72
+ "pass@2": 0.7140426919291342,
73
+ "pass@4": 0.7424857538432698,
74
+ "pass@8": 0.7651357504137184,
75
+ "pass@16": 0.7835360752677899,
76
+ "pass@32": 0.7994032050618843,
77
+ "pass@64": 0.8150506157082538,
78
+ "pass@128": 0.832
79
+ },
80
+ "7": {
81
+ "pass@1": 0.6889375,
82
+ "pass@2": 0.7267791584645674,
83
+ "pass@4": 0.7546797299400082,
84
+ "pass@8": 0.7755084684065878,
85
+ "pass@16": 0.7920948216116958,
86
+ "pass@32": 0.8059412816926678,
87
+ "pass@64": 0.8184436972940266,
88
+ "pass@128": 0.83
89
+ },
90
+ "8": {
91
+ "pass@1": 0.68353125,
92
+ "pass@2": 0.7357874015748035,
93
+ "pass@4": 0.7764557210348707,
94
+ "pass@8": 0.8079708875341801,
95
+ "pass@16": 0.8317391404315082,
96
+ "pass@32": 0.8509228087163498,
97
+ "pass@64": 0.8678658647215115,
98
+ "pass@128": 0.883
99
+ },
100
+ "9": {
101
+ "pass@1": 0.68421875,
102
+ "pass@2": 0.7417383120078741,
103
+ "pass@4": 0.7843575126546682,
104
+ "pass@8": 0.8158893939898693,
105
+ "pass@16": 0.8382860648956109,
106
+ "pass@32": 0.8550870939059481,
107
+ "pass@64": 0.870409897855889,
108
+ "pass@128": 0.886
109
+ }
110
+ },
111
+ "per_op_accuracy": {
112
+ "10": 0.6998359375,
113
+ "2": 0.934484375,
114
+ "3": 0.9164453125,
115
+ "4": 0.8259296875,
116
+ "5": 0.698609375,
117
+ "6": 0.6761796875,
118
+ "7": 0.6889375,
119
+ "8": 0.68353125,
120
+ "9": 0.68421875
121
+ },
122
+ "per_op_avg_response_len": {
123
+ "10": 279.936125,
124
+ "2": 140.577265625,
125
+ "3": 158.07409375,
126
+ "4": 166.3575625,
127
+ "5": 183.2572890625,
128
+ "6": 208.2703046875,
129
+ "7": 227.0995625,
130
+ "8": 247.07909375,
131
+ "9": 261.46315625
132
+ },
133
+ "per_op_avg_loss": {},
134
+ "per_op_length": {
135
+ "10": 279.936125,
136
+ "2": 140.577265625,
137
+ "3": 158.07409375,
138
+ "4": 166.3575625,
139
+ "5": 183.2572890625,
140
+ "6": 208.2703046875,
141
+ "7": 227.0995625,
142
+ "8": 247.07909375,
143
+ "9": 261.46315625
144
+ },
145
+ "per_op_loss": {},
146
+ "per_template": {
147
+ "crazy_zootopia": {
148
+ "count": 379008,
149
+ "correct": 288098,
150
+ "answer_accuracy": 0.7601369891928402,
151
+ "avg_response_len": 198.4661880488011,
152
+ "resp_tokens_sum": 75220273,
153
+ "pass_at_k": {
154
+ "pass@1": 0.7601369891928402,
155
+ "pass@2": 0.8015735483197601,
156
+ "pass@4": 0.8309645480295206,
157
+ "pass@8": 0.8525047419348892,
158
+ "pass@16": 0.8688200752843309,
159
+ "pass@32": 0.8820476346903116,
160
+ "pass@64": 0.8936358462855198,
161
+ "pass@128": 0.9047619047619048
162
+ },
163
+ "per_op_pass_at_k": {
164
+ "10": {
165
+ "pass@1": 0.7018805309734514,
166
+ "pass@2": 0.7549985628179221,
167
+ "pass@4": 0.794857199542535,
168
+ "pass@8": 0.8246423155165492,
169
+ "pass@16": 0.8477130887947533,
170
+ "pass@32": 0.8687864891718109,
171
+ "pass@64": 0.8890464019643919,
172
+ "pass@128": 0.9085545722713865
173
+ },
174
+ "2": {
175
+ "pass@1": 0.9495716510903427,
176
+ "pass@2": 0.9728514331444554,
177
+ "pass@4": 0.9851647276567062,
178
+ "pass@8": 0.992077485746166,
179
+ "pass@16": 0.9954145881797847,
180
+ "pass@32": 0.996585667661237,
181
+ "pass@64": 0.9968750895206262,
182
+ "pass@128": 0.9968847352024922
183
+ },
184
+ "3": {
185
+ "pass@1": 0.9074445718654435,
186
+ "pass@2": 0.9352523236774299,
187
+ "pass@4": 0.953601489034054,
188
+ "pass@8": 0.9679758338381604,
189
+ "pass@16": 0.9790672679207248,
190
+ "pass@32": 0.9871383226541847,
191
+ "pass@64": 0.9930782875203557,
192
+ "pass@128": 0.9969418960244648
193
+ },
194
+ "4": {
195
+ "pass@1": 0.8296630859375,
196
+ "pass@2": 0.8626487912155513,
197
+ "pass@4": 0.8834582688882635,
198
+ "pass@8": 0.895808641913504,
199
+ "pass@16": 0.9041949013259944,
200
+ "pass@32": 0.9123519155789825,
201
+ "pass@64": 0.9217296599331064,
202
+ "pass@128": 0.93125
203
+ },
204
+ "5": {
205
+ "pass@1": 0.6723710317460317,
206
+ "pass@2": 0.7115751156105488,
207
+ "pass@4": 0.7433334842073313,
208
+ "pass@8": 0.768306546362202,
209
+ "pass@16": 0.7871896199409257,
210
+ "pass@32": 0.8020106421329306,
211
+ "pass@64": 0.8145428062453238,
212
+ "pass@128": 0.8285714285714286
213
+ },
214
+ "6": {
215
+ "pass@1": 0.6831259356287425,
216
+ "pass@2": 0.7225941817153095,
217
+ "pass@4": 0.7511133829079745,
218
+ "pass@8": 0.7732136904251459,
219
+ "pass@16": 0.7908965515084061,
220
+ "pass@32": 0.8059498542595275,
221
+ "pass@64": 0.8190138601274597,
222
+ "pass@128": 0.8323353293413174
223
+ },
224
+ "7": {
225
+ "pass@1": 0.7232572115384616,
226
+ "pass@2": 0.7591458265386947,
227
+ "pass@4": 0.7840269109297723,
228
+ "pass@8": 0.8025467115547503,
229
+ "pass@16": 0.8182715609954336,
230
+ "pass@32": 0.8323267215314273,
231
+ "pass@64": 0.8455724214427237,
232
+ "pass@128": 0.8579881656804734
233
+ },
234
+ "8": {
235
+ "pass@1": 0.6790364583333334,
236
+ "pass@2": 0.7368809780688469,
237
+ "pass@4": 0.7800951419534093,
238
+ "pass@8": 0.8111498024624929,
239
+ "pass@16": 0.8340518273159697,
240
+ "pass@32": 0.851499152999998,
241
+ "pass@64": 0.8652018097269851,
242
+ "pass@128": 0.8782051282051282
243
+ },
244
+ "9": {
245
+ "pass@1": 0.7028389084507042,
246
+ "pass@2": 0.7644043889320176,
247
+ "pass@4": 0.808012754743685,
248
+ "pass@8": 0.8407312628937469,
249
+ "pass@16": 0.8654983711299619,
250
+ "pass@32": 0.8838276379384431,
251
+ "pass@64": 0.8988843924487747,
252
+ "pass@128": 0.9126760563380282
253
+ }
254
+ }
255
+ },
256
+ "teachers_in_school": {
257
+ "count": 390656,
258
+ "correct": 296327,
259
+ "answer_accuracy": 0.7585369225098296,
260
+ "avg_response_len": 204.60524604767366,
261
+ "resp_tokens_sum": 79930267,
262
+ "pass_at_k": {
263
+ "pass@1": 0.7585369225098296,
264
+ "pass@2": 0.7960220031269044,
265
+ "pass@4": 0.8238626931033214,
266
+ "pass@8": 0.845128096252563,
267
+ "pass@16": 0.8612886172873727,
268
+ "pass@32": 0.874135681726868,
269
+ "pass@64": 0.8852335883130494,
270
+ "pass@128": 0.8951507208387942
271
+ },
272
+ "per_op_pass_at_k": {
273
+ "10": {
274
+ "pass@1": 0.7098590874233128,
275
+ "pass@2": 0.7535682907830544,
276
+ "pass@4": 0.7864975527407232,
277
+ "pass@8": 0.8135336785898025,
278
+ "pass@16": 0.8366857190055628,
279
+ "pass@32": 0.8580710264821365,
280
+ "pass@64": 0.8776973706717127,
281
+ "pass@128": 0.8957055214723927
282
+ },
283
+ "2": {
284
+ "pass@1": 0.9405395136778115,
285
+ "pass@2": 0.9620412308833738,
286
+ "pass@4": 0.9758225314920738,
287
+ "pass@8": 0.9849524204535265,
288
+ "pass@16": 0.9900649654565532,
289
+ "pass@32": 0.9921000445202469,
290
+ "pass@64": 0.9931668037952412,
291
+ "pass@128": 0.993920972644377
292
+ },
293
+ "3": {
294
+ "pass@1": 0.9234375,
295
+ "pass@2": 0.9456786119064663,
296
+ "pass@4": 0.9617405105611796,
297
+ "pass@8": 0.9726108401727136,
298
+ "pass@16": 0.9779520922364761,
299
+ "pass@32": 0.9803488524936297,
300
+ "pass@64": 0.9825398325449198,
301
+ "pass@128": 0.9848484848484849
302
+ },
303
+ "4": {
304
+ "pass@1": 0.8384106878698225,
305
+ "pass@2": 0.8696401219540603,
306
+ "pass@4": 0.8919865230418679,
307
+ "pass@8": 0.9109039824674823,
308
+ "pass@16": 0.9268934190930932,
309
+ "pass@32": 0.939480395901062,
310
+ "pass@64": 0.9475861428402855,
311
+ "pass@128": 0.9526627218934911
312
+ },
313
+ "5": {
314
+ "pass@1": 0.7247509057971014,
315
+ "pass@2": 0.7569896154285061,
316
+ "pass@4": 0.7794320843046791,
317
+ "pass@8": 0.7960690607966524,
318
+ "pass@16": 0.8107549025408636,
319
+ "pass@32": 0.8248985765960865,
320
+ "pass@64": 0.8371374039094492,
321
+ "pass@128": 0.8463768115942029
322
+ },
323
+ "6": {
324
+ "pass@1": 0.6351181948424068,
325
+ "pass@2": 0.6731091927441734,
326
+ "pass@4": 0.7035691777782792,
327
+ "pass@8": 0.7284657829447492,
328
+ "pass@16": 0.7492790741526567,
329
+ "pass@32": 0.7669448136688961,
330
+ "pass@64": 0.7842548351421045,
331
+ "pass@128": 0.8022922636103151
332
+ },
333
+ "7": {
334
+ "pass@1": 0.6782649927745664,
335
+ "pass@2": 0.7190959816348825,
336
+ "pass@4": 0.7507817206555822,
337
+ "pass@8": 0.7737670955280236,
338
+ "pass@16": 0.7898595118415901,
339
+ "pass@32": 0.8002346239143221,
340
+ "pass@64": 0.8076964146996783,
341
+ "pass@128": 0.815028901734104
342
+ },
343
+ "8": {
344
+ "pass@1": 0.6978653169014084,
345
+ "pass@2": 0.7489800515692578,
346
+ "pass@4": 0.7882377675501828,
347
+ "pass@8": 0.8181867244335217,
348
+ "pass@16": 0.8403183790454402,
349
+ "pass@32": 0.8589868896024869,
350
+ "pass@64": 0.876235649537859,
351
+ "pass@128": 0.8901408450704226
352
+ },
353
+ "9": {
354
+ "pass@1": 0.6945172155688623,
355
+ "pass@2": 0.7500011050733182,
356
+ "pass@4": 0.7906428624752743,
357
+ "pass@8": 0.82085154021908,
358
+ "pass@16": 0.8423189279171873,
359
+ "pass@32": 0.8580696456086162,
360
+ "pass@64": 0.8721121926305166,
361
+ "pass@128": 0.8862275449101796
362
+ }
363
+ }
364
+ },
365
+ "movie_festival_awards": {
366
+ "count": 382336,
367
+ "correct": 287021,
368
+ "answer_accuracy": 0.7507035696350853,
369
+ "avg_response_len": 220.95777012889187,
370
+ "resp_tokens_sum": 84480110,
371
+ "pass_at_k": {
372
+ "pass@1": 0.7507035696350853,
373
+ "pass@2": 0.7901833552348897,
374
+ "pass@4": 0.8199374648285309,
375
+ "pass@8": 0.842527722901281,
376
+ "pass@16": 0.8597900216638895,
377
+ "pass@32": 0.874173577477915,
378
+ "pass@64": 0.8872556485765796,
379
+ "pass@128": 0.8992299966521593
380
+ },
381
+ "per_op_pass_at_k": {
382
+ "10": {
383
+ "pass@1": 0.6880130597014925,
384
+ "pass@2": 0.7406323451639437,
385
+ "pass@4": 0.7818289239964403,
386
+ "pass@8": 0.8151739980189773,
387
+ "pass@16": 0.8430100482640915,
388
+ "pass@32": 0.867084488200749,
389
+ "pass@64": 0.8868682888268872,
390
+ "pass@128": 0.9014925373134328
391
+ },
392
+ "2": {
393
+ "pass@1": 0.9149553571428571,
394
+ "pass@2": 0.941519966254218,
395
+ "pass@4": 0.9592815978359845,
396
+ "pass@8": 0.9720824428768929,
397
+ "pass@16": 0.9815546352101394,
398
+ "pass@32": 0.9888944721407834,
399
+ "pass@64": 0.9941924652953058,
400
+ "pass@128": 0.9971428571428571
401
+ },
402
+ "3": {
403
+ "pass@1": 0.9182990160349854,
404
+ "pass@2": 0.9425432152613579,
405
+ "pass@4": 0.9584297378279611,
406
+ "pass@8": 0.9694135789962994,
407
+ "pass@16": 0.9775193399166878,
408
+ "pass@32": 0.9840654128206252,
409
+ "pass@64": 0.9897341365354874,
410
+ "pass@128": 0.9941690962099126
411
+ },
412
+ "4": {
413
+ "pass@1": 0.8101014254385965,
414
+ "pass@2": 0.8437654688723117,
415
+ "pass@4": 0.8667262329599149,
416
+ "pass@8": 0.8819389004252276,
417
+ "pass@16": 0.8935276816111616,
418
+ "pass@32": 0.9023627399160156,
419
+ "pass@64": 0.9084794471295793,
420
+ "pass@128": 0.9152046783625731
421
+ },
422
+ "5": {
423
+ "pass@1": 0.6963924632352941,
424
+ "pass@2": 0.7404263403195928,
425
+ "pass@4": 0.775511439378901,
426
+ "pass@8": 0.8009202199644325,
427
+ "pass@16": 0.8186978396026182,
428
+ "pass@32": 0.8326293582229927,
429
+ "pass@64": 0.8445105409346427,
430
+ "pass@128": 0.8529411764705882
431
+ },
432
+ "6": {
433
+ "pass@1": 0.714067429022082,
434
+ "pass@2": 0.7500981923296656,
435
+ "pass@4": 0.7762405075942792,
436
+ "pass@8": 0.7969962762271371,
437
+ "pass@16": 0.8134959942735189,
438
+ "pass@32": 0.8282404219834616,
439
+ "pass@64": 0.8447793342621057,
440
+ "pass@128": 0.8643533123028391
441
+ },
442
+ "7": {
443
+ "pass@1": 0.6639141613924051,
444
+ "pass@2": 0.7005717071165156,
445
+ "pass@4": 0.727557464237065,
446
+ "pass@8": 0.7484945090138806,
447
+ "pass@16": 0.7665431737280053,
448
+ "pass@32": 0.7839673099388944,
449
+ "pass@64": 0.8011938587351172,
450
+ "pass@128": 0.8164556962025317
451
+ },
452
+ "8": {
453
+ "pass@1": 0.672461524024024,
454
+ "pass@2": 0.7206985528835925,
455
+ "pass@4": 0.7604853731684436,
456
+ "pass@8": 0.7941016876636111,
457
+ "pass@16": 0.8204262635068897,
458
+ "pass@32": 0.8417859674818847,
459
+ "pass@64": 0.8614391726749312,
460
+ "pass@128": 0.8798798798798799
461
+ },
462
+ "9": {
463
+ "pass@1": 0.6519041398713826,
464
+ "pass@2": 0.7069915911335041,
465
+ "pass@4": 0.7506053654466821,
466
+ "pass@8": 0.7822037981653261,
467
+ "pass@16": 0.8028926727335497,
468
+ "pass@32": 0.8190772373135783,
469
+ "pass@64": 0.8360786694468839,
470
+ "pass@128": 0.8553054662379421
471
+ }
472
+ }
473
+ }
474
+ }
475
+ },
476
+ "ood": {
477
+ "answer_accuracy": 0.34808203125,
478
+ "avg_response_len": 342.315659375,
479
+ "avg_loss": 0.0,
480
+ "count": 1280000,
481
+ "validation_metrics": {},
482
+ "pass_at_k": {
483
+ "pass@1": 0.34808203125,
484
+ "pass@2": 0.43144087106299783,
485
+ "pass@4": 0.5086651600768592,
486
+ "pass@8": 0.5759537329045272,
487
+ "pass@16": 0.6320354790107239,
488
+ "pass@32": 0.6779582503852143,
489
+ "pass@64": 0.7161277856720875,
490
+ "pass@128": 0.7489
491
+ },
492
+ "per_op_pass_at_k": {
493
+ "11": {
494
+ "pass@1": 0.645328125,
495
+ "pass@2": 0.7102300688976382,
496
+ "pass@4": 0.7585581894450686,
497
+ "pass@8": 0.7949712730733313,
498
+ "pass@16": 0.8232239283653476,
499
+ "pass@32": 0.8467896281680857,
500
+ "pass@64": 0.8677760843111625,
501
+ "pass@128": 0.887
502
+ },
503
+ "12": {
504
+ "pass@1": 0.6262265625,
505
+ "pass@2": 0.6924527559055124,
506
+ "pass@4": 0.7439660961754783,
507
+ "pass@8": 0.7881987997956229,
508
+ "pass@16": 0.8274650337042867,
509
+ "pass@32": 0.8621452294532637,
510
+ "pass@64": 0.8899411679493481,
511
+ "pass@128": 0.91
512
+ },
513
+ "13": {
514
+ "pass@1": 0.515390625,
515
+ "pass@2": 0.6064320866141737,
516
+ "pass@4": 0.6819955449006371,
517
+ "pass@8": 0.7454696905246248,
518
+ "pass@16": 0.7970554598668151,
519
+ "pass@32": 0.8370062209346598,
520
+ "pass@64": 0.8680208840016798,
521
+ "pass@128": 0.894
522
+ },
523
+ "14": {
524
+ "pass@1": 0.4509453125,
525
+ "pass@2": 0.5587468011811019,
526
+ "pass@4": 0.6510836965691776,
527
+ "pass@8": 0.7264001469285997,
528
+ "pass@16": 0.7848350439768886,
529
+ "pass@32": 0.8277722436665917,
530
+ "pass@64": 0.8584198739625949,
531
+ "pass@128": 0.88
532
+ },
533
+ "15": {
534
+ "pass@1": 0.342046875,
535
+ "pass@2": 0.45342249015748054,
536
+ "pass@4": 0.5530513783277083,
537
+ "pass@8": 0.634712334515626,
538
+ "pass@16": 0.699178504064928,
539
+ "pass@32": 0.7476314759907243,
540
+ "pass@64": 0.7840614124927014,
541
+ "pass@128": 0.818
542
+ },
543
+ "16": {
544
+ "pass@1": 0.2568984375,
545
+ "pass@2": 0.3605068897637793,
546
+ "pass@4": 0.4633091179227593,
547
+ "pass@8": 0.5545054725122883,
548
+ "pass@16": 0.6304599364492406,
549
+ "pass@32": 0.6926131635972946,
550
+ "pass@64": 0.7438924411504231,
551
+ "pass@128": 0.786
552
+ },
553
+ "17": {
554
+ "pass@1": 0.180546875,
555
+ "pass@2": 0.26008489173228355,
556
+ "pass@4": 0.34560696259842494,
557
+ "pass@8": 0.4284562778464315,
558
+ "pass@16": 0.5018989467657816,
559
+ "pass@32": 0.5645239509858958,
560
+ "pass@64": 0.6209294612489322,
561
+ "pass@128": 0.674
562
+ },
563
+ "18": {
564
+ "pass@1": 0.162046875,
565
+ "pass@2": 0.2337256397637794,
566
+ "pass@4": 0.3077259774090736,
567
+ "pass@8": 0.3768791242808825,
568
+ "pass@16": 0.4392902527391763,
569
+ "pass@32": 0.4956244462569732,
570
+ "pass@64": 0.5464061221745686,
571
+ "pass@128": 0.591
572
+ },
573
+ "19": {
574
+ "pass@1": 0.1514296875,
575
+ "pass@2": 0.2191755659448819,
576
+ "pass@4": 0.2889380630858642,
577
+ "pass@8": 0.3527730781336148,
578
+ "pass@16": 0.40723262416432515,
579
+ "pass@32": 0.45234964946368117,
580
+ "pass@64": 0.4902811911999381,
581
+ "pass@128": 0.522
582
+ },
583
+ "20": {
584
+ "pass@1": 0.1499609375,
585
+ "pass@2": 0.21963152066929123,
586
+ "pass@4": 0.29241657433445783,
587
+ "pass@8": 0.35717113143418483,
588
+ "pass@16": 0.4097150600102705,
589
+ "pass@32": 0.4531264953349681,
590
+ "pass@64": 0.49154921822946673,
591
+ "pass@128": 0.527
592
+ }
593
+ },
594
+ "per_op_accuracy": {
595
+ "11": 0.645328125,
596
+ "12": 0.6262265625,
597
+ "13": 0.515390625,
598
+ "14": 0.4509453125,
599
+ "15": 0.342046875,
600
+ "16": 0.2568984375,
601
+ "17": 0.180546875,
602
+ "18": 0.162046875,
603
+ "19": 0.1514296875,
604
+ "20": 0.1499609375
605
+ },
606
+ "per_op_avg_response_len": {
607
+ "11": 296.783828125,
608
+ "12": 305.03403125,
609
+ "13": 322.5857421875,
610
+ "14": 343.72453125,
611
+ "15": 352.05109375,
612
+ "16": 356.360015625,
613
+ "17": 360.660140625,
614
+ "18": 363.015453125,
615
+ "19": 357.965421875,
616
+ "20": 364.9763359375
617
+ },
618
+ "per_op_avg_loss": {},
619
+ "per_op_length": {
620
+ "11": 296.783828125,
621
+ "12": 305.03403125,
622
+ "13": 322.5857421875,
623
+ "14": 343.72453125,
624
+ "15": 352.05109375,
625
+ "16": 356.360015625,
626
+ "17": 360.660140625,
627
+ "18": 363.015453125,
628
+ "19": 357.965421875,
629
+ "20": 364.9763359375
630
+ },
631
+ "per_op_loss": {},
632
+ "per_template": {
633
+ "crazy_zootopia": {
634
+ "count": 430720,
635
+ "correct": 150791,
636
+ "answer_accuracy": 0.35009054606240714,
637
+ "avg_response_len": 326.6123723068351,
638
+ "resp_tokens_sum": 140678481,
639
+ "pass_at_k": {
640
+ "pass@1": 0.35009054606240714,
641
+ "pass@2": 0.4326315358425677,
642
+ "pass@4": 0.5097523258515436,
643
+ "pass@8": 0.5777330244485334,
644
+ "pass@16": 0.6346750195563541,
645
+ "pass@32": 0.6812510978427386,
646
+ "pass@64": 0.7197941081379721,
647
+ "pass@128": 0.7527488855869242
648
+ },
649
+ "per_op_pass_at_k": {
650
+ "11": {
651
+ "pass@1": 0.607471671388102,
652
+ "pass@2": 0.6684865801565879,
653
+ "pass@4": 0.7145608326933632,
654
+ "pass@8": 0.751410226301821,
655
+ "pass@16": 0.7813128542866512,
656
+ "pass@32": 0.8077145164088809,
657
+ "pass@64": 0.8342378682376506,
658
+ "pass@128": 0.8611898016997167
659
+ },
660
+ "12": {
661
+ "pass@1": 0.6359242584745762,
662
+ "pass@2": 0.7010140019573821,
663
+ "pass@4": 0.7524318914584823,
664
+ "pass@8": 0.7966960762341885,
665
+ "pass@16": 0.8349907268657351,
666
+ "pass@32": 0.8675260403092755,
667
+ "pass@64": 0.8927480390421706,
668
+ "pass@128": 0.9124293785310734
669
+ },
670
+ "13": {
671
+ "pass@1": 0.5220905172413793,
672
+ "pass@2": 0.620534226667983,
673
+ "pass@4": 0.7027034552695014,
674
+ "pass@8": 0.770057213759952,
675
+ "pass@16": 0.8241495227282288,
676
+ "pass@32": 0.8631455967014058,
677
+ "pass@64": 0.8879219996985025,
678
+ "pass@128": 0.9090909090909091
679
+ },
680
+ "14": {
681
+ "pass@1": 0.4550963785046729,
682
+ "pass@2": 0.5581134097431748,
683
+ "pass@4": 0.647930787565105,
684
+ "pass@8": 0.7233182340799785,
685
+ "pass@16": 0.7833892226203504,
686
+ "pass@32": 0.8279559428456201,
687
+ "pass@64": 0.8590417122874536,
688
+ "pass@128": 0.881619937694704
689
+ },
690
+ "15": {
691
+ "pass@1": 0.34934971098265893,
692
+ "pass@2": 0.45711498964544184,
693
+ "pass@4": 0.5511690342356048,
694
+ "pass@8": 0.6290683831435004,
695
+ "pass@16": 0.6933893369608615,
696
+ "pass@32": 0.7435434683159998,
697
+ "pass@64": 0.7816914947898684,
698
+ "pass@128": 0.8179190751445087
699
+ },
700
+ "16": {
701
+ "pass@1": 0.2809475806451613,
702
+ "pass@2": 0.38555094297688597,
703
+ "pass@4": 0.4893392721071152,
704
+ "pass@8": 0.5826136332644718,
705
+ "pass@16": 0.6611631758264305,
706
+ "pass@32": 0.7266388086540002,
707
+ "pass@64": 0.7824484578306036,
708
+ "pass@128": 0.832258064516129
709
+ },
710
+ "17": {
711
+ "pass@1": 0.1666365606936416,
712
+ "pass@2": 0.2457618138682809,
713
+ "pass@4": 0.33538570187758293,
714
+ "pass@8": 0.42521167689535794,
715
+ "pass@16": 0.5048829056797672,
716
+ "pass@32": 0.5706106820406163,
717
+ "pass@64": 0.6265261197812971,
718
+ "pass@128": 0.6734104046242775
719
+ },
720
+ "18": {
721
+ "pass@1": 0.15660714285714286,
722
+ "pass@2": 0.2282670134983126,
723
+ "pass@4": 0.3032608698912637,
724
+ "pass@8": 0.37353183319352135,
725
+ "pass@16": 0.43836449791082005,
726
+ "pass@32": 0.4984786995754745,
727
+ "pass@64": 0.5524853165212575,
728
+ "pass@128": 0.5942857142857143
729
+ },
730
+ "19": {
731
+ "pass@1": 0.16272522522522523,
732
+ "pass@2": 0.23145808209784588,
733
+ "pass@4": 0.30189558653141335,
734
+ "pass@8": 0.36420667880460095,
735
+ "pass@16": 0.4138035379213277,
736
+ "pass@32": 0.4535217258277585,
737
+ "pass@64": 0.48873324942873747,
738
+ "pass@128": 0.5225225225225225
739
+ },
740
+ "20": {
741
+ "pass@1": 0.15388044294294295,
742
+ "pass@2": 0.2248661062637441,
743
+ "pass@4": 0.29865429462083,
744
+ "pass@8": 0.3650207487197481,
745
+ "pass@16": 0.41781596055733294,
746
+ "pass@32": 0.4613199772361069,
747
+ "pass@64": 0.5005061369851226,
748
+ "pass@128": 0.5315315315315315
749
+ }
750
+ }
751
+ },
752
+ "teachers_in_school": {
753
+ "count": 430464,
754
+ "correct": 149593,
755
+ "answer_accuracy": 0.3475157039845376,
756
+ "avg_response_len": 338.59355021558133,
757
+ "resp_tokens_sum": 145752334,
758
+ "pass_at_k": {
759
+ "pass@1": 0.3475157039845376,
760
+ "pass@2": 0.4305480802550229,
761
+ "pass@4": 0.5078043615701046,
762
+ "pass@8": 0.5755099600110637,
763
+ "pass@16": 0.6322037027587826,
764
+ "pass@32": 0.6792683548513664,
765
+ "pass@64": 0.7191436644363755,
766
+ "pass@128": 0.752304490038656
767
+ },
768
+ "per_op_pass_at_k": {
769
+ "11": {
770
+ "pass@1": 0.6834525602409639,
771
+ "pass@2": 0.7452788516269804,
772
+ "pass@4": 0.7915516082891745,
773
+ "pass@8": 0.8261539846629206,
774
+ "pass@16": 0.8526640250865961,
775
+ "pass@32": 0.8738063302776092,
776
+ "pass@64": 0.8889842536640941,
777
+ "pass@128": 0.9006024096385542
778
+ },
779
+ "12": {
780
+ "pass@1": 0.5999706112852664,
781
+ "pass@2": 0.6710307493890849,
782
+ "pass@4": 0.7270021304155156,
783
+ "pass@8": 0.7740386863781831,
784
+ "pass@16": 0.8163447617758682,
785
+ "pass@32": 0.8558721662357881,
786
+ "pass@64": 0.8879737897637945,
787
+ "pass@128": 0.9059561128526645
788
+ },
789
+ "13": {
790
+ "pass@1": 0.5011160714285714,
791
+ "pass@2": 0.5872255398222277,
792
+ "pass@4": 0.6574135097608591,
793
+ "pass@8": 0.7164654558518686,
794
+ "pass@16": 0.7646393709776641,
795
+ "pass@32": 0.8049812812395852,
796
+ "pass@64": 0.8421836892633574,
797
+ "pass@128": 0.8739495798319328
798
+ },
799
+ "14": {
800
+ "pass@1": 0.46562043128654973,
801
+ "pass@2": 0.5736181620389555,
802
+ "pass@4": 0.6658962851683015,
803
+ "pass@8": 0.7406709774148201,
804
+ "pass@16": 0.7999165548996432,
805
+ "pass@32": 0.8459800566229854,
806
+ "pass@64": 0.8804301429076689,
807
+ "pass@128": 0.9035087719298246
808
+ },
809
+ "15": {
810
+ "pass@1": 0.34005681818181815,
811
+ "pass@2": 0.4530862264376045,
812
+ "pass@4": 0.5587200136914704,
813
+ "pass@8": 0.6492672106221214,
814
+ "pass@16": 0.7203189663183454,
815
+ "pass@32": 0.772011087327161,
816
+ "pass@64": 0.8109608705247362,
817
+ "pass@128": 0.8454545454545455
818
+ },
819
+ "16": {
820
+ "pass@1": 0.2512686965811966,
821
+ "pass@2": 0.3525497313637072,
822
+ "pass@4": 0.4530276642770078,
823
+ "pass@8": 0.5447819490663908,
824
+ "pass@16": 0.6222232974153704,
825
+ "pass@32": 0.6833052014073985,
826
+ "pass@64": 0.7317438177167213,
827
+ "pass@128": 0.7663817663817664
828
+ },
829
+ "17": {
830
+ "pass@1": 0.18800951086956522,
831
+ "pass@2": 0.2707564068078446,
832
+ "pass@4": 0.3571127275627813,
833
+ "pass@8": 0.4376570004492196,
834
+ "pass@16": 0.5083689377646803,
835
+ "pass@32": 0.5709991978812625,
836
+ "pass@64": 0.6318509335752782,
837
+ "pass@128": 0.6956521739130435
838
+ },
839
+ "18": {
840
+ "pass@1": 0.16176470588235295,
841
+ "pass@2": 0.23503285664659573,
842
+ "pass@4": 0.31124803204746476,
843
+ "pass@8": 0.3828673713583056,
844
+ "pass@16": 0.44741521106430404,
845
+ "pass@32": 0.5054273561599129,
846
+ "pass@64": 0.5568983926229338,
847
+ "pass@128": 0.6029411764705882
848
+ },
849
+ "19": {
850
+ "pass@1": 0.13955543154761904,
851
+ "pass@2": 0.2029613241704161,
852
+ "pass@4": 0.26906566143517774,
853
+ "pass@8": 0.3307801388013955,
854
+ "pass@16": 0.3853507519948162,
855
+ "pass@32": 0.4335818215143217,
856
+ "pass@64": 0.4757266113676575,
857
+ "pass@128": 0.5089285714285714
858
+ },
859
+ "20": {
860
+ "pass@1": 0.14794629491017963,
861
+ "pass@2": 0.21578361485690029,
862
+ "pass@4": 0.2874044784446853,
863
+ "pass@8": 0.3515917933178076,
864
+ "pass@16": 0.4031600034659726,
865
+ "pass@32": 0.44477466635709534,
866
+ "pass@64": 0.4826415676427111,
867
+ "pass@128": 0.5179640718562875
868
+ }
869
+ }
870
+ },
871
+ "movie_festival_awards": {
872
+ "count": 418816,
873
+ "correct": 145161,
874
+ "answer_accuracy": 0.34659850626528116,
875
+ "avg_response_len": 362.29090817924816,
876
+ "resp_tokens_sum": 151733229,
877
+ "pass_at_k": {
878
+ "pass@1": 0.34659850626528116,
879
+ "pass@2": 0.43113398490894067,
880
+ "pass@4": 0.5084318326155154,
881
+ "pass@8": 0.574579983575393,
882
+ "pass@16": 0.6291480125067703,
883
+ "pass@32": 0.6732252696351326,
884
+ "pass@64": 0.709257499797369,
885
+ "pass@128": 0.741442542787286
886
+ },
887
+ "per_op_pass_at_k": {
888
+ "11": {
889
+ "pass@1": 0.6475694444444444,
890
+ "pass@2": 0.720068975753031,
891
+ "pass@4": 0.7730891477850979,
892
+ "pass@8": 0.8109217151768221,
893
+ "pass@16": 0.8391620142013038,
894
+ "pass@32": 0.8621038165828088,
895
+ "pass@64": 0.8830074431961632,
896
+ "pass@128": 0.9015873015873016
897
+ },
898
+ "12": {
899
+ "pass@1": 0.6413417431192661,
900
+ "pass@2": 0.704082538707891,
901
+ "pass@4": 0.7513502354636861,
902
+ "pass@8": 0.7928136020002462,
903
+ "pass@16": 0.8301661693801712,
904
+ "pass@32": 0.8624397252433138,
905
+ "pass@64": 0.8888217834671825,
906
+ "pass@128": 0.9113149847094801
907
+ },
908
+ "13": {
909
+ "pass@1": 0.5245225694444444,
910
+ "pass@2": 0.613710341328862,
911
+ "pass@4": 0.6886929619908626,
912
+ "pass@8": 0.7532199740619856,
913
+ "pass@16": 0.8060972428317466,
914
+ "pass@32": 0.8465569697048755,
915
+ "pass@64": 0.8768956451569085,
916
+ "pass@128": 0.9012345679012346
917
+ },
918
+ "14": {
919
+ "pass@1": 0.43209847922848665,
920
+ "pass@2": 0.5442581164279539,
921
+ "pass@4": 0.6390545526208331,
922
+ "pass@8": 0.7148531736292517,
923
+ "pass@16": 0.7709069487836203,
924
+ "pass@32": 0.8091193075610889,
925
+ "pass@64": 0.8354907283201769,
926
+ "pass@128": 0.8545994065281899
927
+ },
928
+ "15": {
929
+ "pass@1": 0.3362750771604938,
930
+ "pass@2": 0.4498217562214446,
931
+ "pass@4": 0.5492879258148281,
932
+ "pass@8": 0.6259151063045518,
933
+ "pass@16": 0.6838288104673347,
934
+ "pass@32": 0.7271659788130399,
935
+ "pass@64": 0.7591946543896418,
936
+ "pass@128": 0.7901234567901234
937
+ },
938
+ "16": {
939
+ "pass@1": 0.24073561946902655,
940
+ "pass@2": 0.345844075906441,
941
+ "pass@4": 0.4501511309980055,
942
+ "pass@8": 0.5388695636519144,
943
+ "pass@16": 0.61091142935178,
944
+ "pass@32": 0.6711357145149186,
945
+ "pass@64": 0.7212133899833834,
946
+ "pass@128": 0.7640117994100295
947
+ },
948
+ "17": {
949
+ "pass@1": 0.1878059111445783,
950
+ "pass@2": 0.264661870908832,
951
+ "pass@4": 0.34510003455893307,
952
+ "pass@8": 0.4229141069156303,
953
+ "pass@16": 0.49251404650709835,
954
+ "pass@32": 0.5519003412110756,
955
+ "pass@64": 0.6045042867269972,
956
+ "pass@128": 0.6536144578313253
957
+ },
958
+ "18": {
959
+ "pass@1": 0.16849798387096773,
960
+ "pass@2": 0.23845488315976635,
961
+ "pass@4": 0.3089043291967536,
962
+ "pass@8": 0.3740905690365366,
963
+ "pass@16": 0.4314242151887938,
964
+ "pass@32": 0.4816503235844736,
965
+ "pass@64": 0.5280348641946153,
966
+ "pass@128": 0.5741935483870968
967
+ },
968
+ "19": {
969
+ "pass@1": 0.15211952416918428,
970
+ "pass@2": 0.22327800509075332,
971
+ "pass@4": 0.2960748354340905,
972
+ "pass@8": 0.3635955512217954,
973
+ "pass@16": 0.4228344210461176,
974
+ "pass@32": 0.4702218209493226,
975
+ "pass@64": 0.5066129236272973,
976
+ "pass@128": 0.5347432024169184
977
+ },
978
+ "20": {
979
+ "pass@1": 0.14806212462462462,
980
+ "pass@2": 0.2182563961599395,
981
+ "pass@4": 0.29120600127686735,
982
+ "pass@8": 0.3549176070341172,
983
+ "pass@16": 0.40818890086199433,
984
+ "pass@32": 0.45330992297920386,
985
+ "pass@64": 0.4915266997620285,
986
+ "pass@128": 0.5315315315315315
987
+ }
988
+ }
989
+ }
990
+ }
991
+ },
992
+ "total": {
993
+ "answer_accuracy": 0.5415259046052632,
994
+ "avg_loss": 0.0,
995
+ "count": 2432000,
996
+ "validation_metrics": {
997
+ "id": {},
998
+ "ood": {}
999
+ },
1000
+ "pass_at_k": {
1001
+ "pass@1": 0.5415259046052632,
1002
+ "pass@2": 0.6040844643596877,
1003
+ "pass@4": 0.6584589354883053,
1004
+ "pass@8": 0.70419816463071,
1005
+ "pass@16": 0.7415671940892304,
1006
+ "pass@32": 0.7721233750909188,
1007
+ "pass@64": 0.7978578490211599,
1008
+ "pass@128": 0.8203157894736842
1009
+ },
1010
+ "per_op_pass_at_k": {
1011
+ "10": {
1012
+ "pass@1": 0.6998359375,
1013
+ "pass@2": 0.7497196112204725,
1014
+ "pass@4": 0.7877674823772035,
1015
+ "pass@8": 0.8178490135167437,
1016
+ "pass@16": 0.8425426476657046,
1017
+ "pass@32": 0.8647230780096713,
1018
+ "pass@64": 0.8846169498619155,
1019
+ "pass@128": 0.902
1020
+ },
1021
+ "2": {
1022
+ "pass@1": 0.934484375,
1023
+ "pass@2": 0.9583288631889764,
1024
+ "pass@4": 0.9730320496812903,
1025
+ "pass@8": 0.9827350742606419,
1026
+ "pass@16": 0.9888035787644653,
1027
+ "pass@32": 0.9924179792156923,
1028
+ "pass@64": 0.9947161450381122,
1029
+ "pass@128": 0.996
1030
+ },
1031
+ "3": {
1032
+ "pass@1": 0.9164453125,
1033
+ "pass@2": 0.9411937746062994,
1034
+ "pass@4": 0.9579434554743155,
1035
+ "pass@8": 0.9699985325178049,
1036
+ "pass@16": 0.9781683206395383,
1037
+ "pass@32": 0.9838437894282911,
1038
+ "pass@64": 0.9884535535906521,
1039
+ "pass@128": 0.992
1040
+ },
1041
+ "4": {
1042
+ "pass@1": 0.8259296875,
1043
+ "pass@2": 0.8585537647637796,
1044
+ "pass@4": 0.8806184625046869,
1045
+ "pass@8": 0.8961674154317582,
1046
+ "pass@16": 0.9082188111888009,
1047
+ "pass@32": 0.9181050438511107,
1048
+ "pass@64": 0.9259375783769267,
1049
+ "pass@128": 0.933
1050
+ },
1051
+ "5": {
1052
+ "pass@1": 0.698609375,
1053
+ "pass@2": 0.7370525344488192,
1054
+ "pass@4": 0.7667280059992501,
1055
+ "pass@8": 0.7889732628668457,
1056
+ "pass@16": 0.8060324371228792,
1057
+ "pass@32": 0.8203173429933404,
1058
+ "pass@64": 0.8325269722338159,
1059
+ "pass@128": 0.843
1060
+ },
1061
+ "6": {
1062
+ "pass@1": 0.6761796875,
1063
+ "pass@2": 0.7140426919291342,
1064
+ "pass@4": 0.7424857538432698,
1065
+ "pass@8": 0.7651357504137184,
1066
+ "pass@16": 0.7835360752677899,
1067
+ "pass@32": 0.7994032050618843,
1068
+ "pass@64": 0.8150506157082538,
1069
+ "pass@128": 0.832
1070
+ },
1071
+ "7": {
1072
+ "pass@1": 0.6889375,
1073
+ "pass@2": 0.7267791584645674,
1074
+ "pass@4": 0.7546797299400082,
1075
+ "pass@8": 0.7755084684065878,
1076
+ "pass@16": 0.7920948216116958,
1077
+ "pass@32": 0.8059412816926678,
1078
+ "pass@64": 0.8184436972940266,
1079
+ "pass@128": 0.83
1080
+ },
1081
+ "8": {
1082
+ "pass@1": 0.68353125,
1083
+ "pass@2": 0.7357874015748035,
1084
+ "pass@4": 0.7764557210348707,
1085
+ "pass@8": 0.8079708875341801,
1086
+ "pass@16": 0.8317391404315082,
1087
+ "pass@32": 0.8509228087163498,
1088
+ "pass@64": 0.8678658647215115,
1089
+ "pass@128": 0.883
1090
+ },
1091
+ "9": {
1092
+ "pass@1": 0.68421875,
1093
+ "pass@2": 0.7417383120078741,
1094
+ "pass@4": 0.7843575126546682,
1095
+ "pass@8": 0.8158893939898693,
1096
+ "pass@16": 0.8382860648956109,
1097
+ "pass@32": 0.8550870939059481,
1098
+ "pass@64": 0.870409897855889,
1099
+ "pass@128": 0.886
1100
+ },
1101
+ "11": {
1102
+ "pass@1": 0.645328125,
1103
+ "pass@2": 0.7102300688976382,
1104
+ "pass@4": 0.7585581894450686,
1105
+ "pass@8": 0.7949712730733313,
1106
+ "pass@16": 0.8232239283653476,
1107
+ "pass@32": 0.8467896281680857,
1108
+ "pass@64": 0.8677760843111625,
1109
+ "pass@128": 0.887
1110
+ },
1111
+ "12": {
1112
+ "pass@1": 0.6262265625,
1113
+ "pass@2": 0.6924527559055124,
1114
+ "pass@4": 0.7439660961754783,
1115
+ "pass@8": 0.7881987997956229,
1116
+ "pass@16": 0.8274650337042867,
1117
+ "pass@32": 0.8621452294532637,
1118
+ "pass@64": 0.8899411679493481,
1119
+ "pass@128": 0.91
1120
+ },
1121
+ "13": {
1122
+ "pass@1": 0.515390625,
1123
+ "pass@2": 0.6064320866141737,
1124
+ "pass@4": 0.6819955449006371,
1125
+ "pass@8": 0.7454696905246248,
1126
+ "pass@16": 0.7970554598668151,
1127
+ "pass@32": 0.8370062209346598,
1128
+ "pass@64": 0.8680208840016798,
1129
+ "pass@128": 0.894
1130
+ },
1131
+ "14": {
1132
+ "pass@1": 0.4509453125,
1133
+ "pass@2": 0.5587468011811019,
1134
+ "pass@4": 0.6510836965691776,
1135
+ "pass@8": 0.7264001469285997,
1136
+ "pass@16": 0.7848350439768886,
1137
+ "pass@32": 0.8277722436665917,
1138
+ "pass@64": 0.8584198739625949,
1139
+ "pass@128": 0.88
1140
+ },
1141
+ "15": {
1142
+ "pass@1": 0.342046875,
1143
+ "pass@2": 0.45342249015748054,
1144
+ "pass@4": 0.5530513783277083,
1145
+ "pass@8": 0.634712334515626,
1146
+ "pass@16": 0.699178504064928,
1147
+ "pass@32": 0.7476314759907243,
1148
+ "pass@64": 0.7840614124927014,
1149
+ "pass@128": 0.818
1150
+ },
1151
+ "16": {
1152
+ "pass@1": 0.2568984375,
1153
+ "pass@2": 0.3605068897637793,
1154
+ "pass@4": 0.4633091179227593,
1155
+ "pass@8": 0.5545054725122883,
1156
+ "pass@16": 0.6304599364492406,
1157
+ "pass@32": 0.6926131635972946,
1158
+ "pass@64": 0.7438924411504231,
1159
+ "pass@128": 0.786
1160
+ },
1161
+ "17": {
1162
+ "pass@1": 0.180546875,
1163
+ "pass@2": 0.26008489173228355,
1164
+ "pass@4": 0.34560696259842494,
1165
+ "pass@8": 0.4284562778464315,
1166
+ "pass@16": 0.5018989467657816,
1167
+ "pass@32": 0.5645239509858958,
1168
+ "pass@64": 0.6209294612489322,
1169
+ "pass@128": 0.674
1170
+ },
1171
+ "18": {
1172
+ "pass@1": 0.162046875,
1173
+ "pass@2": 0.2337256397637794,
1174
+ "pass@4": 0.3077259774090736,
1175
+ "pass@8": 0.3768791242808825,
1176
+ "pass@16": 0.4392902527391763,
1177
+ "pass@32": 0.4956244462569732,
1178
+ "pass@64": 0.5464061221745686,
1179
+ "pass@128": 0.591
1180
+ },
1181
+ "19": {
1182
+ "pass@1": 0.1514296875,
1183
+ "pass@2": 0.2191755659448819,
1184
+ "pass@4": 0.2889380630858642,
1185
+ "pass@8": 0.3527730781336148,
1186
+ "pass@16": 0.40723262416432515,
1187
+ "pass@32": 0.45234964946368117,
1188
+ "pass@64": 0.4902811911999381,
1189
+ "pass@128": 0.522
1190
+ },
1191
+ "20": {
1192
+ "pass@1": 0.1499609375,
1193
+ "pass@2": 0.21963152066929123,
1194
+ "pass@4": 0.29241657433445783,
1195
+ "pass@8": 0.35717113143418483,
1196
+ "pass@16": 0.4097150600102705,
1197
+ "pass@32": 0.4531264953349681,
1198
+ "pass@64": 0.49154921822946673,
1199
+ "pass@128": 0.527
1200
+ }
1201
+ },
1202
+ "per_op_accuracy": {
1203
+ "10": 0.6998359375,
1204
+ "2": 0.934484375,
1205
+ "3": 0.9164453125,
1206
+ "4": 0.8259296875,
1207
+ "5": 0.698609375,
1208
+ "6": 0.6761796875,
1209
+ "7": 0.6889375,
1210
+ "8": 0.68353125,
1211
+ "9": 0.68421875,
1212
+ "11": 0.645328125,
1213
+ "12": 0.6262265625,
1214
+ "13": 0.515390625,
1215
+ "14": 0.4509453125,
1216
+ "15": 0.342046875,
1217
+ "16": 0.2568984375,
1218
+ "17": 0.180546875,
1219
+ "18": 0.162046875,
1220
+ "19": 0.1514296875,
1221
+ "20": 0.1499609375
1222
+ },
1223
+ "per_op_avg_response_len": {
1224
+ "10": 279.936125,
1225
+ "2": 140.577265625,
1226
+ "3": 158.07409375,
1227
+ "4": 166.3575625,
1228
+ "5": 183.2572890625,
1229
+ "6": 208.2703046875,
1230
+ "7": 227.0995625,
1231
+ "8": 247.07909375,
1232
+ "9": 261.46315625,
1233
+ "11": 296.783828125,
1234
+ "12": 305.03403125,
1235
+ "13": 322.5857421875,
1236
+ "14": 343.72453125,
1237
+ "15": 352.05109375,
1238
+ "16": 356.360015625,
1239
+ "17": 360.660140625,
1240
+ "18": 363.015453125,
1241
+ "19": 357.965421875,
1242
+ "20": 364.9763359375
1243
+ },
1244
+ "per_op_avg_loss": {},
1245
+ "per_op_length": {
1246
+ "10": 279.936125,
1247
+ "2": 140.577265625,
1248
+ "3": 158.07409375,
1249
+ "4": 166.3575625,
1250
+ "5": 183.2572890625,
1251
+ "6": 208.2703046875,
1252
+ "7": 227.0995625,
1253
+ "8": 247.07909375,
1254
+ "9": 261.46315625,
1255
+ "11": 296.783828125,
1256
+ "12": 305.03403125,
1257
+ "13": 322.5857421875,
1258
+ "14": 343.72453125,
1259
+ "15": 352.05109375,
1260
+ "16": 356.360015625,
1261
+ "17": 360.660140625,
1262
+ "18": 363.015453125,
1263
+ "19": 357.965421875,
1264
+ "20": 364.9763359375
1265
+ },
1266
+ "per_op_loss": {},
1267
+ "per_template": {
1268
+ "crazy_zootopia": {
1269
+ "count": 809728,
1270
+ "correct": 438889,
1271
+ "answer_accuracy": 0.5420202833544103,
1272
+ "avg_response_len": 266.6312070226051,
1273
+ "resp_tokens_sum": 215898754,
1274
+ "pass_at_k": {
1275
+ "pass@1": 0.5420202833544103,
1276
+ "pass@2": 0.6053215925837938,
1277
+ "pass@4": 0.6601015812845192,
1278
+ "pass@8": 0.7063449522824156,
1279
+ "pass@16": 0.7442708953088986,
1280
+ "pass@32": 0.7752375894022824,
1281
+ "pass@64": 0.8011639131735242,
1282
+ "pass@128": 0.8239013594688587
1283
+ },
1284
+ "per_op_pass_at_k": {
1285
+ "10": {
1286
+ "pass@1": 0.7018805309734514,
1287
+ "pass@2": 0.7549985628179221,
1288
+ "pass@4": 0.794857199542535,
1289
+ "pass@8": 0.8246423155165492,
1290
+ "pass@16": 0.8477130887947533,
1291
+ "pass@32": 0.8687864891718109,
1292
+ "pass@64": 0.8890464019643919,
1293
+ "pass@128": 0.9085545722713865
1294
+ },
1295
+ "2": {
1296
+ "pass@1": 0.9495716510903427,
1297
+ "pass@2": 0.9728514331444554,
1298
+ "pass@4": 0.9851647276567062,
1299
+ "pass@8": 0.992077485746166,
1300
+ "pass@16": 0.9954145881797847,
1301
+ "pass@32": 0.996585667661237,
1302
+ "pass@64": 0.9968750895206262,
1303
+ "pass@128": 0.9968847352024922
1304
+ },
1305
+ "3": {
1306
+ "pass@1": 0.9074445718654435,
1307
+ "pass@2": 0.9352523236774299,
1308
+ "pass@4": 0.953601489034054,
1309
+ "pass@8": 0.9679758338381604,
1310
+ "pass@16": 0.9790672679207248,
1311
+ "pass@32": 0.9871383226541847,
1312
+ "pass@64": 0.9930782875203557,
1313
+ "pass@128": 0.9969418960244648
1314
+ },
1315
+ "4": {
1316
+ "pass@1": 0.8296630859375,
1317
+ "pass@2": 0.8626487912155513,
1318
+ "pass@4": 0.8834582688882635,
1319
+ "pass@8": 0.895808641913504,
1320
+ "pass@16": 0.9041949013259944,
1321
+ "pass@32": 0.9123519155789825,
1322
+ "pass@64": 0.9217296599331064,
1323
+ "pass@128": 0.93125
1324
+ },
1325
+ "5": {
1326
+ "pass@1": 0.6723710317460317,
1327
+ "pass@2": 0.7115751156105488,
1328
+ "pass@4": 0.7433334842073313,
1329
+ "pass@8": 0.768306546362202,
1330
+ "pass@16": 0.7871896199409257,
1331
+ "pass@32": 0.8020106421329306,
1332
+ "pass@64": 0.8145428062453238,
1333
+ "pass@128": 0.8285714285714286
1334
+ },
1335
+ "6": {
1336
+ "pass@1": 0.6831259356287425,
1337
+ "pass@2": 0.7225941817153095,
1338
+ "pass@4": 0.7511133829079745,
1339
+ "pass@8": 0.7732136904251459,
1340
+ "pass@16": 0.7908965515084061,
1341
+ "pass@32": 0.8059498542595275,
1342
+ "pass@64": 0.8190138601274597,
1343
+ "pass@128": 0.8323353293413174
1344
+ },
1345
+ "7": {
1346
+ "pass@1": 0.7232572115384616,
1347
+ "pass@2": 0.7591458265386947,
1348
+ "pass@4": 0.7840269109297723,
1349
+ "pass@8": 0.8025467115547503,
1350
+ "pass@16": 0.8182715609954336,
1351
+ "pass@32": 0.8323267215314273,
1352
+ "pass@64": 0.8455724214427237,
1353
+ "pass@128": 0.8579881656804734
1354
+ },
1355
+ "8": {
1356
+ "pass@1": 0.6790364583333334,
1357
+ "pass@2": 0.7368809780688469,
1358
+ "pass@4": 0.7800951419534093,
1359
+ "pass@8": 0.8111498024624929,
1360
+ "pass@16": 0.8340518273159697,
1361
+ "pass@32": 0.851499152999998,
1362
+ "pass@64": 0.8652018097269851,
1363
+ "pass@128": 0.8782051282051282
1364
+ },
1365
+ "9": {
1366
+ "pass@1": 0.7028389084507042,
1367
+ "pass@2": 0.7644043889320176,
1368
+ "pass@4": 0.808012754743685,
1369
+ "pass@8": 0.8407312628937469,
1370
+ "pass@16": 0.8654983711299619,
1371
+ "pass@32": 0.8838276379384431,
1372
+ "pass@64": 0.8988843924487747,
1373
+ "pass@128": 0.9126760563380282
1374
+ },
1375
+ "11": {
1376
+ "pass@1": 0.607471671388102,
1377
+ "pass@2": 0.6684865801565879,
1378
+ "pass@4": 0.7145608326933632,
1379
+ "pass@8": 0.751410226301821,
1380
+ "pass@16": 0.7813128542866512,
1381
+ "pass@32": 0.8077145164088809,
1382
+ "pass@64": 0.8342378682376506,
1383
+ "pass@128": 0.8611898016997167
1384
+ },
1385
+ "12": {
1386
+ "pass@1": 0.6359242584745762,
1387
+ "pass@2": 0.7010140019573821,
1388
+ "pass@4": 0.7524318914584823,
1389
+ "pass@8": 0.7966960762341885,
1390
+ "pass@16": 0.8349907268657351,
1391
+ "pass@32": 0.8675260403092755,
1392
+ "pass@64": 0.8927480390421706,
1393
+ "pass@128": 0.9124293785310734
1394
+ },
1395
+ "13": {
1396
+ "pass@1": 0.5220905172413793,
1397
+ "pass@2": 0.620534226667983,
1398
+ "pass@4": 0.7027034552695014,
1399
+ "pass@8": 0.770057213759952,
1400
+ "pass@16": 0.8241495227282288,
1401
+ "pass@32": 0.8631455967014058,
1402
+ "pass@64": 0.8879219996985025,
1403
+ "pass@128": 0.9090909090909091
1404
+ },
1405
+ "14": {
1406
+ "pass@1": 0.4550963785046729,
1407
+ "pass@2": 0.5581134097431748,
1408
+ "pass@4": 0.647930787565105,
1409
+ "pass@8": 0.7233182340799785,
1410
+ "pass@16": 0.7833892226203504,
1411
+ "pass@32": 0.8279559428456201,
1412
+ "pass@64": 0.8590417122874536,
1413
+ "pass@128": 0.881619937694704
1414
+ },
1415
+ "15": {
1416
+ "pass@1": 0.34934971098265893,
1417
+ "pass@2": 0.45711498964544184,
1418
+ "pass@4": 0.5511690342356048,
1419
+ "pass@8": 0.6290683831435004,
1420
+ "pass@16": 0.6933893369608615,
1421
+ "pass@32": 0.7435434683159998,
1422
+ "pass@64": 0.7816914947898684,
1423
+ "pass@128": 0.8179190751445087
1424
+ },
1425
+ "16": {
1426
+ "pass@1": 0.2809475806451613,
1427
+ "pass@2": 0.38555094297688597,
1428
+ "pass@4": 0.4893392721071152,
1429
+ "pass@8": 0.5826136332644718,
1430
+ "pass@16": 0.6611631758264305,
1431
+ "pass@32": 0.7266388086540002,
1432
+ "pass@64": 0.7824484578306036,
1433
+ "pass@128": 0.832258064516129
1434
+ },
1435
+ "17": {
1436
+ "pass@1": 0.1666365606936416,
1437
+ "pass@2": 0.2457618138682809,
1438
+ "pass@4": 0.33538570187758293,
1439
+ "pass@8": 0.42521167689535794,
1440
+ "pass@16": 0.5048829056797672,
1441
+ "pass@32": 0.5706106820406163,
1442
+ "pass@64": 0.6265261197812971,
1443
+ "pass@128": 0.6734104046242775
1444
+ },
1445
+ "18": {
1446
+ "pass@1": 0.15660714285714286,
1447
+ "pass@2": 0.2282670134983126,
1448
+ "pass@4": 0.3032608698912637,
1449
+ "pass@8": 0.37353183319352135,
1450
+ "pass@16": 0.43836449791082005,
1451
+ "pass@32": 0.4984786995754745,
1452
+ "pass@64": 0.5524853165212575,
1453
+ "pass@128": 0.5942857142857143
1454
+ },
1455
+ "19": {
1456
+ "pass@1": 0.16272522522522523,
1457
+ "pass@2": 0.23145808209784588,
1458
+ "pass@4": 0.30189558653141335,
1459
+ "pass@8": 0.36420667880460095,
1460
+ "pass@16": 0.4138035379213277,
1461
+ "pass@32": 0.4535217258277585,
1462
+ "pass@64": 0.48873324942873747,
1463
+ "pass@128": 0.5225225225225225
1464
+ },
1465
+ "20": {
1466
+ "pass@1": 0.15388044294294295,
1467
+ "pass@2": 0.2248661062637441,
1468
+ "pass@4": 0.29865429462083,
1469
+ "pass@8": 0.3650207487197481,
1470
+ "pass@16": 0.41781596055733294,
1471
+ "pass@32": 0.4613199772361069,
1472
+ "pass@64": 0.5005061369851226,
1473
+ "pass@128": 0.5315315315315315
1474
+ }
1475
+ }
1476
+ },
1477
+ "teachers_in_school": {
1478
+ "count": 821120,
1479
+ "correct": 445920,
1480
+ "answer_accuracy": 0.5430631332813718,
1481
+ "avg_response_len": 274.84728297934527,
1482
+ "resp_tokens_sum": 225682601,
1483
+ "pass_at_k": {
1484
+ "pass@1": 0.5430631332813718,
1485
+ "pass@2": 0.6044259310118462,
1486
+ "pass@4": 0.6581722536728921,
1487
+ "pass@8": 0.7037834676975927,
1488
+ "pass@16": 0.7411931274105865,
1489
+ "pass@32": 0.771978422134926,
1490
+ "pass@64": 0.7981625962635988,
1491
+ "pass@128": 0.8202650038971161
1492
+ },
1493
+ "per_op_pass_at_k": {
1494
+ "10": {
1495
+ "pass@1": 0.7098590874233128,
1496
+ "pass@2": 0.7535682907830544,
1497
+ "pass@4": 0.7864975527407232,
1498
+ "pass@8": 0.8135336785898025,
1499
+ "pass@16": 0.8366857190055628,
1500
+ "pass@32": 0.8580710264821365,
1501
+ "pass@64": 0.8776973706717127,
1502
+ "pass@128": 0.8957055214723927
1503
+ },
1504
+ "2": {
1505
+ "pass@1": 0.9405395136778115,
1506
+ "pass@2": 0.9620412308833738,
1507
+ "pass@4": 0.9758225314920738,
1508
+ "pass@8": 0.9849524204535265,
1509
+ "pass@16": 0.9900649654565532,
1510
+ "pass@32": 0.9921000445202469,
1511
+ "pass@64": 0.9931668037952412,
1512
+ "pass@128": 0.993920972644377
1513
+ },
1514
+ "3": {
1515
+ "pass@1": 0.9234375,
1516
+ "pass@2": 0.9456786119064663,
1517
+ "pass@4": 0.9617405105611796,
1518
+ "pass@8": 0.9726108401727136,
1519
+ "pass@16": 0.9779520922364761,
1520
+ "pass@32": 0.9803488524936297,
1521
+ "pass@64": 0.9825398325449198,
1522
+ "pass@128": 0.9848484848484849
1523
+ },
1524
+ "4": {
1525
+ "pass@1": 0.8384106878698225,
1526
+ "pass@2": 0.8696401219540603,
1527
+ "pass@4": 0.8919865230418679,
1528
+ "pass@8": 0.9109039824674823,
1529
+ "pass@16": 0.9268934190930932,
1530
+ "pass@32": 0.939480395901062,
1531
+ "pass@64": 0.9475861428402855,
1532
+ "pass@128": 0.9526627218934911
1533
+ },
1534
+ "5": {
1535
+ "pass@1": 0.7247509057971014,
1536
+ "pass@2": 0.7569896154285061,
1537
+ "pass@4": 0.7794320843046791,
1538
+ "pass@8": 0.7960690607966524,
1539
+ "pass@16": 0.8107549025408636,
1540
+ "pass@32": 0.8248985765960865,
1541
+ "pass@64": 0.8371374039094492,
1542
+ "pass@128": 0.8463768115942029
1543
+ },
1544
+ "6": {
1545
+ "pass@1": 0.6351181948424068,
1546
+ "pass@2": 0.6731091927441734,
1547
+ "pass@4": 0.7035691777782792,
1548
+ "pass@8": 0.7284657829447492,
1549
+ "pass@16": 0.7492790741526567,
1550
+ "pass@32": 0.7669448136688961,
1551
+ "pass@64": 0.7842548351421045,
1552
+ "pass@128": 0.8022922636103151
1553
+ },
1554
+ "7": {
1555
+ "pass@1": 0.6782649927745664,
1556
+ "pass@2": 0.7190959816348825,
1557
+ "pass@4": 0.7507817206555822,
1558
+ "pass@8": 0.7737670955280236,
1559
+ "pass@16": 0.7898595118415901,
1560
+ "pass@32": 0.8002346239143221,
1561
+ "pass@64": 0.8076964146996783,
1562
+ "pass@128": 0.815028901734104
1563
+ },
1564
+ "8": {
1565
+ "pass@1": 0.6978653169014084,
1566
+ "pass@2": 0.7489800515692578,
1567
+ "pass@4": 0.7882377675501828,
1568
+ "pass@8": 0.8181867244335217,
1569
+ "pass@16": 0.8403183790454402,
1570
+ "pass@32": 0.8589868896024869,
1571
+ "pass@64": 0.876235649537859,
1572
+ "pass@128": 0.8901408450704226
1573
+ },
1574
+ "9": {
1575
+ "pass@1": 0.6945172155688623,
1576
+ "pass@2": 0.7500011050733182,
1577
+ "pass@4": 0.7906428624752743,
1578
+ "pass@8": 0.82085154021908,
1579
+ "pass@16": 0.8423189279171873,
1580
+ "pass@32": 0.8580696456086162,
1581
+ "pass@64": 0.8721121926305166,
1582
+ "pass@128": 0.8862275449101796
1583
+ },
1584
+ "11": {
1585
+ "pass@1": 0.6834525602409639,
1586
+ "pass@2": 0.7452788516269804,
1587
+ "pass@4": 0.7915516082891745,
1588
+ "pass@8": 0.8261539846629206,
1589
+ "pass@16": 0.8526640250865961,
1590
+ "pass@32": 0.8738063302776092,
1591
+ "pass@64": 0.8889842536640941,
1592
+ "pass@128": 0.9006024096385542
1593
+ },
1594
+ "12": {
1595
+ "pass@1": 0.5999706112852664,
1596
+ "pass@2": 0.6710307493890849,
1597
+ "pass@4": 0.7270021304155156,
1598
+ "pass@8": 0.7740386863781831,
1599
+ "pass@16": 0.8163447617758682,
1600
+ "pass@32": 0.8558721662357881,
1601
+ "pass@64": 0.8879737897637945,
1602
+ "pass@128": 0.9059561128526645
1603
+ },
1604
+ "13": {
1605
+ "pass@1": 0.5011160714285714,
1606
+ "pass@2": 0.5872255398222277,
1607
+ "pass@4": 0.6574135097608591,
1608
+ "pass@8": 0.7164654558518686,
1609
+ "pass@16": 0.7646393709776641,
1610
+ "pass@32": 0.8049812812395852,
1611
+ "pass@64": 0.8421836892633574,
1612
+ "pass@128": 0.8739495798319328
1613
+ },
1614
+ "14": {
1615
+ "pass@1": 0.46562043128654973,
1616
+ "pass@2": 0.5736181620389555,
1617
+ "pass@4": 0.6658962851683015,
1618
+ "pass@8": 0.7406709774148201,
1619
+ "pass@16": 0.7999165548996432,
1620
+ "pass@32": 0.8459800566229854,
1621
+ "pass@64": 0.8804301429076689,
1622
+ "pass@128": 0.9035087719298246
1623
+ },
1624
+ "15": {
1625
+ "pass@1": 0.34005681818181815,
1626
+ "pass@2": 0.4530862264376045,
1627
+ "pass@4": 0.5587200136914704,
1628
+ "pass@8": 0.6492672106221214,
1629
+ "pass@16": 0.7203189663183454,
1630
+ "pass@32": 0.772011087327161,
1631
+ "pass@64": 0.8109608705247362,
1632
+ "pass@128": 0.8454545454545455
1633
+ },
1634
+ "16": {
1635
+ "pass@1": 0.2512686965811966,
1636
+ "pass@2": 0.3525497313637072,
1637
+ "pass@4": 0.4530276642770078,
1638
+ "pass@8": 0.5447819490663908,
1639
+ "pass@16": 0.6222232974153704,
1640
+ "pass@32": 0.6833052014073985,
1641
+ "pass@64": 0.7317438177167213,
1642
+ "pass@128": 0.7663817663817664
1643
+ },
1644
+ "17": {
1645
+ "pass@1": 0.18800951086956522,
1646
+ "pass@2": 0.2707564068078446,
1647
+ "pass@4": 0.3571127275627813,
1648
+ "pass@8": 0.4376570004492196,
1649
+ "pass@16": 0.5083689377646803,
1650
+ "pass@32": 0.5709991978812625,
1651
+ "pass@64": 0.6318509335752782,
1652
+ "pass@128": 0.6956521739130435
1653
+ },
1654
+ "18": {
1655
+ "pass@1": 0.16176470588235295,
1656
+ "pass@2": 0.23503285664659573,
1657
+ "pass@4": 0.31124803204746476,
1658
+ "pass@8": 0.3828673713583056,
1659
+ "pass@16": 0.44741521106430404,
1660
+ "pass@32": 0.5054273561599129,
1661
+ "pass@64": 0.5568983926229338,
1662
+ "pass@128": 0.6029411764705882
1663
+ },
1664
+ "19": {
1665
+ "pass@1": 0.13955543154761904,
1666
+ "pass@2": 0.2029613241704161,
1667
+ "pass@4": 0.26906566143517774,
1668
+ "pass@8": 0.3307801388013955,
1669
+ "pass@16": 0.3853507519948162,
1670
+ "pass@32": 0.4335818215143217,
1671
+ "pass@64": 0.4757266113676575,
1672
+ "pass@128": 0.5089285714285714
1673
+ },
1674
+ "20": {
1675
+ "pass@1": 0.14794629491017963,
1676
+ "pass@2": 0.21578361485690029,
1677
+ "pass@4": 0.2874044784446853,
1678
+ "pass@8": 0.3515917933178076,
1679
+ "pass@16": 0.4031600034659726,
1680
+ "pass@32": 0.44477466635709534,
1681
+ "pass@64": 0.4826415676427111,
1682
+ "pass@128": 0.5179640718562875
1683
+ }
1684
+ }
1685
+ },
1686
+ "movie_festival_awards": {
1687
+ "count": 801152,
1688
+ "correct": 432182,
1689
+ "answer_accuracy": 0.5394506910049529,
1690
+ "avg_response_len": 294.8421011243809,
1691
+ "resp_tokens_sum": 236213339,
1692
+ "pass_at_k": {
1693
+ "pass@1": 0.5394506910049529,
1694
+ "pass@2": 0.6024841157866588,
1695
+ "pass@4": 0.6570925329542748,
1696
+ "pass@8": 0.702453429392051,
1697
+ "pass@16": 0.7392179408263716,
1698
+ "pass@32": 0.7691243901857652,
1699
+ "pass@64": 0.7942040520267212,
1700
+ "pass@128": 0.8167438888001278
1701
+ },
1702
+ "per_op_pass_at_k": {
1703
+ "10": {
1704
+ "pass@1": 0.6880130597014925,
1705
+ "pass@2": 0.7406323451639437,
1706
+ "pass@4": 0.7818289239964403,
1707
+ "pass@8": 0.8151739980189773,
1708
+ "pass@16": 0.8430100482640915,
1709
+ "pass@32": 0.867084488200749,
1710
+ "pass@64": 0.8868682888268872,
1711
+ "pass@128": 0.9014925373134328
1712
+ },
1713
+ "2": {
1714
+ "pass@1": 0.9149553571428571,
1715
+ "pass@2": 0.941519966254218,
1716
+ "pass@4": 0.9592815978359845,
1717
+ "pass@8": 0.9720824428768929,
1718
+ "pass@16": 0.9815546352101394,
1719
+ "pass@32": 0.9888944721407834,
1720
+ "pass@64": 0.9941924652953058,
1721
+ "pass@128": 0.9971428571428571
1722
+ },
1723
+ "3": {
1724
+ "pass@1": 0.9182990160349854,
1725
+ "pass@2": 0.9425432152613579,
1726
+ "pass@4": 0.9584297378279611,
1727
+ "pass@8": 0.9694135789962994,
1728
+ "pass@16": 0.9775193399166878,
1729
+ "pass@32": 0.9840654128206252,
1730
+ "pass@64": 0.9897341365354874,
1731
+ "pass@128": 0.9941690962099126
1732
+ },
1733
+ "4": {
1734
+ "pass@1": 0.8101014254385965,
1735
+ "pass@2": 0.8437654688723117,
1736
+ "pass@4": 0.8667262329599149,
1737
+ "pass@8": 0.8819389004252276,
1738
+ "pass@16": 0.8935276816111616,
1739
+ "pass@32": 0.9023627399160156,
1740
+ "pass@64": 0.9084794471295793,
1741
+ "pass@128": 0.9152046783625731
1742
+ },
1743
+ "5": {
1744
+ "pass@1": 0.6963924632352941,
1745
+ "pass@2": 0.7404263403195928,
1746
+ "pass@4": 0.775511439378901,
1747
+ "pass@8": 0.8009202199644325,
1748
+ "pass@16": 0.8186978396026182,
1749
+ "pass@32": 0.8326293582229927,
1750
+ "pass@64": 0.8445105409346427,
1751
+ "pass@128": 0.8529411764705882
1752
+ },
1753
+ "6": {
1754
+ "pass@1": 0.714067429022082,
1755
+ "pass@2": 0.7500981923296656,
1756
+ "pass@4": 0.7762405075942792,
1757
+ "pass@8": 0.7969962762271371,
1758
+ "pass@16": 0.8134959942735189,
1759
+ "pass@32": 0.8282404219834616,
1760
+ "pass@64": 0.8447793342621057,
1761
+ "pass@128": 0.8643533123028391
1762
+ },
1763
+ "7": {
1764
+ "pass@1": 0.6639141613924051,
1765
+ "pass@2": 0.7005717071165156,
1766
+ "pass@4": 0.727557464237065,
1767
+ "pass@8": 0.7484945090138806,
1768
+ "pass@16": 0.7665431737280053,
1769
+ "pass@32": 0.7839673099388944,
1770
+ "pass@64": 0.8011938587351172,
1771
+ "pass@128": 0.8164556962025317
1772
+ },
1773
+ "8": {
1774
+ "pass@1": 0.672461524024024,
1775
+ "pass@2": 0.7206985528835925,
1776
+ "pass@4": 0.7604853731684436,
1777
+ "pass@8": 0.7941016876636111,
1778
+ "pass@16": 0.8204262635068897,
1779
+ "pass@32": 0.8417859674818847,
1780
+ "pass@64": 0.8614391726749312,
1781
+ "pass@128": 0.8798798798798799
1782
+ },
1783
+ "9": {
1784
+ "pass@1": 0.6519041398713826,
1785
+ "pass@2": 0.7069915911335041,
1786
+ "pass@4": 0.7506053654466821,
1787
+ "pass@8": 0.7822037981653261,
1788
+ "pass@16": 0.8028926727335497,
1789
+ "pass@32": 0.8190772373135783,
1790
+ "pass@64": 0.8360786694468839,
1791
+ "pass@128": 0.8553054662379421
1792
+ },
1793
+ "11": {
1794
+ "pass@1": 0.6475694444444444,
1795
+ "pass@2": 0.720068975753031,
1796
+ "pass@4": 0.7730891477850979,
1797
+ "pass@8": 0.8109217151768221,
1798
+ "pass@16": 0.8391620142013038,
1799
+ "pass@32": 0.8621038165828088,
1800
+ "pass@64": 0.8830074431961632,
1801
+ "pass@128": 0.9015873015873016
1802
+ },
1803
+ "12": {
1804
+ "pass@1": 0.6413417431192661,
1805
+ "pass@2": 0.704082538707891,
1806
+ "pass@4": 0.7513502354636861,
1807
+ "pass@8": 0.7928136020002462,
1808
+ "pass@16": 0.8301661693801712,
1809
+ "pass@32": 0.8624397252433138,
1810
+ "pass@64": 0.8888217834671825,
1811
+ "pass@128": 0.9113149847094801
1812
+ },
1813
+ "13": {
1814
+ "pass@1": 0.5245225694444444,
1815
+ "pass@2": 0.613710341328862,
1816
+ "pass@4": 0.6886929619908626,
1817
+ "pass@8": 0.7532199740619856,
1818
+ "pass@16": 0.8060972428317466,
1819
+ "pass@32": 0.8465569697048755,
1820
+ "pass@64": 0.8768956451569085,
1821
+ "pass@128": 0.9012345679012346
1822
+ },
1823
+ "14": {
1824
+ "pass@1": 0.43209847922848665,
1825
+ "pass@2": 0.5442581164279539,
1826
+ "pass@4": 0.6390545526208331,
1827
+ "pass@8": 0.7148531736292517,
1828
+ "pass@16": 0.7709069487836203,
1829
+ "pass@32": 0.8091193075610889,
1830
+ "pass@64": 0.8354907283201769,
1831
+ "pass@128": 0.8545994065281899
1832
+ },
1833
+ "15": {
1834
+ "pass@1": 0.3362750771604938,
1835
+ "pass@2": 0.4498217562214446,
1836
+ "pass@4": 0.5492879258148281,
1837
+ "pass@8": 0.6259151063045518,
1838
+ "pass@16": 0.6838288104673347,
1839
+ "pass@32": 0.7271659788130399,
1840
+ "pass@64": 0.7591946543896418,
1841
+ "pass@128": 0.7901234567901234
1842
+ },
1843
+ "16": {
1844
+ "pass@1": 0.24073561946902655,
1845
+ "pass@2": 0.345844075906441,
1846
+ "pass@4": 0.4501511309980055,
1847
+ "pass@8": 0.5388695636519144,
1848
+ "pass@16": 0.61091142935178,
1849
+ "pass@32": 0.6711357145149186,
1850
+ "pass@64": 0.7212133899833834,
1851
+ "pass@128": 0.7640117994100295
1852
+ },
1853
+ "17": {
1854
+ "pass@1": 0.1878059111445783,
1855
+ "pass@2": 0.264661870908832,
1856
+ "pass@4": 0.34510003455893307,
1857
+ "pass@8": 0.4229141069156303,
1858
+ "pass@16": 0.49251404650709835,
1859
+ "pass@32": 0.5519003412110756,
1860
+ "pass@64": 0.6045042867269972,
1861
+ "pass@128": 0.6536144578313253
1862
+ },
1863
+ "18": {
1864
+ "pass@1": 0.16849798387096773,
1865
+ "pass@2": 0.23845488315976635,
1866
+ "pass@4": 0.3089043291967536,
1867
+ "pass@8": 0.3740905690365366,
1868
+ "pass@16": 0.4314242151887938,
1869
+ "pass@32": 0.4816503235844736,
1870
+ "pass@64": 0.5280348641946153,
1871
+ "pass@128": 0.5741935483870968
1872
+ },
1873
+ "19": {
1874
+ "pass@1": 0.15211952416918428,
1875
+ "pass@2": 0.22327800509075332,
1876
+ "pass@4": 0.2960748354340905,
1877
+ "pass@8": 0.3635955512217954,
1878
+ "pass@16": 0.4228344210461176,
1879
+ "pass@32": 0.4702218209493226,
1880
+ "pass@64": 0.5066129236272973,
1881
+ "pass@128": 0.5347432024169184
1882
+ },
1883
+ "20": {
1884
+ "pass@1": 0.14806212462462462,
1885
+ "pass@2": 0.2182563961599395,
1886
+ "pass@4": 0.29120600127686735,
1887
+ "pass@8": 0.3549176070341172,
1888
+ "pass@16": 0.40818890086199433,
1889
+ "pass@32": 0.45330992297920386,
1890
+ "pass@64": 0.4915266997620285,
1891
+ "pass@128": 0.5315315315315315
1892
+ }
1893
+ }
1894
+ }
1895
+ }
1896
+ }
1897
+ }