ZachW commited on
Commit
ce97c6b
·
verified ·
1 Parent(s): f76a04b

Upload extrapolation artifact: 050302 pretrain checkpoint-18586 metrics

Browse files
050302/evals/pretrain_050302_ckpt18586/pretrain_050302_ckpt18586_metrics.json ADDED
@@ -0,0 +1,1897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "pretrain_050302_ckpt18586",
3
+ "id": {
4
+ "answer_accuracy": 0.6591371527777777,
5
+ "avg_response_len": 202.56687413194445,
6
+ "avg_loss": 0.0,
7
+ "count": 1152000,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {
10
+ "pass@1": 0.6591371527777777,
11
+ "pass@2": 0.707459249234457,
12
+ "pass@4": 0.7458439854289,
13
+ "pass@8": 0.7785815581728577,
14
+ "pass@16": 0.8076446647276339,
15
+ "pass@32": 0.8338631196139237,
16
+ "pass@64": 0.8576606399876234,
17
+ "pass@128": 0.8797777777777778
18
+ },
19
+ "per_op_pass_at_k": {
20
+ "10": {
21
+ "pass@1": 0.5284453125,
22
+ "pass@2": 0.6050858759842528,
23
+ "pass@4": 0.6711347938695169,
24
+ "pass@8": 0.7306809613281867,
25
+ "pass@16": 0.7840277284668682,
26
+ "pass@32": 0.8297723522006395,
27
+ "pass@64": 0.867259596306628,
28
+ "pass@128": 0.897
29
+ },
30
+ "2": {
31
+ "pass@1": 0.966640625,
32
+ "pass@2": 0.9790877214566934,
33
+ "pass@4": 0.9861680459317581,
34
+ "pass@8": 0.9908336649877861,
35
+ "pass@16": 0.9935962945915,
36
+ "pass@32": 0.9948623964784657,
37
+ "pass@64": 0.9957516213419054,
38
+ "pass@128": 0.997
39
+ },
40
+ "3": {
41
+ "pass@1": 0.944015625,
42
+ "pass@2": 0.9593532234251971,
43
+ "pass@4": 0.9692371754780653,
44
+ "pass@8": 0.976086913788811,
45
+ "pass@16": 0.9806792820711105,
46
+ "pass@32": 0.9835232837747843,
47
+ "pass@64": 0.9852268502493713,
48
+ "pass@128": 0.986
49
+ },
50
+ "4": {
51
+ "pass@1": 0.7210234375,
52
+ "pass@2": 0.7622476624015748,
53
+ "pass@4": 0.7931688311773528,
54
+ "pass@8": 0.8180216883664996,
55
+ "pass@16": 0.8392104673213089,
56
+ "pass@32": 0.8579530107784558,
57
+ "pass@64": 0.8754655959158385,
58
+ "pass@128": 0.892
59
+ },
60
+ "5": {
61
+ "pass@1": 0.5931328125,
62
+ "pass@2": 0.6308367372047244,
63
+ "pass@4": 0.6605045965504309,
64
+ "pass@8": 0.6864019188998131,
65
+ "pass@16": 0.7111635789832254,
66
+ "pass@32": 0.735359380629359,
67
+ "pass@64": 0.7584082913454374,
68
+ "pass@128": 0.781
69
+ },
70
+ "6": {
71
+ "pass@1": 0.58290625,
72
+ "pass@2": 0.634886318897638,
73
+ "pass@4": 0.6761626663854513,
74
+ "pass@8": 0.71331148317756,
75
+ "pass@16": 0.7479597662974471,
76
+ "pass@32": 0.7795488087145294,
77
+ "pass@64": 0.8077825276680214,
78
+ "pass@128": 0.832
79
+ },
80
+ "7": {
81
+ "pass@1": 0.548109375,
82
+ "pass@2": 0.599172244094488,
83
+ "pass@4": 0.6401481196100498,
84
+ "pass@8": 0.6750420158660341,
85
+ "pass@16": 0.7067478460514601,
86
+ "pass@32": 0.7380706785193885,
87
+ "pass@64": 0.7693794860547157,
88
+ "pass@128": 0.803
89
+ },
90
+ "8": {
91
+ "pass@1": 0.535453125,
92
+ "pass@2": 0.60276316437008,
93
+ "pass@4": 0.6567131941319833,
94
+ "pass@8": 0.7034993490119114,
95
+ "pass@16": 0.7472323803234155,
96
+ "pass@32": 0.7884353184853741,
97
+ "pass@64": 0.8256183039969759,
98
+ "pass@128": 0.859
99
+ },
100
+ "9": {
101
+ "pass@1": 0.5125078125,
102
+ "pass@2": 0.5937002952755916,
103
+ "pass@4": 0.6593584457255333,
104
+ "pass@8": 0.7133560281291572,
105
+ "pass@16": 0.7581846384423049,
106
+ "pass@32": 0.7972428469443084,
107
+ "pass@64": 0.8340534870096861,
108
+ "pass@128": 0.871
109
+ }
110
+ },
111
+ "per_op_accuracy": {
112
+ "10": 0.5284453125,
113
+ "2": 0.966640625,
114
+ "3": 0.944015625,
115
+ "4": 0.7210234375,
116
+ "5": 0.5931328125,
117
+ "6": 0.58290625,
118
+ "7": 0.548109375,
119
+ "8": 0.535453125,
120
+ "9": 0.5125078125
121
+ },
122
+ "per_op_avg_response_len": {
123
+ "10": 278.1583203125,
124
+ "2": 121.0686328125,
125
+ "3": 147.5285625,
126
+ "4": 161.5481484375,
127
+ "5": 177.994421875,
128
+ "6": 203.069109375,
129
+ "7": 226.3078828125,
130
+ "8": 246.1722421875,
131
+ "9": 261.254546875
132
+ },
133
+ "per_op_avg_loss": {},
134
+ "per_op_length": {
135
+ "10": 278.1583203125,
136
+ "2": 121.0686328125,
137
+ "3": 147.5285625,
138
+ "4": 161.5481484375,
139
+ "5": 177.994421875,
140
+ "6": 203.069109375,
141
+ "7": 226.3078828125,
142
+ "8": 246.1722421875,
143
+ "9": 261.254546875
144
+ },
145
+ "per_op_loss": {},
146
+ "per_template": {
147
+ "crazy_zootopia": {
148
+ "count": 379008,
149
+ "correct": 251959,
150
+ "answer_accuracy": 0.6647854398851739,
151
+ "avg_response_len": 193.0696739910503,
152
+ "resp_tokens_sum": 73174951,
153
+ "pass_at_k": {
154
+ "pass@1": 0.6647854398851739,
155
+ "pass@2": 0.7134728172276359,
156
+ "pass@4": 0.7511316364393166,
157
+ "pass@8": 0.7825762495072922,
158
+ "pass@16": 0.809940422012813,
159
+ "pass@32": 0.8334142775850467,
160
+ "pass@64": 0.8541131730711384,
161
+ "pass@128": 0.8747044917257684
162
+ },
163
+ "per_op_pass_at_k": {
164
+ "10": {
165
+ "pass@1": 0.5485342920353983,
166
+ "pass@2": 0.6212945381274241,
167
+ "pass@4": 0.6798927465372133,
168
+ "pass@8": 0.7329713189633855,
169
+ "pass@16": 0.7809185330850646,
170
+ "pass@32": 0.8214210274584312,
171
+ "pass@64": 0.8545478788781176,
172
+ "pass@128": 0.8820058997050148
173
+ },
174
+ "2": {
175
+ "pass@1": 0.9669976635514018,
176
+ "pass@2": 0.9762200431721738,
177
+ "pass@4": 0.9828518164201436,
178
+ "pass@8": 0.988660768625104,
179
+ "pass@16": 0.9928562993950985,
180
+ "pass@32": 0.9949676562841662,
181
+ "pass@64": 0.9961110638984689,
182
+ "pass@128": 0.9968847352024922
183
+ },
184
+ "3": {
185
+ "pass@1": 0.9523604740061162,
186
+ "pass@2": 0.965793481663416,
187
+ "pass@4": 0.9743357550489675,
188
+ "pass@8": 0.9804021050148144,
189
+ "pass@16": 0.9847239199729213,
190
+ "pass@32": 0.9876441175361615,
191
+ "pass@64": 0.9896933766236177,
192
+ "pass@128": 0.9908256880733946
193
+ },
194
+ "4": {
195
+ "pass@1": 0.7410400390625,
196
+ "pass@2": 0.7776855468750001,
197
+ "pass@4": 0.804752966816648,
198
+ "pass@8": 0.8265181231781614,
199
+ "pass@16": 0.8441561704150973,
200
+ "pass@32": 0.8584107299081515,
201
+ "pass@64": 0.8725424263973889,
202
+ "pass@128": 0.890625
203
+ },
204
+ "5": {
205
+ "pass@1": 0.5860615079365079,
206
+ "pass@2": 0.6280289182602176,
207
+ "pass@4": 0.6568151829235631,
208
+ "pass@8": 0.6784618285484385,
209
+ "pass@16": 0.6985429050145969,
210
+ "pass@32": 0.7172269713114492,
211
+ "pass@64": 0.7336378808087067,
212
+ "pass@128": 0.7523809523809524
213
+ },
214
+ "6": {
215
+ "pass@1": 0.6110357410179641,
216
+ "pass@2": 0.6640256642227358,
217
+ "pass@4": 0.7050190184872099,
218
+ "pass@8": 0.7401048358206398,
219
+ "pass@16": 0.7712992571383563,
220
+ "pass@32": 0.7976646629926885,
221
+ "pass@64": 0.8199997055863216,
222
+ "pass@128": 0.8383233532934131
223
+ },
224
+ "7": {
225
+ "pass@1": 0.5465051775147929,
226
+ "pass@2": 0.597871919116619,
227
+ "pass@4": 0.6383466158815945,
228
+ "pass@8": 0.6704945279926731,
229
+ "pass@16": 0.6982883012525174,
230
+ "pass@32": 0.7252874533852414,
231
+ "pass@64": 0.7523146294968283,
232
+ "pass@128": 0.7810650887573964
233
+ },
234
+ "8": {
235
+ "pass@1": 0.5412159455128205,
236
+ "pass@2": 0.6090149751413282,
237
+ "pass@4": 0.6620202931364347,
238
+ "pass@8": 0.7088100090850898,
239
+ "pass@16": 0.7524414216865002,
240
+ "pass@32": 0.7903939600021158,
241
+ "pass@64": 0.8223663408035892,
242
+ "pass@128": 0.8557692307692307
243
+ },
244
+ "9": {
245
+ "pass@1": 0.5105413732394366,
246
+ "pass@2": 0.5978214761006982,
247
+ "pass@4": 0.6684741905501246,
248
+ "pass@8": 0.7256541935888985,
249
+ "pass@16": 0.7724577522280122,
250
+ "pass@32": 0.8116780331865703,
251
+ "pass@64": 0.8476223526429457,
252
+ "pass@128": 0.8845070422535212
253
+ }
254
+ }
255
+ },
256
+ "teachers_in_school": {
257
+ "count": 390656,
258
+ "correct": 255040,
259
+ "answer_accuracy": 0.6528505897771952,
260
+ "avg_response_len": 200.41877764580602,
261
+ "resp_tokens_sum": 78294798,
262
+ "pass_at_k": {
263
+ "pass@1": 0.6528505897771952,
264
+ "pass@2": 0.698921128264932,
265
+ "pass@4": 0.7353569988404961,
266
+ "pass@8": 0.7662138431913184,
267
+ "pass@16": 0.7938974098470803,
268
+ "pass@32": 0.820121240427082,
269
+ "pass@64": 0.8452291379587921,
270
+ "pass@128": 0.8689384010484927
271
+ },
272
+ "per_op_pass_at_k": {
273
+ "10": {
274
+ "pass@1": 0.5434240797546013,
275
+ "pass@2": 0.6273198607555187,
276
+ "pass@4": 0.6973762240609492,
277
+ "pass@8": 0.7554670303495753,
278
+ "pass@16": 0.8036698672416946,
279
+ "pass@32": 0.8436957147979897,
280
+ "pass@64": 0.877881304694098,
281
+ "pass@128": 0.9049079754601227
282
+ },
283
+ "2": {
284
+ "pass@1": 0.9736892097264438,
285
+ "pass@2": 0.9847328758107362,
286
+ "pass@4": 0.9901901966851409,
287
+ "pass@8": 0.9927113374191188,
288
+ "pass@16": 0.993648342971609,
289
+ "pass@32": 0.9939055659835022,
290
+ "pass@64": 0.9939209658267383,
291
+ "pass@128": 0.993920972644377
292
+ },
293
+ "3": {
294
+ "pass@1": 0.9402225378787878,
295
+ "pass@2": 0.9562224111190645,
296
+ "pass@4": 0.966675997602572,
297
+ "pass@8": 0.9732372569438384,
298
+ "pass@16": 0.9772623665321553,
299
+ "pass@32": 0.979753273368313,
300
+ "pass@64": 0.9813605854303431,
301
+ "pass@128": 0.9818181818181818
302
+ },
303
+ "4": {
304
+ "pass@1": 0.7226100221893491,
305
+ "pass@2": 0.7654976005218281,
306
+ "pass@4": 0.7946490924026508,
307
+ "pass@8": 0.8150740090995962,
308
+ "pass@16": 0.8315740698722591,
309
+ "pass@32": 0.8478944811913116,
310
+ "pass@64": 0.8650704580299757,
311
+ "pass@128": 0.878698224852071
312
+ },
313
+ "5": {
314
+ "pass@1": 0.6107789855072464,
315
+ "pass@2": 0.6436579938377267,
316
+ "pass@4": 0.670637732783402,
317
+ "pass@8": 0.6953935801960935,
318
+ "pass@16": 0.7181272383034838,
319
+ "pass@32": 0.7399590031087759,
320
+ "pass@64": 0.7619561041686341,
321
+ "pass@128": 0.7855072463768116
322
+ },
323
+ "6": {
324
+ "pass@1": 0.5464272922636103,
325
+ "pass@2": 0.5969523441554045,
326
+ "pass@4": 0.6346913039881474,
327
+ "pass@8": 0.6666373773924721,
328
+ "pass@16": 0.6975353973976225,
329
+ "pass@32": 0.7284124145362817,
330
+ "pass@64": 0.756892771403926,
331
+ "pass@128": 0.7822349570200573
332
+ },
333
+ "7": {
334
+ "pass@1": 0.5512779985549133,
335
+ "pass@2": 0.5971500785125847,
336
+ "pass@4": 0.6349748599885993,
337
+ "pass@8": 0.6696253642996941,
338
+ "pass@16": 0.7024347855609688,
339
+ "pass@32": 0.7341433776297595,
340
+ "pass@64": 0.7661180630503324,
341
+ "pass@128": 0.8034682080924855
342
+ },
343
+ "8": {
344
+ "pass@1": 0.5112676056338028,
345
+ "pass@2": 0.5702083564378394,
346
+ "pass@4": 0.6210338083619829,
347
+ "pass@8": 0.6679787729679809,
348
+ "pass@16": 0.7144538059651695,
349
+ "pass@32": 0.7619468940389942,
350
+ "pass@64": 0.8071898332537814,
351
+ "pass@128": 0.847887323943662
352
+ },
353
+ "9": {
354
+ "pass@1": 0.4994620134730539,
355
+ "pass@2": 0.5715450251072656,
356
+ "pass@4": 0.6303960414379337,
357
+ "pass@8": 0.6812784051466028,
358
+ "pass@16": 0.7264440386053771,
359
+ "pass@32": 0.7696312332548112,
360
+ "pass@64": 0.8130165993358162,
361
+ "pass@128": 0.8562874251497006
362
+ }
363
+ }
364
+ },
365
+ "movie_festival_awards": {
366
+ "count": 382336,
367
+ "correct": 252327,
368
+ "answer_accuracy": 0.6599613952125879,
369
+ "avg_response_len": 214.17624811683964,
370
+ "resp_tokens_sum": 81887290,
371
+ "pass_at_k": {
372
+ "pass@1": 0.6599613952125879,
373
+ "pass@2": 0.7102219443704902,
374
+ "pass@4": 0.7513175537000822,
375
+ "pass@8": 0.7872584865566714,
376
+ "pass@16": 0.819415299000816,
377
+ "pass@32": 0.8483489704762341,
378
+ "pass@64": 0.8738792518830564,
379
+ "pass@128": 0.8958821560093739
380
+ },
381
+ "per_op_pass_at_k": {
382
+ "10": {
383
+ "pass@1": 0.4935401119402985,
384
+ "pass@2": 0.5670470237395698,
385
+ "pass@4": 0.6367358320881525,
386
+ "pass@8": 0.7042430815093631,
387
+ "pass@16": 0.7680596090454958,
388
+ "pass@32": 0.8246740921435434,
389
+ "pass@64": 0.8697867463781164,
390
+ "pass@128": 0.9044776119402985
391
+ },
392
+ "2": {
393
+ "pass@1": 0.9596875,
394
+ "pass@2": 0.9764113470191226,
395
+ "pass@4": 0.9854286804328029,
396
+ "pass@8": 0.991061509280679,
397
+ "pass@16": 0.9942260504228976,
398
+ "pass@32": 0.9956652788647895,
399
+ "pass@64": 0.9971427773528576,
400
+ "pass@128": 1.0
401
+ },
402
+ "3": {
403
+ "pass@1": 0.9397093658892128,
404
+ "pass@2": 0.9562255371777508,
405
+ "pass@4": 0.9668405375166294,
406
+ "pass@8": 0.9747146666399417,
407
+ "pass@16": 0.9801107267765427,
408
+ "pass@32": 0.9832217992096687,
409
+ "pass@64": 0.984688404960452,
410
+ "pass@128": 0.9854227405247813
411
+ },
412
+ "4": {
413
+ "pass@1": 0.7007264254385965,
414
+ "pass@2": 0.7445909310678271,
415
+ "pass@4": 0.7808669256255251,
416
+ "pass@8": 0.8129850113269721,
417
+ "pass@16": 0.842129991729984,
418
+ "pass@32": 0.8674656215356255,
419
+ "pass@64": 0.8884742826156206,
420
+ "pass@128": 0.9064327485380117
421
+ },
422
+ "5": {
423
+ "pass@1": 0.5817784926470588,
424
+ "pass@2": 0.6204282943492355,
425
+ "pass@4": 0.6536405768212797,
426
+ "pass@8": 0.68463428746883,
427
+ "pass@16": 0.7157901961439006,
428
+ "pass@32": 0.7474912605697788,
429
+ "pass@64": 0.7777573616250464,
430
+ "pass@128": 0.8029411764705883
431
+ },
432
+ "6": {
433
+ "pass@1": 0.5934296135646687,
434
+ "pass@2": 0.64594756762463,
435
+ "pass@4": 0.6914165587345744,
436
+ "pass@8": 0.7364669504842055,
437
+ "pass@16": 0.778883156850051,
438
+ "pass@32": 0.8167598694063396,
439
+ "pass@64": 0.8509370624044789,
440
+ "pass@128": 0.8801261829652997
441
+ },
442
+ "7": {
443
+ "pass@1": 0.5463558148734177,
444
+ "pass@2": 0.6027772413535332,
445
+ "pass@4": 0.6477394363481457,
446
+ "pass@8": 0.6858369916354995,
447
+ "pass@16": 0.7205188747595422,
448
+ "pass@32": 0.7560440209344339,
449
+ "pass@64": 0.7912034540169977,
450
+ "pass@128": 0.8259493670886076
451
+ },
452
+ "8": {
453
+ "pass@1": 0.5558370870870871,
454
+ "pass@2": 0.6316111879596129,
455
+ "pass@4": 0.6897773594742097,
456
+ "pass@8": 0.736390876197389,
457
+ "pass@16": 0.7772959628816623,
458
+ "pass@32": 0.8148386053479606,
459
+ "pass@64": 0.8483111557392278,
460
+ "pass@128": 0.8738738738738738
461
+ },
462
+ "9": {
463
+ "pass@1": 0.5287630627009646,
464
+ "pass@2": 0.6127898163151629,
465
+ "pass@4": 0.680057331961318,
466
+ "pass@8": 0.7337678523669866,
467
+ "pass@16": 0.7759801527561564,
468
+ "pass@32": 0.8104190136847873,
469
+ "pass@64": 0.841157580846551,
470
+ "pass@128": 0.8713826366559485
471
+ }
472
+ }
473
+ }
474
+ }
475
+ },
476
+ "ood": {
477
+ "answer_accuracy": 0.1743203125,
478
+ "avg_response_len": 301.8644203125,
479
+ "avg_loss": 0.0,
480
+ "count": 1280000,
481
+ "validation_metrics": {},
482
+ "pass_at_k": {
483
+ "pass@1": 0.1743203125,
484
+ "pass@2": 0.24371594488189435,
485
+ "pass@4": 0.3202035408511474,
486
+ "pass@8": 0.39943672447112044,
487
+ "pass@16": 0.475507100100325,
488
+ "pass@32": 0.5423402745555124,
489
+ "pass@64": 0.5967209250393932,
490
+ "pass@128": 0.6402
491
+ },
492
+ "per_op_pass_at_k": {
493
+ "11": {
494
+ "pass@1": 0.436265625,
495
+ "pass@2": 0.5340772637795281,
496
+ "pass@4": 0.6203406313273335,
497
+ "pass@8": 0.6955876483796678,
498
+ "pass@16": 0.760309368115832,
499
+ "pass@32": 0.8152574932269782,
500
+ "pass@64": 0.8607695816340385,
501
+ "pass@128": 0.898
502
+ },
503
+ "12": {
504
+ "pass@1": 0.3764765625,
505
+ "pass@2": 0.49078690944881975,
506
+ "pass@4": 0.5897831228908883,
507
+ "pass@8": 0.6752086536338234,
508
+ "pass@16": 0.7505092156215493,
509
+ "pass@32": 0.8143294232531485,
510
+ "pass@64": 0.8625700947965467,
511
+ "pass@128": 0.897
512
+ },
513
+ "13": {
514
+ "pass@1": 0.1875546875,
515
+ "pass@2": 0.2765618848425194,
516
+ "pass@4": 0.3746293088676413,
517
+ "pass@8": 0.47422989814980515,
518
+ "pass@16": 0.570095611224325,
519
+ "pass@32": 0.6559800165690604,
520
+ "pass@64": 0.7268097914764812,
521
+ "pass@128": 0.784
522
+ },
523
+ "14": {
524
+ "pass@1": 0.1202734375,
525
+ "pass@2": 0.1822513533464566,
526
+ "pass@4": 0.2575197916197975,
527
+ "pass@8": 0.34121546101434963,
528
+ "pass@16": 0.4264310580383103,
529
+ "pass@32": 0.5056039816537211,
530
+ "pass@64": 0.5742541439181715,
531
+ "pass@128": 0.633
532
+ },
533
+ "15": {
534
+ "pass@1": 0.10128125,
535
+ "pass@2": 0.15608230807086593,
536
+ "pass@4": 0.22474121944131967,
537
+ "pass@8": 0.3040805629049734,
538
+ "pass@16": 0.38644958428799214,
539
+ "pass@32": 0.4625041528444439,
540
+ "pass@64": 0.5274360687349702,
541
+ "pass@128": 0.582
542
+ },
543
+ "16": {
544
+ "pass@1": 0.1056875,
545
+ "pass@2": 0.15908267716535426,
546
+ "pass@4": 0.2240340161229844,
547
+ "pass@8": 0.2959581191868564,
548
+ "pass@16": 0.3675161749690196,
549
+ "pass@32": 0.431845456719432,
550
+ "pass@64": 0.4875276074685491,
551
+ "pass@128": 0.535
552
+ },
553
+ "17": {
554
+ "pass@1": 0.11615625,
555
+ "pass@2": 0.1744458661417323,
556
+ "pass@4": 0.24267885104986836,
557
+ "pass@8": 0.3148497774165068,
558
+ "pass@16": 0.38519763091511355,
559
+ "pass@32": 0.44891700121322214,
560
+ "pass@64": 0.5027736794825323,
561
+ "pass@128": 0.548
562
+ },
563
+ "18": {
564
+ "pass@1": 0.1071171875,
565
+ "pass@2": 0.16290108267716508,
566
+ "pass@4": 0.23030277652793374,
567
+ "pass@8": 0.3041988013351102,
568
+ "pass@16": 0.3763951476213891,
569
+ "pass@32": 0.4395935170131949,
570
+ "pass@64": 0.4896073566301441,
571
+ "pass@128": 0.526
572
+ },
573
+ "19": {
574
+ "pass@1": 0.092125,
575
+ "pass@2": 0.14406188484251944,
576
+ "pass@4": 0.20898489004499407,
577
+ "pass@8": 0.2810631586756763,
578
+ "pass@16": 0.35164518784344284,
579
+ "pass@32": 0.4126498000665176,
580
+ "pass@64": 0.46098748128887995,
581
+ "pass@128": 0.499
582
+ },
583
+ "20": {
584
+ "pass@1": 0.100265625,
585
+ "pass@2": 0.15690821850393682,
586
+ "pass@4": 0.2290208006186725,
587
+ "pass@8": 0.3079751640143204,
588
+ "pass@16": 0.38052202236614363,
589
+ "pass@32": 0.4367219029954224,
590
+ "pass@64": 0.4744734449635939,
591
+ "pass@128": 0.5
592
+ }
593
+ },
594
+ "per_op_accuracy": {
595
+ "11": 0.436265625,
596
+ "12": 0.3764765625,
597
+ "13": 0.1875546875,
598
+ "14": 0.1202734375,
599
+ "15": 0.10128125,
600
+ "16": 0.1056875,
601
+ "17": 0.11615625,
602
+ "18": 0.1071171875,
603
+ "19": 0.092125,
604
+ "20": 0.100265625
605
+ },
606
+ "per_op_avg_response_len": {
607
+ "11": 291.024140625,
608
+ "12": 294.8629609375,
609
+ "13": 297.5098203125,
610
+ "14": 300.940375,
611
+ "15": 300.8659140625,
612
+ "16": 302.8726171875,
613
+ "17": 307.435828125,
614
+ "18": 305.881828125,
615
+ "19": 306.381546875,
616
+ "20": 310.869171875
617
+ },
618
+ "per_op_avg_loss": {},
619
+ "per_op_length": {
620
+ "11": 291.024140625,
621
+ "12": 294.8629609375,
622
+ "13": 297.5098203125,
623
+ "14": 300.940375,
624
+ "15": 300.8659140625,
625
+ "16": 302.8726171875,
626
+ "17": 307.435828125,
627
+ "18": 305.881828125,
628
+ "19": 306.381546875,
629
+ "20": 310.869171875
630
+ },
631
+ "per_op_loss": {},
632
+ "per_template": {
633
+ "crazy_zootopia": {
634
+ "count": 430720,
635
+ "correct": 74714,
636
+ "answer_accuracy": 0.17346303863298662,
637
+ "avg_response_len": 287.7524122399703,
638
+ "resp_tokens_sum": 123940719,
639
+ "pass_at_k": {
640
+ "pass@1": 0.17346303863298662,
641
+ "pass@2": 0.24261010313439718,
642
+ "pass@4": 0.3188155590785188,
643
+ "pass@8": 0.3973496364292067,
644
+ "pass@16": 0.4722451193690256,
645
+ "pass@32": 0.5382268181213222,
646
+ "pass@64": 0.5930338995005733,
647
+ "pass@128": 0.6389301634472511
648
+ },
649
+ "per_op_pass_at_k": {
650
+ "11": {
651
+ "pass@1": 0.41977248583569404,
652
+ "pass@2": 0.5152966836563977,
653
+ "pass@4": 0.5988550168835131,
654
+ "pass@8": 0.6740992829329473,
655
+ "pass@16": 0.7422318681552861,
656
+ "pass@32": 0.8019395682288849,
657
+ "pass@64": 0.8536206281446311,
658
+ "pass@128": 0.8980169971671388
659
+ },
660
+ "12": {
661
+ "pass@1": 0.3938912429378531,
662
+ "pass@2": 0.5104100632812844,
663
+ "pass@4": 0.6080495266481519,
664
+ "pass@8": 0.68796901487411,
665
+ "pass@16": 0.7544555654168367,
666
+ "pass@32": 0.8096660100272961,
667
+ "pass@64": 0.8520104842456188,
668
+ "pass@128": 0.884180790960452
669
+ },
670
+ "13": {
671
+ "pass@1": 0.17650372257053293,
672
+ "pass@2": 0.26551083911337076,
673
+ "pass@4": 0.3665973341302558,
674
+ "pass@8": 0.4710679392988577,
675
+ "pass@16": 0.5713484839319661,
676
+ "pass@32": 0.6590817812645573,
677
+ "pass@64": 0.7304700839844447,
678
+ "pass@128": 0.7899686520376176
679
+ },
680
+ "14": {
681
+ "pass@1": 0.11207651869158879,
682
+ "pass@2": 0.17378371292957537,
683
+ "pass@4": 0.25215351060089447,
684
+ "pass@8": 0.3400416779501758,
685
+ "pass@16": 0.42831800765620487,
686
+ "pass@32": 0.5089598246736358,
687
+ "pass@64": 0.5781006993506586,
688
+ "pass@128": 0.6386292834890965
689
+ },
690
+ "15": {
691
+ "pass@1": 0.0994400289017341,
692
+ "pass@2": 0.15031248577670558,
693
+ "pass@4": 0.21217345971204465,
694
+ "pass@8": 0.2830047929574361,
695
+ "pass@16": 0.35940929952555856,
696
+ "pass@32": 0.4361782265948589,
697
+ "pass@64": 0.5084135452231018,
698
+ "pass@128": 0.5751445086705202
699
+ },
700
+ "16": {
701
+ "pass@1": 0.09947076612903226,
702
+ "pass@2": 0.15187325374650743,
703
+ "pass@4": 0.2179497756328846,
704
+ "pass@8": 0.2926177158647718,
705
+ "pass@16": 0.3690935358079865,
706
+ "pass@32": 0.4399879447067403,
707
+ "pass@64": 0.49926141465739254,
708
+ "pass@128": 0.5451612903225806
709
+ },
710
+ "17": {
711
+ "pass@1": 0.10357207369942197,
712
+ "pass@2": 0.1580094215101725,
713
+ "pass@4": 0.22143721648233278,
714
+ "pass@8": 0.2883468654318178,
715
+ "pass@16": 0.35307511467549146,
716
+ "pass@32": 0.4110660972285916,
717
+ "pass@64": 0.45948745382879763,
718
+ "pass@128": 0.5
719
+ },
720
+ "18": {
721
+ "pass@1": 0.11919642857142858,
722
+ "pass@2": 0.17823186164229474,
723
+ "pass@4": 0.24767612941239495,
724
+ "pass@8": 0.32057904558995565,
725
+ "pass@16": 0.3888781464313517,
726
+ "pass@32": 0.44704524193215434,
727
+ "pass@64": 0.4943156741255045,
728
+ "pass@128": 0.5314285714285715
729
+ },
730
+ "19": {
731
+ "pass@1": 0.08955048798798798,
732
+ "pass@2": 0.14224620782199515,
733
+ "pass@4": 0.21036005099587773,
734
+ "pass@8": 0.2858482595544025,
735
+ "pass@16": 0.3575037208007843,
736
+ "pass@32": 0.4184082663584716,
737
+ "pass@64": 0.4677929962465217,
738
+ "pass@128": 0.5075075075075075
739
+ },
740
+ "20": {
741
+ "pass@1": 0.09365615615615616,
742
+ "pass@2": 0.14956861388948,
743
+ "pass@4": 0.22208081000010135,
744
+ "pass@8": 0.301411696925859,
745
+ "pass@16": 0.37289719680717975,
746
+ "pass@32": 0.4278229117854127,
747
+ "pass@64": 0.4672031815887534,
748
+ "pass@128": 0.5015015015015015
749
+ }
750
+ }
751
+ },
752
+ "teachers_in_school": {
753
+ "count": 430464,
754
+ "correct": 75838,
755
+ "answer_accuracy": 0.176177334225394,
756
+ "avg_response_len": 298.75116618346715,
757
+ "resp_tokens_sum": 128601622,
758
+ "pass_at_k": {
759
+ "pass@1": 0.176177334225394,
760
+ "pass@2": 0.2462267414499163,
761
+ "pass@4": 0.323831370766433,
762
+ "pass@8": 0.40464838810354997,
763
+ "pass@16": 0.4823127322919932,
764
+ "pass@32": 0.5500811664576809,
765
+ "pass@64": 0.6043942535502497,
766
+ "pass@128": 0.6458519179304193
767
+ },
768
+ "per_op_pass_at_k": {
769
+ "11": {
770
+ "pass@1": 0.4587019954819277,
771
+ "pass@2": 0.5552945047908173,
772
+ "pass@4": 0.6409966999606971,
773
+ "pass@8": 0.7148040088201414,
774
+ "pass@16": 0.7767393045658387,
775
+ "pass@32": 0.8297072308383238,
776
+ "pass@64": 0.8746833656520376,
777
+ "pass@128": 0.9126506024096386
778
+ },
779
+ "12": {
780
+ "pass@1": 0.3535952194357367,
781
+ "pass@2": 0.46705609927677505,
782
+ "pass@4": 0.5703037878494028,
783
+ "pass@8": 0.6629286050294428,
784
+ "pass@16": 0.7478458624123885,
785
+ "pass@32": 0.8187617469065239,
786
+ "pass@64": 0.8679221722274657,
787
+ "pass@128": 0.9028213166144201
788
+ },
789
+ "13": {
790
+ "pass@1": 0.20017069327731093,
791
+ "pass@2": 0.29245896193122883,
792
+ "pass@4": 0.39196500384930894,
793
+ "pass@8": 0.49107469341212867,
794
+ "pass@16": 0.5849278755630151,
795
+ "pass@32": 0.66909629091827,
796
+ "pass@64": 0.7394157298961205,
797
+ "pass@128": 0.7927170868347339
798
+ },
799
+ "14": {
800
+ "pass@1": 0.12148209064327485,
801
+ "pass@2": 0.18374897833494486,
802
+ "pass@4": 0.2576632594719518,
803
+ "pass@8": 0.33923709003618235,
804
+ "pass@16": 0.42476920910139276,
805
+ "pass@32": 0.5074564490761422,
806
+ "pass@64": 0.5803711048868923,
807
+ "pass@128": 0.6403508771929824
808
+ },
809
+ "15": {
810
+ "pass@1": 0.09886363636363636,
811
+ "pass@2": 0.15513675137198757,
812
+ "pass@4": 0.22539836440899438,
813
+ "pass@8": 0.3055976387358603,
814
+ "pass@16": 0.38618518854570155,
815
+ "pass@32": 0.45698552030349504,
816
+ "pass@64": 0.5147245657272281,
817
+ "pass@128": 0.5575757575757576
818
+ },
819
+ "16": {
820
+ "pass@1": 0.10803952991452992,
821
+ "pass@2": 0.15944531372680978,
822
+ "pass@4": 0.22162470449740793,
823
+ "pass@8": 0.29181218523663216,
824
+ "pass@16": 0.36106886230415675,
825
+ "pass@32": 0.42176417751644407,
826
+ "pass@64": 0.47626160158795905,
827
+ "pass@128": 0.5299145299145299
828
+ },
829
+ "17": {
830
+ "pass@1": 0.12783870341614906,
831
+ "pass@2": 0.19124290849513367,
832
+ "pass@4": 0.2656787292116435,
833
+ "pass@8": 0.3460024308333159,
834
+ "pass@16": 0.42595468112596735,
835
+ "pass@32": 0.4986195373338599,
836
+ "pass@64": 0.5599232484124532,
837
+ "pass@128": 0.6118012422360248
838
+ },
839
+ "18": {
840
+ "pass@1": 0.11353400735294118,
841
+ "pass@2": 0.17243262216303845,
842
+ "pass@4": 0.2442672787225127,
843
+ "pass@8": 0.3239806135278011,
844
+ "pass@16": 0.39993769990663836,
845
+ "pass@32": 0.4633629313271958,
846
+ "pass@64": 0.5111777845425836,
847
+ "pass@128": 0.5441176470588235
848
+ },
849
+ "19": {
850
+ "pass@1": 0.10014415922619048,
851
+ "pass@2": 0.1556139711051743,
852
+ "pass@4": 0.22348070581579088,
853
+ "pass@8": 0.29794332466542117,
854
+ "pass@16": 0.37017448624899146,
855
+ "pass@32": 0.43114841416760863,
856
+ "pass@64": 0.4765716994613993,
857
+ "pass@128": 0.5059523809523809
858
+ },
859
+ "20": {
860
+ "pass@1": 0.09110684880239521,
861
+ "pass@2": 0.14313720000942995,
862
+ "pass@4": 0.21076069402252862,
863
+ "pass@8": 0.28671310390061244,
864
+ "pass@16": 0.3586706072592117,
865
+ "pass@32": 0.4163128162045844,
866
+ "pass@64": 0.4539123901453239,
867
+ "pass@128": 0.47005988023952094
868
+ }
869
+ }
870
+ },
871
+ "movie_festival_awards": {
872
+ "count": 418816,
873
+ "correct": 72578,
874
+ "answer_accuracy": 0.17329328392420537,
875
+ "avg_response_len": 319.57737287973714,
876
+ "resp_tokens_sum": 133844117,
877
+ "pass_at_k": {
878
+ "pass@1": 0.17329328392420537,
879
+ "pass@2": 0.24227259177126953,
880
+ "pass@4": 0.31790224704300396,
881
+ "pass@8": 0.396226524735505,
882
+ "pass@16": 0.471866887416981,
883
+ "pass@32": 0.5386144681478255,
884
+ "pass@64": 0.5926260097448032,
885
+ "pass@128": 0.6356968215158925
886
+ },
887
+ "per_op_pass_at_k": {
888
+ "11": {
889
+ "pass@1": 0.4311011904761905,
890
+ "pass@2": 0.5327611392325959,
891
+ "pass@4": 0.6226473205730233,
892
+ "pass@8": 0.699414827225559,
893
+ "pass@16": 0.7632510144163813,
894
+ "pass@32": 0.8149524603297097,
895
+ "pass@64": 0.8541162619127182,
896
+ "pass@128": 0.8825396825396825
897
+ },
898
+ "12": {
899
+ "pass@1": 0.37994552752293576,
900
+ "pass@2": 0.49269373510077263,
901
+ "pass@4": 0.5890112603470439,
902
+ "pass@8": 0.673374303865433,
903
+ "pass@16": 0.7488352151511932,
904
+ "pass@32": 0.8150540013464985,
905
+ "pass@64": 0.8687804600398654,
906
+ "pass@128": 0.9051987767584098
907
+ },
908
+ "13": {
909
+ "pass@1": 0.18453414351851852,
910
+ "pass@2": 0.26992616591328855,
911
+ "pass@4": 0.36343596575891,
912
+ "pass@8": 0.45878259248561787,
913
+ "pass@16": 0.5525191150433114,
914
+ "pass@32": 0.6384739274316215,
915
+ "pass@64": 0.709316108372,
916
+ "pass@128": 0.7685185185185185
917
+ },
918
+ "14": {
919
+ "pass@1": 0.12685459940652818,
920
+ "pass@2": 0.18879712434869975,
921
+ "pass@4": 0.2624856972626196,
922
+ "pass@8": 0.3443412391690488,
923
+ "pass@16": 0.4263202019821751,
924
+ "pass@32": 0.5005275143900397,
925
+ "pass@64": 0.5643824971967146,
926
+ "pass@128": 0.6201780415430267
927
+ },
928
+ "15": {
929
+ "pass@1": 0.10570987654320987,
930
+ "pass@2": 0.16320697543015447,
931
+ "pass@4": 0.23749303125303775,
932
+ "pass@8": 0.32504223382366393,
933
+ "pass@16": 0.41559522972860385,
934
+ "pass@32": 0.49623847142737515,
935
+ "pass@64": 0.5606971462894773,
936
+ "pass@128": 0.6141975308641975
937
+ },
938
+ "16": {
939
+ "pass@1": 0.10893713126843659,
940
+ "pass@2": 0.1652998919935893,
941
+ "pass@4": 0.23209237285604775,
942
+ "pass@8": 0.3033054579667241,
943
+ "pass@16": 0.3727492867250306,
944
+ "pass@32": 0.43483766239548877,
945
+ "pass@64": 0.48846237984478985,
946
+ "pass@128": 0.5309734513274337
947
+ },
948
+ "17": {
949
+ "pass@1": 0.11794051204819277,
950
+ "pass@2": 0.17528430651740828,
951
+ "pass@4": 0.24250910644829024,
952
+ "pass@8": 0.31225596159268304,
953
+ "pass@16": 0.37914528287600097,
954
+ "pass@32": 0.44015855587538144,
955
+ "pass@64": 0.49245703153300757,
956
+ "pass@128": 0.536144578313253
957
+ },
958
+ "18": {
959
+ "pass@1": 0.08644153225806452,
960
+ "pass@2": 0.1351381921513843,
961
+ "pass@4": 0.19537179505787589,
962
+ "pass@8": 0.2640087960618504,
963
+ "pass@16": 0.3364805754908361,
964
+ "pass@32": 0.40511059898611057,
965
+ "pass@64": 0.4606336256185126,
966
+ "pass@128": 0.5
967
+ },
968
+ "19": {
969
+ "pass@1": 0.08657477341389729,
970
+ "pass@2": 0.13416194364488426,
971
+ "pass@4": 0.19288663416695273,
972
+ "pass@8": 0.25911399140930275,
973
+ "pass@16": 0.32694205872241794,
974
+ "pass@32": 0.3880784900568889,
975
+ "pass@64": 0.43832121607177427,
976
+ "pass@128": 0.48338368580060426
977
+ },
978
+ "20": {
979
+ "pass@1": 0.11606137387387387,
980
+ "pass@2": 0.17806019602279444,
981
+ "pass@4": 0.2542757329883313,
982
+ "pass@8": 0.33586454124686205,
983
+ "pass@16": 0.41006388289722623,
984
+ "pass@32": 0.4660912695451911,
985
+ "pass@64": 0.5023665080660694,
986
+ "pass@128": 0.5285285285285285
987
+ }
988
+ }
989
+ }
990
+ }
991
+ },
992
+ "total": {
993
+ "answer_accuracy": 0.4039703947368421,
994
+ "avg_loss": 0.0,
995
+ "count": 2432000,
996
+ "validation_metrics": {
997
+ "id": {},
998
+ "ood": {}
999
+ },
1000
+ "pass_at_k": {
1001
+ "pass@1": 0.4039703947368421,
1002
+ "pass@2": 0.46338382589098,
1003
+ "pass@4": 0.5218226988090147,
1004
+ "pass@8": 0.5790316456982592,
1005
+ "pass@16": 0.6328354201869616,
1006
+ "pass@32": 0.680430043267418,
1007
+ "pass@64": 0.7203239479096087,
1008
+ "pass@128": 0.7536842105263157
1009
+ },
1010
+ "per_op_pass_at_k": {
1011
+ "10": {
1012
+ "pass@1": 0.5284453125,
1013
+ "pass@2": 0.6050858759842528,
1014
+ "pass@4": 0.6711347938695169,
1015
+ "pass@8": 0.7306809613281867,
1016
+ "pass@16": 0.7840277284668682,
1017
+ "pass@32": 0.8297723522006395,
1018
+ "pass@64": 0.867259596306628,
1019
+ "pass@128": 0.897
1020
+ },
1021
+ "2": {
1022
+ "pass@1": 0.966640625,
1023
+ "pass@2": 0.9790877214566934,
1024
+ "pass@4": 0.9861680459317581,
1025
+ "pass@8": 0.9908336649877861,
1026
+ "pass@16": 0.9935962945915,
1027
+ "pass@32": 0.9948623964784657,
1028
+ "pass@64": 0.9957516213419054,
1029
+ "pass@128": 0.997
1030
+ },
1031
+ "3": {
1032
+ "pass@1": 0.944015625,
1033
+ "pass@2": 0.9593532234251971,
1034
+ "pass@4": 0.9692371754780653,
1035
+ "pass@8": 0.976086913788811,
1036
+ "pass@16": 0.9806792820711105,
1037
+ "pass@32": 0.9835232837747843,
1038
+ "pass@64": 0.9852268502493713,
1039
+ "pass@128": 0.986
1040
+ },
1041
+ "4": {
1042
+ "pass@1": 0.7210234375,
1043
+ "pass@2": 0.7622476624015748,
1044
+ "pass@4": 0.7931688311773528,
1045
+ "pass@8": 0.8180216883664996,
1046
+ "pass@16": 0.8392104673213089,
1047
+ "pass@32": 0.8579530107784558,
1048
+ "pass@64": 0.8754655959158385,
1049
+ "pass@128": 0.892
1050
+ },
1051
+ "5": {
1052
+ "pass@1": 0.5931328125,
1053
+ "pass@2": 0.6308367372047244,
1054
+ "pass@4": 0.6605045965504309,
1055
+ "pass@8": 0.6864019188998131,
1056
+ "pass@16": 0.7111635789832254,
1057
+ "pass@32": 0.735359380629359,
1058
+ "pass@64": 0.7584082913454374,
1059
+ "pass@128": 0.781
1060
+ },
1061
+ "6": {
1062
+ "pass@1": 0.58290625,
1063
+ "pass@2": 0.634886318897638,
1064
+ "pass@4": 0.6761626663854513,
1065
+ "pass@8": 0.71331148317756,
1066
+ "pass@16": 0.7479597662974471,
1067
+ "pass@32": 0.7795488087145294,
1068
+ "pass@64": 0.8077825276680214,
1069
+ "pass@128": 0.832
1070
+ },
1071
+ "7": {
1072
+ "pass@1": 0.548109375,
1073
+ "pass@2": 0.599172244094488,
1074
+ "pass@4": 0.6401481196100498,
1075
+ "pass@8": 0.6750420158660341,
1076
+ "pass@16": 0.7067478460514601,
1077
+ "pass@32": 0.7380706785193885,
1078
+ "pass@64": 0.7693794860547157,
1079
+ "pass@128": 0.803
1080
+ },
1081
+ "8": {
1082
+ "pass@1": 0.535453125,
1083
+ "pass@2": 0.60276316437008,
1084
+ "pass@4": 0.6567131941319833,
1085
+ "pass@8": 0.7034993490119114,
1086
+ "pass@16": 0.7472323803234155,
1087
+ "pass@32": 0.7884353184853741,
1088
+ "pass@64": 0.8256183039969759,
1089
+ "pass@128": 0.859
1090
+ },
1091
+ "9": {
1092
+ "pass@1": 0.5125078125,
1093
+ "pass@2": 0.5937002952755916,
1094
+ "pass@4": 0.6593584457255333,
1095
+ "pass@8": 0.7133560281291572,
1096
+ "pass@16": 0.7581846384423049,
1097
+ "pass@32": 0.7972428469443084,
1098
+ "pass@64": 0.8340534870096861,
1099
+ "pass@128": 0.871
1100
+ },
1101
+ "11": {
1102
+ "pass@1": 0.436265625,
1103
+ "pass@2": 0.5340772637795281,
1104
+ "pass@4": 0.6203406313273335,
1105
+ "pass@8": 0.6955876483796678,
1106
+ "pass@16": 0.760309368115832,
1107
+ "pass@32": 0.8152574932269782,
1108
+ "pass@64": 0.8607695816340385,
1109
+ "pass@128": 0.898
1110
+ },
1111
+ "12": {
1112
+ "pass@1": 0.3764765625,
1113
+ "pass@2": 0.49078690944881975,
1114
+ "pass@4": 0.5897831228908883,
1115
+ "pass@8": 0.6752086536338234,
1116
+ "pass@16": 0.7505092156215493,
1117
+ "pass@32": 0.8143294232531485,
1118
+ "pass@64": 0.8625700947965467,
1119
+ "pass@128": 0.897
1120
+ },
1121
+ "13": {
1122
+ "pass@1": 0.1875546875,
1123
+ "pass@2": 0.2765618848425194,
1124
+ "pass@4": 0.3746293088676413,
1125
+ "pass@8": 0.47422989814980515,
1126
+ "pass@16": 0.570095611224325,
1127
+ "pass@32": 0.6559800165690604,
1128
+ "pass@64": 0.7268097914764812,
1129
+ "pass@128": 0.784
1130
+ },
1131
+ "14": {
1132
+ "pass@1": 0.1202734375,
1133
+ "pass@2": 0.1822513533464566,
1134
+ "pass@4": 0.2575197916197975,
1135
+ "pass@8": 0.34121546101434963,
1136
+ "pass@16": 0.4264310580383103,
1137
+ "pass@32": 0.5056039816537211,
1138
+ "pass@64": 0.5742541439181715,
1139
+ "pass@128": 0.633
1140
+ },
1141
+ "15": {
1142
+ "pass@1": 0.10128125,
1143
+ "pass@2": 0.15608230807086593,
1144
+ "pass@4": 0.22474121944131967,
1145
+ "pass@8": 0.3040805629049734,
1146
+ "pass@16": 0.38644958428799214,
1147
+ "pass@32": 0.4625041528444439,
1148
+ "pass@64": 0.5274360687349702,
1149
+ "pass@128": 0.582
1150
+ },
1151
+ "16": {
1152
+ "pass@1": 0.1056875,
1153
+ "pass@2": 0.15908267716535426,
1154
+ "pass@4": 0.2240340161229844,
1155
+ "pass@8": 0.2959581191868564,
1156
+ "pass@16": 0.3675161749690196,
1157
+ "pass@32": 0.431845456719432,
1158
+ "pass@64": 0.4875276074685491,
1159
+ "pass@128": 0.535
1160
+ },
1161
+ "17": {
1162
+ "pass@1": 0.11615625,
1163
+ "pass@2": 0.1744458661417323,
1164
+ "pass@4": 0.24267885104986836,
1165
+ "pass@8": 0.3148497774165068,
1166
+ "pass@16": 0.38519763091511355,
1167
+ "pass@32": 0.44891700121322214,
1168
+ "pass@64": 0.5027736794825323,
1169
+ "pass@128": 0.548
1170
+ },
1171
+ "18": {
1172
+ "pass@1": 0.1071171875,
1173
+ "pass@2": 0.16290108267716508,
1174
+ "pass@4": 0.23030277652793374,
1175
+ "pass@8": 0.3041988013351102,
1176
+ "pass@16": 0.3763951476213891,
1177
+ "pass@32": 0.4395935170131949,
1178
+ "pass@64": 0.4896073566301441,
1179
+ "pass@128": 0.526
1180
+ },
1181
+ "19": {
1182
+ "pass@1": 0.092125,
1183
+ "pass@2": 0.14406188484251944,
1184
+ "pass@4": 0.20898489004499407,
1185
+ "pass@8": 0.2810631586756763,
1186
+ "pass@16": 0.35164518784344284,
1187
+ "pass@32": 0.4126498000665176,
1188
+ "pass@64": 0.46098748128887995,
1189
+ "pass@128": 0.499
1190
+ },
1191
+ "20": {
1192
+ "pass@1": 0.100265625,
1193
+ "pass@2": 0.15690821850393682,
1194
+ "pass@4": 0.2290208006186725,
1195
+ "pass@8": 0.3079751640143204,
1196
+ "pass@16": 0.38052202236614363,
1197
+ "pass@32": 0.4367219029954224,
1198
+ "pass@64": 0.4744734449635939,
1199
+ "pass@128": 0.5
1200
+ }
1201
+ },
1202
+ "per_op_accuracy": {
1203
+ "10": 0.5284453125,
1204
+ "2": 0.966640625,
1205
+ "3": 0.944015625,
1206
+ "4": 0.7210234375,
1207
+ "5": 0.5931328125,
1208
+ "6": 0.58290625,
1209
+ "7": 0.548109375,
1210
+ "8": 0.535453125,
1211
+ "9": 0.5125078125,
1212
+ "11": 0.436265625,
1213
+ "12": 0.3764765625,
1214
+ "13": 0.1875546875,
1215
+ "14": 0.1202734375,
1216
+ "15": 0.10128125,
1217
+ "16": 0.1056875,
1218
+ "17": 0.11615625,
1219
+ "18": 0.1071171875,
1220
+ "19": 0.092125,
1221
+ "20": 0.100265625
1222
+ },
1223
+ "per_op_avg_response_len": {
1224
+ "10": 278.1583203125,
1225
+ "2": 121.0686328125,
1226
+ "3": 147.5285625,
1227
+ "4": 161.5481484375,
1228
+ "5": 177.994421875,
1229
+ "6": 203.069109375,
1230
+ "7": 226.3078828125,
1231
+ "8": 246.1722421875,
1232
+ "9": 261.254546875,
1233
+ "11": 291.024140625,
1234
+ "12": 294.8629609375,
1235
+ "13": 297.5098203125,
1236
+ "14": 300.940375,
1237
+ "15": 300.8659140625,
1238
+ "16": 302.8726171875,
1239
+ "17": 307.435828125,
1240
+ "18": 305.881828125,
1241
+ "19": 306.381546875,
1242
+ "20": 310.869171875
1243
+ },
1244
+ "per_op_avg_loss": {},
1245
+ "per_op_length": {
1246
+ "10": 278.1583203125,
1247
+ "2": 121.0686328125,
1248
+ "3": 147.5285625,
1249
+ "4": 161.5481484375,
1250
+ "5": 177.994421875,
1251
+ "6": 203.069109375,
1252
+ "7": 226.3078828125,
1253
+ "8": 246.1722421875,
1254
+ "9": 261.254546875,
1255
+ "11": 291.024140625,
1256
+ "12": 294.8629609375,
1257
+ "13": 297.5098203125,
1258
+ "14": 300.940375,
1259
+ "15": 300.8659140625,
1260
+ "16": 302.8726171875,
1261
+ "17": 307.435828125,
1262
+ "18": 305.881828125,
1263
+ "19": 306.381546875,
1264
+ "20": 310.869171875
1265
+ },
1266
+ "per_op_loss": {},
1267
+ "per_template": {
1268
+ "crazy_zootopia": {
1269
+ "count": 809728,
1270
+ "correct": 326673,
1271
+ "answer_accuracy": 0.40343547462851725,
1272
+ "avg_response_len": 243.4344248932975,
1273
+ "resp_tokens_sum": 197115670,
1274
+ "pass_at_k": {
1275
+ "pass@1": 0.40343547462851725,
1276
+ "pass@2": 0.46300600835572403,
1277
+ "pass@4": 0.5211690059747147,
1278
+ "pass@8": 0.5776619983204903,
1279
+ "pass@16": 0.6303095820829564,
1280
+ "pass@32": 0.6763947073834303,
1281
+ "pass@64": 0.7152368285303596,
1282
+ "pass@128": 0.7492886500158078
1283
+ },
1284
+ "per_op_pass_at_k": {
1285
+ "10": {
1286
+ "pass@1": 0.5485342920353983,
1287
+ "pass@2": 0.6212945381274241,
1288
+ "pass@4": 0.6798927465372133,
1289
+ "pass@8": 0.7329713189633855,
1290
+ "pass@16": 0.7809185330850646,
1291
+ "pass@32": 0.8214210274584312,
1292
+ "pass@64": 0.8545478788781176,
1293
+ "pass@128": 0.8820058997050148
1294
+ },
1295
+ "2": {
1296
+ "pass@1": 0.9669976635514018,
1297
+ "pass@2": 0.9762200431721738,
1298
+ "pass@4": 0.9828518164201436,
1299
+ "pass@8": 0.988660768625104,
1300
+ "pass@16": 0.9928562993950985,
1301
+ "pass@32": 0.9949676562841662,
1302
+ "pass@64": 0.9961110638984689,
1303
+ "pass@128": 0.9968847352024922
1304
+ },
1305
+ "3": {
1306
+ "pass@1": 0.9523604740061162,
1307
+ "pass@2": 0.965793481663416,
1308
+ "pass@4": 0.9743357550489675,
1309
+ "pass@8": 0.9804021050148144,
1310
+ "pass@16": 0.9847239199729213,
1311
+ "pass@32": 0.9876441175361615,
1312
+ "pass@64": 0.9896933766236177,
1313
+ "pass@128": 0.9908256880733946
1314
+ },
1315
+ "4": {
1316
+ "pass@1": 0.7410400390625,
1317
+ "pass@2": 0.7776855468750001,
1318
+ "pass@4": 0.804752966816648,
1319
+ "pass@8": 0.8265181231781614,
1320
+ "pass@16": 0.8441561704150973,
1321
+ "pass@32": 0.8584107299081515,
1322
+ "pass@64": 0.8725424263973889,
1323
+ "pass@128": 0.890625
1324
+ },
1325
+ "5": {
1326
+ "pass@1": 0.5860615079365079,
1327
+ "pass@2": 0.6280289182602176,
1328
+ "pass@4": 0.6568151829235631,
1329
+ "pass@8": 0.6784618285484385,
1330
+ "pass@16": 0.6985429050145969,
1331
+ "pass@32": 0.7172269713114492,
1332
+ "pass@64": 0.7336378808087067,
1333
+ "pass@128": 0.7523809523809524
1334
+ },
1335
+ "6": {
1336
+ "pass@1": 0.6110357410179641,
1337
+ "pass@2": 0.6640256642227358,
1338
+ "pass@4": 0.7050190184872099,
1339
+ "pass@8": 0.7401048358206398,
1340
+ "pass@16": 0.7712992571383563,
1341
+ "pass@32": 0.7976646629926885,
1342
+ "pass@64": 0.8199997055863216,
1343
+ "pass@128": 0.8383233532934131
1344
+ },
1345
+ "7": {
1346
+ "pass@1": 0.5465051775147929,
1347
+ "pass@2": 0.597871919116619,
1348
+ "pass@4": 0.6383466158815945,
1349
+ "pass@8": 0.6704945279926731,
1350
+ "pass@16": 0.6982883012525174,
1351
+ "pass@32": 0.7252874533852414,
1352
+ "pass@64": 0.7523146294968283,
1353
+ "pass@128": 0.7810650887573964
1354
+ },
1355
+ "8": {
1356
+ "pass@1": 0.5412159455128205,
1357
+ "pass@2": 0.6090149751413282,
1358
+ "pass@4": 0.6620202931364347,
1359
+ "pass@8": 0.7088100090850898,
1360
+ "pass@16": 0.7524414216865002,
1361
+ "pass@32": 0.7903939600021158,
1362
+ "pass@64": 0.8223663408035892,
1363
+ "pass@128": 0.8557692307692307
1364
+ },
1365
+ "9": {
1366
+ "pass@1": 0.5105413732394366,
1367
+ "pass@2": 0.5978214761006982,
1368
+ "pass@4": 0.6684741905501246,
1369
+ "pass@8": 0.7256541935888985,
1370
+ "pass@16": 0.7724577522280122,
1371
+ "pass@32": 0.8116780331865703,
1372
+ "pass@64": 0.8476223526429457,
1373
+ "pass@128": 0.8845070422535212
1374
+ },
1375
+ "11": {
1376
+ "pass@1": 0.41977248583569404,
1377
+ "pass@2": 0.5152966836563977,
1378
+ "pass@4": 0.5988550168835131,
1379
+ "pass@8": 0.6740992829329473,
1380
+ "pass@16": 0.7422318681552861,
1381
+ "pass@32": 0.8019395682288849,
1382
+ "pass@64": 0.8536206281446311,
1383
+ "pass@128": 0.8980169971671388
1384
+ },
1385
+ "12": {
1386
+ "pass@1": 0.3938912429378531,
1387
+ "pass@2": 0.5104100632812844,
1388
+ "pass@4": 0.6080495266481519,
1389
+ "pass@8": 0.68796901487411,
1390
+ "pass@16": 0.7544555654168367,
1391
+ "pass@32": 0.8096660100272961,
1392
+ "pass@64": 0.8520104842456188,
1393
+ "pass@128": 0.884180790960452
1394
+ },
1395
+ "13": {
1396
+ "pass@1": 0.17650372257053293,
1397
+ "pass@2": 0.26551083911337076,
1398
+ "pass@4": 0.3665973341302558,
1399
+ "pass@8": 0.4710679392988577,
1400
+ "pass@16": 0.5713484839319661,
1401
+ "pass@32": 0.6590817812645573,
1402
+ "pass@64": 0.7304700839844447,
1403
+ "pass@128": 0.7899686520376176
1404
+ },
1405
+ "14": {
1406
+ "pass@1": 0.11207651869158879,
1407
+ "pass@2": 0.17378371292957537,
1408
+ "pass@4": 0.25215351060089447,
1409
+ "pass@8": 0.3400416779501758,
1410
+ "pass@16": 0.42831800765620487,
1411
+ "pass@32": 0.5089598246736358,
1412
+ "pass@64": 0.5781006993506586,
1413
+ "pass@128": 0.6386292834890965
1414
+ },
1415
+ "15": {
1416
+ "pass@1": 0.0994400289017341,
1417
+ "pass@2": 0.15031248577670558,
1418
+ "pass@4": 0.21217345971204465,
1419
+ "pass@8": 0.2830047929574361,
1420
+ "pass@16": 0.35940929952555856,
1421
+ "pass@32": 0.4361782265948589,
1422
+ "pass@64": 0.5084135452231018,
1423
+ "pass@128": 0.5751445086705202
1424
+ },
1425
+ "16": {
1426
+ "pass@1": 0.09947076612903226,
1427
+ "pass@2": 0.15187325374650743,
1428
+ "pass@4": 0.2179497756328846,
1429
+ "pass@8": 0.2926177158647718,
1430
+ "pass@16": 0.3690935358079865,
1431
+ "pass@32": 0.4399879447067403,
1432
+ "pass@64": 0.49926141465739254,
1433
+ "pass@128": 0.5451612903225806
1434
+ },
1435
+ "17": {
1436
+ "pass@1": 0.10357207369942197,
1437
+ "pass@2": 0.1580094215101725,
1438
+ "pass@4": 0.22143721648233278,
1439
+ "pass@8": 0.2883468654318178,
1440
+ "pass@16": 0.35307511467549146,
1441
+ "pass@32": 0.4110660972285916,
1442
+ "pass@64": 0.45948745382879763,
1443
+ "pass@128": 0.5
1444
+ },
1445
+ "18": {
1446
+ "pass@1": 0.11919642857142858,
1447
+ "pass@2": 0.17823186164229474,
1448
+ "pass@4": 0.24767612941239495,
1449
+ "pass@8": 0.32057904558995565,
1450
+ "pass@16": 0.3888781464313517,
1451
+ "pass@32": 0.44704524193215434,
1452
+ "pass@64": 0.4943156741255045,
1453
+ "pass@128": 0.5314285714285715
1454
+ },
1455
+ "19": {
1456
+ "pass@1": 0.08955048798798798,
1457
+ "pass@2": 0.14224620782199515,
1458
+ "pass@4": 0.21036005099587773,
1459
+ "pass@8": 0.2858482595544025,
1460
+ "pass@16": 0.3575037208007843,
1461
+ "pass@32": 0.4184082663584716,
1462
+ "pass@64": 0.4677929962465217,
1463
+ "pass@128": 0.5075075075075075
1464
+ },
1465
+ "20": {
1466
+ "pass@1": 0.09365615615615616,
1467
+ "pass@2": 0.14956861388948,
1468
+ "pass@4": 0.22208081000010135,
1469
+ "pass@8": 0.301411696925859,
1470
+ "pass@16": 0.37289719680717975,
1471
+ "pass@32": 0.4278229117854127,
1472
+ "pass@64": 0.4672031815887534,
1473
+ "pass@128": 0.5015015015015015
1474
+ }
1475
+ }
1476
+ },
1477
+ "teachers_in_school": {
1478
+ "count": 821120,
1479
+ "correct": 330878,
1480
+ "answer_accuracy": 0.4029593725643024,
1481
+ "avg_response_len": 251.9685551441933,
1482
+ "resp_tokens_sum": 206896420,
1483
+ "pass_at_k": {
1484
+ "pass@1": 0.4029593725643024,
1485
+ "pass@2": 0.4616005946937903,
1486
+ "pass@4": 0.5196187779187411,
1487
+ "pass@8": 0.5766667464711145,
1488
+ "pass@16": 0.6305522390571069,
1489
+ "pass@32": 0.6785554152113233,
1490
+ "pass@64": 0.7189738431394698,
1491
+ "pass@128": 0.751987529228371
1492
+ },
1493
+ "per_op_pass_at_k": {
1494
+ "10": {
1495
+ "pass@1": 0.5434240797546013,
1496
+ "pass@2": 0.6273198607555187,
1497
+ "pass@4": 0.6973762240609492,
1498
+ "pass@8": 0.7554670303495753,
1499
+ "pass@16": 0.8036698672416946,
1500
+ "pass@32": 0.8436957147979897,
1501
+ "pass@64": 0.877881304694098,
1502
+ "pass@128": 0.9049079754601227
1503
+ },
1504
+ "2": {
1505
+ "pass@1": 0.9736892097264438,
1506
+ "pass@2": 0.9847328758107362,
1507
+ "pass@4": 0.9901901966851409,
1508
+ "pass@8": 0.9927113374191188,
1509
+ "pass@16": 0.993648342971609,
1510
+ "pass@32": 0.9939055659835022,
1511
+ "pass@64": 0.9939209658267383,
1512
+ "pass@128": 0.993920972644377
1513
+ },
1514
+ "3": {
1515
+ "pass@1": 0.9402225378787878,
1516
+ "pass@2": 0.9562224111190645,
1517
+ "pass@4": 0.966675997602572,
1518
+ "pass@8": 0.9732372569438384,
1519
+ "pass@16": 0.9772623665321553,
1520
+ "pass@32": 0.979753273368313,
1521
+ "pass@64": 0.9813605854303431,
1522
+ "pass@128": 0.9818181818181818
1523
+ },
1524
+ "4": {
1525
+ "pass@1": 0.7226100221893491,
1526
+ "pass@2": 0.7654976005218281,
1527
+ "pass@4": 0.7946490924026508,
1528
+ "pass@8": 0.8150740090995962,
1529
+ "pass@16": 0.8315740698722591,
1530
+ "pass@32": 0.8478944811913116,
1531
+ "pass@64": 0.8650704580299757,
1532
+ "pass@128": 0.878698224852071
1533
+ },
1534
+ "5": {
1535
+ "pass@1": 0.6107789855072464,
1536
+ "pass@2": 0.6436579938377267,
1537
+ "pass@4": 0.670637732783402,
1538
+ "pass@8": 0.6953935801960935,
1539
+ "pass@16": 0.7181272383034838,
1540
+ "pass@32": 0.7399590031087759,
1541
+ "pass@64": 0.7619561041686341,
1542
+ "pass@128": 0.7855072463768116
1543
+ },
1544
+ "6": {
1545
+ "pass@1": 0.5464272922636103,
1546
+ "pass@2": 0.5969523441554045,
1547
+ "pass@4": 0.6346913039881474,
1548
+ "pass@8": 0.6666373773924721,
1549
+ "pass@16": 0.6975353973976225,
1550
+ "pass@32": 0.7284124145362817,
1551
+ "pass@64": 0.756892771403926,
1552
+ "pass@128": 0.7822349570200573
1553
+ },
1554
+ "7": {
1555
+ "pass@1": 0.5512779985549133,
1556
+ "pass@2": 0.5971500785125847,
1557
+ "pass@4": 0.6349748599885993,
1558
+ "pass@8": 0.6696253642996941,
1559
+ "pass@16": 0.7024347855609688,
1560
+ "pass@32": 0.7341433776297595,
1561
+ "pass@64": 0.7661180630503324,
1562
+ "pass@128": 0.8034682080924855
1563
+ },
1564
+ "8": {
1565
+ "pass@1": 0.5112676056338028,
1566
+ "pass@2": 0.5702083564378394,
1567
+ "pass@4": 0.6210338083619829,
1568
+ "pass@8": 0.6679787729679809,
1569
+ "pass@16": 0.7144538059651695,
1570
+ "pass@32": 0.7619468940389942,
1571
+ "pass@64": 0.8071898332537814,
1572
+ "pass@128": 0.847887323943662
1573
+ },
1574
+ "9": {
1575
+ "pass@1": 0.4994620134730539,
1576
+ "pass@2": 0.5715450251072656,
1577
+ "pass@4": 0.6303960414379337,
1578
+ "pass@8": 0.6812784051466028,
1579
+ "pass@16": 0.7264440386053771,
1580
+ "pass@32": 0.7696312332548112,
1581
+ "pass@64": 0.8130165993358162,
1582
+ "pass@128": 0.8562874251497006
1583
+ },
1584
+ "11": {
1585
+ "pass@1": 0.4587019954819277,
1586
+ "pass@2": 0.5552945047908173,
1587
+ "pass@4": 0.6409966999606971,
1588
+ "pass@8": 0.7148040088201414,
1589
+ "pass@16": 0.7767393045658387,
1590
+ "pass@32": 0.8297072308383238,
1591
+ "pass@64": 0.8746833656520376,
1592
+ "pass@128": 0.9126506024096386
1593
+ },
1594
+ "12": {
1595
+ "pass@1": 0.3535952194357367,
1596
+ "pass@2": 0.46705609927677505,
1597
+ "pass@4": 0.5703037878494028,
1598
+ "pass@8": 0.6629286050294428,
1599
+ "pass@16": 0.7478458624123885,
1600
+ "pass@32": 0.8187617469065239,
1601
+ "pass@64": 0.8679221722274657,
1602
+ "pass@128": 0.9028213166144201
1603
+ },
1604
+ "13": {
1605
+ "pass@1": 0.20017069327731093,
1606
+ "pass@2": 0.29245896193122883,
1607
+ "pass@4": 0.39196500384930894,
1608
+ "pass@8": 0.49107469341212867,
1609
+ "pass@16": 0.5849278755630151,
1610
+ "pass@32": 0.66909629091827,
1611
+ "pass@64": 0.7394157298961205,
1612
+ "pass@128": 0.7927170868347339
1613
+ },
1614
+ "14": {
1615
+ "pass@1": 0.12148209064327485,
1616
+ "pass@2": 0.18374897833494486,
1617
+ "pass@4": 0.2576632594719518,
1618
+ "pass@8": 0.33923709003618235,
1619
+ "pass@16": 0.42476920910139276,
1620
+ "pass@32": 0.5074564490761422,
1621
+ "pass@64": 0.5803711048868923,
1622
+ "pass@128": 0.6403508771929824
1623
+ },
1624
+ "15": {
1625
+ "pass@1": 0.09886363636363636,
1626
+ "pass@2": 0.15513675137198757,
1627
+ "pass@4": 0.22539836440899438,
1628
+ "pass@8": 0.3055976387358603,
1629
+ "pass@16": 0.38618518854570155,
1630
+ "pass@32": 0.45698552030349504,
1631
+ "pass@64": 0.5147245657272281,
1632
+ "pass@128": 0.5575757575757576
1633
+ },
1634
+ "16": {
1635
+ "pass@1": 0.10803952991452992,
1636
+ "pass@2": 0.15944531372680978,
1637
+ "pass@4": 0.22162470449740793,
1638
+ "pass@8": 0.29181218523663216,
1639
+ "pass@16": 0.36106886230415675,
1640
+ "pass@32": 0.42176417751644407,
1641
+ "pass@64": 0.47626160158795905,
1642
+ "pass@128": 0.5299145299145299
1643
+ },
1644
+ "17": {
1645
+ "pass@1": 0.12783870341614906,
1646
+ "pass@2": 0.19124290849513367,
1647
+ "pass@4": 0.2656787292116435,
1648
+ "pass@8": 0.3460024308333159,
1649
+ "pass@16": 0.42595468112596735,
1650
+ "pass@32": 0.4986195373338599,
1651
+ "pass@64": 0.5599232484124532,
1652
+ "pass@128": 0.6118012422360248
1653
+ },
1654
+ "18": {
1655
+ "pass@1": 0.11353400735294118,
1656
+ "pass@2": 0.17243262216303845,
1657
+ "pass@4": 0.2442672787225127,
1658
+ "pass@8": 0.3239806135278011,
1659
+ "pass@16": 0.39993769990663836,
1660
+ "pass@32": 0.4633629313271958,
1661
+ "pass@64": 0.5111777845425836,
1662
+ "pass@128": 0.5441176470588235
1663
+ },
1664
+ "19": {
1665
+ "pass@1": 0.10014415922619048,
1666
+ "pass@2": 0.1556139711051743,
1667
+ "pass@4": 0.22348070581579088,
1668
+ "pass@8": 0.29794332466542117,
1669
+ "pass@16": 0.37017448624899146,
1670
+ "pass@32": 0.43114841416760863,
1671
+ "pass@64": 0.4765716994613993,
1672
+ "pass@128": 0.5059523809523809
1673
+ },
1674
+ "20": {
1675
+ "pass@1": 0.09110684880239521,
1676
+ "pass@2": 0.14313720000942995,
1677
+ "pass@4": 0.21076069402252862,
1678
+ "pass@8": 0.28671310390061244,
1679
+ "pass@16": 0.3586706072592117,
1680
+ "pass@32": 0.4163128162045844,
1681
+ "pass@64": 0.4539123901453239,
1682
+ "pass@128": 0.47005988023952094
1683
+ }
1684
+ }
1685
+ },
1686
+ "movie_festival_awards": {
1687
+ "count": 801152,
1688
+ "correct": 324905,
1689
+ "answer_accuracy": 0.4055472619428024,
1690
+ "avg_response_len": 269.2765005891516,
1691
+ "resp_tokens_sum": 215731407,
1692
+ "pass_at_k": {
1693
+ "pass@1": 0.4055472619428024,
1694
+ "pass@2": 0.4655933644528339,
1695
+ "pass@4": 0.5247422408095355,
1696
+ "pass@8": 0.5828397968172885,
1697
+ "pass@16": 0.6377283837264522,
1698
+ "pass@32": 0.6864299272395259,
1699
+ "pass@64": 0.726849277721629,
1700
+ "pass@128": 0.7598657932577089
1701
+ },
1702
+ "per_op_pass_at_k": {
1703
+ "10": {
1704
+ "pass@1": 0.4935401119402985,
1705
+ "pass@2": 0.5670470237395698,
1706
+ "pass@4": 0.6367358320881525,
1707
+ "pass@8": 0.7042430815093631,
1708
+ "pass@16": 0.7680596090454958,
1709
+ "pass@32": 0.8246740921435434,
1710
+ "pass@64": 0.8697867463781164,
1711
+ "pass@128": 0.9044776119402985
1712
+ },
1713
+ "2": {
1714
+ "pass@1": 0.9596875,
1715
+ "pass@2": 0.9764113470191226,
1716
+ "pass@4": 0.9854286804328029,
1717
+ "pass@8": 0.991061509280679,
1718
+ "pass@16": 0.9942260504228976,
1719
+ "pass@32": 0.9956652788647895,
1720
+ "pass@64": 0.9971427773528576,
1721
+ "pass@128": 1.0
1722
+ },
1723
+ "3": {
1724
+ "pass@1": 0.9397093658892128,
1725
+ "pass@2": 0.9562255371777508,
1726
+ "pass@4": 0.9668405375166294,
1727
+ "pass@8": 0.9747146666399417,
1728
+ "pass@16": 0.9801107267765427,
1729
+ "pass@32": 0.9832217992096687,
1730
+ "pass@64": 0.984688404960452,
1731
+ "pass@128": 0.9854227405247813
1732
+ },
1733
+ "4": {
1734
+ "pass@1": 0.7007264254385965,
1735
+ "pass@2": 0.7445909310678271,
1736
+ "pass@4": 0.7808669256255251,
1737
+ "pass@8": 0.8129850113269721,
1738
+ "pass@16": 0.842129991729984,
1739
+ "pass@32": 0.8674656215356255,
1740
+ "pass@64": 0.8884742826156206,
1741
+ "pass@128": 0.9064327485380117
1742
+ },
1743
+ "5": {
1744
+ "pass@1": 0.5817784926470588,
1745
+ "pass@2": 0.6204282943492355,
1746
+ "pass@4": 0.6536405768212797,
1747
+ "pass@8": 0.68463428746883,
1748
+ "pass@16": 0.7157901961439006,
1749
+ "pass@32": 0.7474912605697788,
1750
+ "pass@64": 0.7777573616250464,
1751
+ "pass@128": 0.8029411764705883
1752
+ },
1753
+ "6": {
1754
+ "pass@1": 0.5934296135646687,
1755
+ "pass@2": 0.64594756762463,
1756
+ "pass@4": 0.6914165587345744,
1757
+ "pass@8": 0.7364669504842055,
1758
+ "pass@16": 0.778883156850051,
1759
+ "pass@32": 0.8167598694063396,
1760
+ "pass@64": 0.8509370624044789,
1761
+ "pass@128": 0.8801261829652997
1762
+ },
1763
+ "7": {
1764
+ "pass@1": 0.5463558148734177,
1765
+ "pass@2": 0.6027772413535332,
1766
+ "pass@4": 0.6477394363481457,
1767
+ "pass@8": 0.6858369916354995,
1768
+ "pass@16": 0.7205188747595422,
1769
+ "pass@32": 0.7560440209344339,
1770
+ "pass@64": 0.7912034540169977,
1771
+ "pass@128": 0.8259493670886076
1772
+ },
1773
+ "8": {
1774
+ "pass@1": 0.5558370870870871,
1775
+ "pass@2": 0.6316111879596129,
1776
+ "pass@4": 0.6897773594742097,
1777
+ "pass@8": 0.736390876197389,
1778
+ "pass@16": 0.7772959628816623,
1779
+ "pass@32": 0.8148386053479606,
1780
+ "pass@64": 0.8483111557392278,
1781
+ "pass@128": 0.8738738738738738
1782
+ },
1783
+ "9": {
1784
+ "pass@1": 0.5287630627009646,
1785
+ "pass@2": 0.6127898163151629,
1786
+ "pass@4": 0.680057331961318,
1787
+ "pass@8": 0.7337678523669866,
1788
+ "pass@16": 0.7759801527561564,
1789
+ "pass@32": 0.8104190136847873,
1790
+ "pass@64": 0.841157580846551,
1791
+ "pass@128": 0.8713826366559485
1792
+ },
1793
+ "11": {
1794
+ "pass@1": 0.4311011904761905,
1795
+ "pass@2": 0.5327611392325959,
1796
+ "pass@4": 0.6226473205730233,
1797
+ "pass@8": 0.699414827225559,
1798
+ "pass@16": 0.7632510144163813,
1799
+ "pass@32": 0.8149524603297097,
1800
+ "pass@64": 0.8541162619127182,
1801
+ "pass@128": 0.8825396825396825
1802
+ },
1803
+ "12": {
1804
+ "pass@1": 0.37994552752293576,
1805
+ "pass@2": 0.49269373510077263,
1806
+ "pass@4": 0.5890112603470439,
1807
+ "pass@8": 0.673374303865433,
1808
+ "pass@16": 0.7488352151511932,
1809
+ "pass@32": 0.8150540013464985,
1810
+ "pass@64": 0.8687804600398654,
1811
+ "pass@128": 0.9051987767584098
1812
+ },
1813
+ "13": {
1814
+ "pass@1": 0.18453414351851852,
1815
+ "pass@2": 0.26992616591328855,
1816
+ "pass@4": 0.36343596575891,
1817
+ "pass@8": 0.45878259248561787,
1818
+ "pass@16": 0.5525191150433114,
1819
+ "pass@32": 0.6384739274316215,
1820
+ "pass@64": 0.709316108372,
1821
+ "pass@128": 0.7685185185185185
1822
+ },
1823
+ "14": {
1824
+ "pass@1": 0.12685459940652818,
1825
+ "pass@2": 0.18879712434869975,
1826
+ "pass@4": 0.2624856972626196,
1827
+ "pass@8": 0.3443412391690488,
1828
+ "pass@16": 0.4263202019821751,
1829
+ "pass@32": 0.5005275143900397,
1830
+ "pass@64": 0.5643824971967146,
1831
+ "pass@128": 0.6201780415430267
1832
+ },
1833
+ "15": {
1834
+ "pass@1": 0.10570987654320987,
1835
+ "pass@2": 0.16320697543015447,
1836
+ "pass@4": 0.23749303125303775,
1837
+ "pass@8": 0.32504223382366393,
1838
+ "pass@16": 0.41559522972860385,
1839
+ "pass@32": 0.49623847142737515,
1840
+ "pass@64": 0.5606971462894773,
1841
+ "pass@128": 0.6141975308641975
1842
+ },
1843
+ "16": {
1844
+ "pass@1": 0.10893713126843659,
1845
+ "pass@2": 0.1652998919935893,
1846
+ "pass@4": 0.23209237285604775,
1847
+ "pass@8": 0.3033054579667241,
1848
+ "pass@16": 0.3727492867250306,
1849
+ "pass@32": 0.43483766239548877,
1850
+ "pass@64": 0.48846237984478985,
1851
+ "pass@128": 0.5309734513274337
1852
+ },
1853
+ "17": {
1854
+ "pass@1": 0.11794051204819277,
1855
+ "pass@2": 0.17528430651740828,
1856
+ "pass@4": 0.24250910644829024,
1857
+ "pass@8": 0.31225596159268304,
1858
+ "pass@16": 0.37914528287600097,
1859
+ "pass@32": 0.44015855587538144,
1860
+ "pass@64": 0.49245703153300757,
1861
+ "pass@128": 0.536144578313253
1862
+ },
1863
+ "18": {
1864
+ "pass@1": 0.08644153225806452,
1865
+ "pass@2": 0.1351381921513843,
1866
+ "pass@4": 0.19537179505787589,
1867
+ "pass@8": 0.2640087960618504,
1868
+ "pass@16": 0.3364805754908361,
1869
+ "pass@32": 0.40511059898611057,
1870
+ "pass@64": 0.4606336256185126,
1871
+ "pass@128": 0.5
1872
+ },
1873
+ "19": {
1874
+ "pass@1": 0.08657477341389729,
1875
+ "pass@2": 0.13416194364488426,
1876
+ "pass@4": 0.19288663416695273,
1877
+ "pass@8": 0.25911399140930275,
1878
+ "pass@16": 0.32694205872241794,
1879
+ "pass@32": 0.3880784900568889,
1880
+ "pass@64": 0.43832121607177427,
1881
+ "pass@128": 0.48338368580060426
1882
+ },
1883
+ "20": {
1884
+ "pass@1": 0.11606137387387387,
1885
+ "pass@2": 0.17806019602279444,
1886
+ "pass@4": 0.2542757329883313,
1887
+ "pass@8": 0.33586454124686205,
1888
+ "pass@16": 0.41006388289722623,
1889
+ "pass@32": 0.4660912695451911,
1890
+ "pass@64": 0.5023665080660694,
1891
+ "pass@128": 0.5285285285285285
1892
+ }
1893
+ }
1894
+ }
1895
+ }
1896
+ }
1897
+ }