ZachW commited on
Commit
c48e6e0
·
verified ·
1 Parent(s): 96989ab

Upload extrapolation artifact: 050302 RL OP15-20 step200 metrics

Browse files
050302/evals/rl050302_op15_20_step200_merged_hf/rl050302_op15_20_step200_merged_hf_metrics.json ADDED
@@ -0,0 +1,1897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "rl050302_op15_20_step200_merged_hf",
3
+ "id": {
4
+ "answer_accuracy": 0.7198645833333334,
5
+ "avg_response_len": 203.807328125,
6
+ "avg_loss": 0.0,
7
+ "count": 1152000,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {
10
+ "pass@1": 0.7198645833333334,
11
+ "pass@2": 0.7771044537401421,
12
+ "pass@4": 0.8192827575719673,
13
+ "pass@8": 0.8497876776252181,
14
+ "pass@16": 0.8719250396099119,
15
+ "pass@32": 0.8890832813618992,
16
+ "pass@64": 0.9034296326095764,
17
+ "pass@128": 0.9163333333333333
18
+ },
19
+ "per_op_pass_at_k": {
20
+ "10": {
21
+ "pass@1": 0.6137109375,
22
+ "pass@2": 0.6928655265748027,
23
+ "pass@4": 0.7514061629171355,
24
+ "pass@8": 0.7965425200019911,
25
+ "pass@16": 0.8332050121400162,
26
+ "pass@32": 0.863816716890956,
27
+ "pass@64": 0.8889742483433577,
28
+ "pass@128": 0.91
29
+ },
30
+ "2": {
31
+ "pass@1": 0.92753125,
32
+ "pass@2": 0.9454394685039371,
33
+ "pass@4": 0.9582661204536929,
34
+ "pass@8": 0.9675005822154368,
35
+ "pass@16": 0.9738152511151082,
36
+ "pass@32": 0.978352349440396,
37
+ "pass@64": 0.982220783659935,
38
+ "pass@128": 0.986
39
+ },
40
+ "3": {
41
+ "pass@1": 0.87525,
42
+ "pass@2": 0.8933171751968509,
43
+ "pass@4": 0.903821142950881,
44
+ "pass@8": 0.9113115359504261,
45
+ "pass@16": 0.9173621733495384,
46
+ "pass@32": 0.9227675643594974,
47
+ "pass@64": 0.9282661218073456,
48
+ "pass@128": 0.934
49
+ },
50
+ "4": {
51
+ "pass@1": 0.7896640625,
52
+ "pass@2": 0.8281980807086614,
53
+ "pass@4": 0.8554755267154103,
54
+ "pass@8": 0.8741056836959573,
55
+ "pass@16": 0.8877460253022831,
56
+ "pass@32": 0.8997405892622435,
57
+ "pass@64": 0.911213861081669,
58
+ "pass@128": 0.923
59
+ },
60
+ "5": {
61
+ "pass@1": 0.698640625,
62
+ "pass@2": 0.7646503444881888,
63
+ "pass@4": 0.8178164796587929,
64
+ "pass@8": 0.8575505477175563,
65
+ "pass@16": 0.8860211077579768,
66
+ "pass@32": 0.906900117552899,
67
+ "pass@64": 0.9237245923609523,
68
+ "pass@128": 0.94
69
+ },
70
+ "6": {
71
+ "pass@1": 0.672421875,
72
+ "pass@2": 0.7338943159448824,
73
+ "pass@4": 0.7815458341769774,
74
+ "pass@8": 0.8147635720348121,
75
+ "pass@16": 0.8370253152434817,
76
+ "pass@32": 0.8540255270767936,
77
+ "pass@64": 0.8681434397988514,
78
+ "pass@128": 0.879
79
+ },
80
+ "7": {
81
+ "pass@1": 0.66359375,
82
+ "pass@2": 0.7303299704724414,
83
+ "pass@4": 0.7794709802212226,
84
+ "pass@8": 0.8147819203326597,
85
+ "pass@16": 0.8397578705147839,
86
+ "pass@32": 0.8577877159512644,
87
+ "pass@64": 0.872366823689421,
88
+ "pass@128": 0.885
89
+ },
90
+ "8": {
91
+ "pass@1": 0.6255859375,
92
+ "pass@2": 0.7037237942913391,
93
+ "pass@4": 0.7589305926134238,
94
+ "pass@8": 0.7973030953082081,
95
+ "pass@16": 0.8247642585556798,
96
+ "pass@32": 0.8463344723730156,
97
+ "pass@64": 0.8662206995650144,
98
+ "pass@128": 0.888
99
+ },
100
+ "9": {
101
+ "pass@1": 0.6123828125,
102
+ "pass@2": 0.7015214074803167,
103
+ "pass@4": 0.7668119784401946,
104
+ "pass@8": 0.8142296413699427,
105
+ "pass@16": 0.8476283425102903,
106
+ "pass@32": 0.8720244793500149,
107
+ "pass@64": 0.8897361231796059,
108
+ "pass@128": 0.902
109
+ }
110
+ },
111
+ "per_op_accuracy": {
112
+ "10": 0.6137109375,
113
+ "2": 0.92753125,
114
+ "3": 0.87525,
115
+ "4": 0.7896640625,
116
+ "5": 0.698640625,
117
+ "6": 0.672421875,
118
+ "7": 0.66359375,
119
+ "8": 0.6255859375,
120
+ "9": 0.6123828125
121
+ },
122
+ "per_op_avg_response_len": {
123
+ "10": 281.0300546875,
124
+ "2": 123.2331015625,
125
+ "3": 148.998984375,
126
+ "4": 162.370984375,
127
+ "5": 179.260140625,
128
+ "6": 203.594375,
129
+ "7": 226.4866328125,
130
+ "8": 247.12684375,
131
+ "9": 262.1648359375
132
+ },
133
+ "per_op_avg_loss": {},
134
+ "per_op_length": {
135
+ "10": 281.0300546875,
136
+ "2": 123.2331015625,
137
+ "3": 148.998984375,
138
+ "4": 162.370984375,
139
+ "5": 179.260140625,
140
+ "6": 203.594375,
141
+ "7": 226.4866328125,
142
+ "8": 247.12684375,
143
+ "9": 262.1648359375
144
+ },
145
+ "per_op_loss": {},
146
+ "per_template": {
147
+ "crazy_zootopia": {
148
+ "count": 379008,
149
+ "correct": 272932,
150
+ "answer_accuracy": 0.7201220027017899,
151
+ "avg_response_len": 194.61431684819317,
152
+ "resp_tokens_sum": 73760383,
153
+ "pass_at_k": {
154
+ "pass@1": 0.7201220027017899,
155
+ "pass@2": 0.7788133032573089,
156
+ "pass@4": 0.8215033461812726,
157
+ "pass@8": 0.8524991320069447,
158
+ "pass@16": 0.8754489236011029,
159
+ "pass@32": 0.8936120545406552,
160
+ "pass@64": 0.908694780640936,
161
+ "pass@128": 0.9223235393448159
162
+ },
163
+ "per_op_pass_at_k": {
164
+ "10": {
165
+ "pass@1": 0.5904775073746312,
166
+ "pass@2": 0.6759274179499686,
167
+ "pass@4": 0.739327368647503,
168
+ "pass@8": 0.7888088892182306,
169
+ "pass@16": 0.8301737990227551,
170
+ "pass@32": 0.8653796021525334,
171
+ "pass@64": 0.8950191494801747,
172
+ "pass@128": 0.9203539823008849
173
+ },
174
+ "2": {
175
+ "pass@1": 0.9306853582554517,
176
+ "pass@2": 0.951853291264994,
177
+ "pass@4": 0.9672763547664015,
178
+ "pass@8": 0.9777387432602255,
179
+ "pass@16": 0.9849406600666408,
180
+ "pass@32": 0.9906709430385632,
181
+ "pass@64": 0.9946393301197776,
182
+ "pass@128": 0.9968847352024922
183
+ },
184
+ "3": {
185
+ "pass@1": 0.8729692278287462,
186
+ "pass@2": 0.8916826193744132,
187
+ "pass@4": 0.9012887589166031,
188
+ "pass@8": 0.9069931487274436,
189
+ "pass@16": 0.9113667010227782,
190
+ "pass@32": 0.9158878607999068,
191
+ "pass@64": 0.921283309008342,
192
+ "pass@128": 0.926605504587156
193
+ },
194
+ "4": {
195
+ "pass@1": 0.79365234375,
196
+ "pass@2": 0.8259854053887796,
197
+ "pass@4": 0.8509953552680912,
198
+ "pass@8": 0.8705874877072775,
199
+ "pass@16": 0.8861866677459165,
200
+ "pass@32": 0.9001222965392037,
201
+ "pass@64": 0.9150975754710673,
202
+ "pass@128": 0.934375
203
+ },
204
+ "5": {
205
+ "pass@1": 0.711954365079365,
206
+ "pass@2": 0.7870863798275217,
207
+ "pass@4": 0.8438272906362893,
208
+ "pass@8": 0.8820148031784064,
209
+ "pass@16": 0.9061775402294971,
210
+ "pass@32": 0.9233827284096477,
211
+ "pass@64": 0.939203360121733,
212
+ "pass@128": 0.9587301587301588
213
+ },
214
+ "6": {
215
+ "pass@1": 0.6886695359281437,
216
+ "pass@2": 0.7508096503842708,
217
+ "pass@4": 0.7981536092606685,
218
+ "pass@8": 0.8300540071307995,
219
+ "pass@16": 0.8515233346236275,
220
+ "pass@32": 0.867618886093707,
221
+ "pass@64": 0.8782963968718454,
222
+ "pass@128": 0.8832335329341318
223
+ },
224
+ "7": {
225
+ "pass@1": 0.6789478550295858,
226
+ "pass@2": 0.7444577587010199,
227
+ "pass@4": 0.7924073042090151,
228
+ "pass@8": 0.8294529774774302,
229
+ "pass@16": 0.85758601479046,
230
+ "pass@32": 0.8766141160571631,
231
+ "pass@64": 0.8892577006853214,
232
+ "pass@128": 0.8994082840236687
233
+ },
234
+ "8": {
235
+ "pass@1": 0.6259515224358975,
236
+ "pass@2": 0.6997999949525543,
237
+ "pass@4": 0.7484703330352931,
238
+ "pass@8": 0.7817076118380484,
239
+ "pass@16": 0.8062330519425672,
240
+ "pass@32": 0.8273824576064478,
241
+ "pass@64": 0.8497525386355358,
242
+ "pass@128": 0.875
243
+ },
244
+ "9": {
245
+ "pass@1": 0.6052596830985916,
246
+ "pass@2": 0.6952658866585334,
247
+ "pass@4": 0.7621365283564899,
248
+ "pass@8": 0.8126598351468055,
249
+ "pass@16": 0.8499978703599235,
250
+ "pass@32": 0.8788524080867729,
251
+ "pass@64": 0.8985118669285895,
252
+ "pass@128": 0.9098591549295775
253
+ }
254
+ }
255
+ },
256
+ "teachers_in_school": {
257
+ "count": 390656,
258
+ "correct": 282545,
259
+ "answer_accuracy": 0.7232578022608126,
260
+ "avg_response_len": 201.10431684141545,
261
+ "resp_tokens_sum": 78562608,
262
+ "pass_at_k": {
263
+ "pass@1": 0.7232578022608126,
264
+ "pass@2": 0.7784499450470092,
265
+ "pass@4": 0.818904463950937,
266
+ "pass@8": 0.8477502623120604,
267
+ "pass@16": 0.8682079868323207,
268
+ "pass@32": 0.8840742985733938,
269
+ "pass@64": 0.897939053516026,
270
+ "pass@128": 0.9108781127129751
271
+ },
272
+ "per_op_pass_at_k": {
273
+ "10": {
274
+ "pass@1": 0.6425421779141104,
275
+ "pass@2": 0.7180774781411522,
276
+ "pass@4": 0.7713729633489064,
277
+ "pass@8": 0.8115314019701301,
278
+ "pass@16": 0.8422782157793477,
279
+ "pass@32": 0.8665394346696862,
280
+ "pass@64": 0.8875838773290864,
281
+ "pass@128": 0.9079754601226994
282
+ },
283
+ "2": {
284
+ "pass@1": 0.9346267097264438,
285
+ "pass@2": 0.9484187648086544,
286
+ "pass@4": 0.9577753805660308,
287
+ "pass@8": 0.9642355108332087,
288
+ "pass@16": 0.9687406308251917,
289
+ "pass@32": 0.9722132661616432,
290
+ "pass@64": 0.976268636177634,
291
+ "pass@128": 0.9817629179331308
292
+ },
293
+ "3": {
294
+ "pass@1": 0.8785274621212121,
295
+ "pass@2": 0.8940225334049151,
296
+ "pass@4": 0.9043032615241278,
297
+ "pass@8": 0.9116606528207251,
298
+ "pass@16": 0.9163981804252643,
299
+ "pass@32": 0.919534935837756,
300
+ "pass@64": 0.9226355633253187,
301
+ "pass@128": 0.9272727272727272
302
+ },
303
+ "4": {
304
+ "pass@1": 0.791350776627219,
305
+ "pass@2": 0.8298940327540418,
306
+ "pass@4": 0.8566257274423535,
307
+ "pass@8": 0.8740505626426824,
308
+ "pass@16": 0.8863419432111916,
309
+ "pass@32": 0.8973476695044501,
310
+ "pass@64": 0.9076491744197636,
311
+ "pass@128": 0.9171597633136095
312
+ },
313
+ "5": {
314
+ "pass@1": 0.7036231884057971,
315
+ "pass@2": 0.760960144927536,
316
+ "pass@4": 0.8089746806105759,
317
+ "pass@8": 0.848232289495532,
318
+ "pass@16": 0.8788940387283235,
319
+ "pass@32": 0.903333779760882,
320
+ "pass@64": 0.9245269760364313,
321
+ "pass@128": 0.9449275362318841
322
+ },
323
+ "6": {
324
+ "pass@1": 0.6609733166189111,
325
+ "pass@2": 0.7211168580646615,
326
+ "pass@4": 0.7662969808752413,
327
+ "pass@8": 0.7963462083773434,
328
+ "pass@16": 0.8159017311926985,
329
+ "pass@32": 0.8326346080497815,
330
+ "pass@64": 0.8482729333475885,
331
+ "pass@128": 0.8595988538681948
332
+ },
333
+ "7": {
334
+ "pass@1": 0.669549313583815,
335
+ "pass@2": 0.7348724597196304,
336
+ "pass@4": 0.7796674157601687,
337
+ "pass@8": 0.8099202435632497,
338
+ "pass@16": 0.8312183082099309,
339
+ "pass@32": 0.8471171620508142,
340
+ "pass@64": 0.8607780565256175,
341
+ "pass@128": 0.8728323699421965
342
+ },
343
+ "8": {
344
+ "pass@1": 0.6231073943661972,
345
+ "pass@2": 0.7033488826660749,
346
+ "pass@4": 0.7635277449473741,
347
+ "pass@8": 0.8050630470213901,
348
+ "pass@16": 0.8335760394037561,
349
+ "pass@32": 0.8552420543936154,
350
+ "pass@64": 0.8752369946603544,
351
+ "pass@128": 0.8985915492957747
352
+ },
353
+ "9": {
354
+ "pass@1": 0.6189651946107785,
355
+ "pass@2": 0.7066431850865202,
356
+ "pass@4": 0.7706885601687509,
357
+ "pass@8": 0.8163749170683283,
358
+ "pass@16": 0.8472702317498962,
359
+ "pass@32": 0.8686874199142173,
360
+ "pass@64": 0.883720682738893,
361
+ "pass@128": 0.8922155688622755
362
+ }
363
+ }
364
+ },
365
+ "movie_festival_awards": {
366
+ "count": 382336,
367
+ "correct": 273807,
368
+ "answer_accuracy": 0.7161423459993305,
369
+ "avg_response_len": 215.68215130147306,
370
+ "resp_tokens_sum": 82463051,
371
+ "pass_at_k": {
372
+ "pass@1": 0.7161423459993305,
373
+ "pass@2": 0.7740357082132844,
374
+ "pass@4": 0.8174680234773142,
375
+ "pass@8": 0.8491815762229757,
376
+ "pass@16": 0.8722297682939683,
377
+ "pass@32": 0.8897119114550385,
378
+ "pass@64": 0.9038203738458033,
379
+ "pass@128": 0.9159691998660864
380
+ },
381
+ "per_op_pass_at_k": {
382
+ "10": {
383
+ "pass@1": 0.6091651119402985,
384
+ "pass@2": 0.6854712657186507,
385
+ "pass@4": 0.7441988026683222,
386
+ "pass@8": 0.7897822970589528,
387
+ "pass@16": 0.8274429729171232,
388
+ "pass@32": 0.8595856001759086,
389
+ "pass@64": 0.8842101870456582,
390
+ "pass@128": 0.9014925373134328
391
+ },
392
+ "2": {
393
+ "pass@1": 0.91796875,
394
+ "pass@2": 0.9367565382452198,
395
+ "pass@4": 0.9504637439070117,
396
+ "pass@8": 0.9611798930422266,
397
+ "pass@16": 0.9683818048349383,
398
+ "pass@32": 0.9728251775938187,
399
+ "pass@64": 0.9764262211115565,
400
+ "pass@128": 0.98
401
+ },
402
+ "3": {
403
+ "pass@1": 0.874271137026239,
404
+ "pass@2": 0.8941968590023187,
405
+ "pass@4": 0.9057715524262092,
406
+ "pass@8": 0.915092597334439,
407
+ "pass@16": 0.9240054302470929,
408
+ "pass@32": 0.9324364578760008,
409
+ "pass@64": 0.9403403611202984,
410
+ "pass@128": 0.9475218658892128
411
+ },
412
+ "4": {
413
+ "pass@1": 0.784265350877193,
414
+ "pass@2": 0.8285923038403095,
415
+ "pass@4": 0.8585307519125894,
416
+ "pass@8": 0.8774520393462032,
417
+ "pass@16": 0.8905927333865701,
418
+ "pass@32": 0.9017483686467666,
419
+ "pass@64": 0.9111029706931209,
420
+ "pass@128": 0.9181286549707602
421
+ },
422
+ "5": {
423
+ "pass@1": 0.68125,
424
+ "pass@2": 0.7476084848309403,
425
+ "pass@4": 0.8026900538168023,
426
+ "pass@8": 0.8443404260011751,
427
+ "pass@16": 0.8745786447776861,
428
+ "pass@32": 0.8952482473128105,
429
+ "pass@64": 0.9085697858530525,
430
+ "pass@128": 0.9176470588235294
431
+ },
432
+ "6": {
433
+ "pass@1": 0.6679071372239748,
434
+ "pass@2": 0.7301391459052637,
435
+ "pass@4": 0.7808355910361361,
436
+ "pass@8": 0.818929674856254,
437
+ "pass@16": 0.8450057327852958,
438
+ "pass@32": 0.8632534413631587,
439
+ "pass@64": 0.8793223643701782,
440
+ "pass@128": 0.8958990536277602
441
+ },
442
+ "7": {
443
+ "pass@1": 0.6406497231012658,
444
+ "pass@2": 0.71024486382438,
445
+ "pass@4": 0.7654189415998627,
446
+ "pass@8": 0.8044126888367252,
447
+ "pass@16": 0.8300388065663724,
448
+ "pass@32": 0.8493341982733007,
449
+ "pass@64": 0.8669889661389816,
450
+ "pass@128": 0.8829113924050633
451
+ },
452
+ "8": {
453
+ "pass@1": 0.6278856981981982,
454
+ "pass@2": 0.7077998273864413,
455
+ "pass@4": 0.7638303280783592,
456
+ "pass@8": 0.8036424586250549,
457
+ "pass@16": 0.8327328899737727,
458
+ "pass@32": 0.8545952441143289,
459
+ "pass@64": 0.8720383615804834,
460
+ "pass@128": 0.8888888888888888
461
+ },
462
+ "9": {
463
+ "pass@1": 0.6134445337620579,
464
+ "pass@2": 0.7031613951692531,
465
+ "pass@4": 0.7679856327243657,
466
+ "pass@8": 0.8137176128360283,
467
+ "pass@16": 0.8453081708297483,
468
+ "pass@32": 0.8678143930156336,
469
+ "pass@64": 0.8861791073478019,
470
+ "pass@128": 0.9035369774919614
471
+ }
472
+ }
473
+ }
474
+ }
475
+ },
476
+ "ood": {
477
+ "answer_accuracy": 0.29179921875,
478
+ "avg_response_len": 332.50551015625,
479
+ "avg_loss": 0.0,
480
+ "count": 1280000,
481
+ "validation_metrics": {},
482
+ "pass_at_k": {
483
+ "pass@1": 0.29179921875,
484
+ "pass@2": 0.3952800196850465,
485
+ "pass@4": 0.49449603572365797,
486
+ "pass@8": 0.5800142683512608,
487
+ "pass@16": 0.6497950549933502,
488
+ "pass@32": 0.7067447065449153,
489
+ "pass@64": 0.7535298202183753,
490
+ "pass@128": 0.7923
491
+ },
492
+ "per_op_pass_at_k": {
493
+ "11": {
494
+ "pass@1": 0.5468203125,
495
+ "pass@2": 0.648342027559055,
496
+ "pass@4": 0.7251446963817036,
497
+ "pass@8": 0.7825192329955811,
498
+ "pass@16": 0.8260392111057165,
499
+ "pass@32": 0.8579424969970187,
500
+ "pass@64": 0.8808024280651586,
501
+ "pass@128": 0.899
502
+ },
503
+ "12": {
504
+ "pass@1": 0.5172734375,
505
+ "pass@2": 0.6321119586614163,
506
+ "pass@4": 0.718914105455568,
507
+ "pass@8": 0.7837164778935699,
508
+ "pass@16": 0.831747640474528,
509
+ "pass@32": 0.867396036690709,
510
+ "pass@64": 0.8938193211530763,
511
+ "pass@128": 0.912
512
+ },
513
+ "13": {
514
+ "pass@1": 0.40490625,
515
+ "pass@2": 0.5382737450787406,
516
+ "pass@4": 0.6505229554743155,
517
+ "pass@8": 0.7362787088452506,
518
+ "pass@16": 0.8000323808896687,
519
+ "pass@32": 0.84706737223152,
520
+ "pass@64": 0.8816356236418902,
521
+ "pass@128": 0.908
522
+ },
523
+ "14": {
524
+ "pass@1": 0.33403125,
525
+ "pass@2": 0.4654560777559057,
526
+ "pass@4": 0.5895357703412076,
527
+ "pass@8": 0.6903756004044173,
528
+ "pass@16": 0.7653717718563208,
529
+ "pass@32": 0.8203046486849622,
530
+ "pass@64": 0.8601269996361208,
531
+ "pass@128": 0.89
532
+ },
533
+ "15": {
534
+ "pass@1": 0.2540703125,
535
+ "pass@2": 0.37107271161417366,
536
+ "pass@4": 0.4906143659542563,
537
+ "pass@8": 0.5954665228777352,
538
+ "pass@16": 0.6796684614746239,
539
+ "pass@32": 0.7467914324913115,
540
+ "pass@64": 0.7994782479135412,
541
+ "pass@128": 0.838
542
+ },
543
+ "16": {
544
+ "pass@1": 0.2188359375,
545
+ "pass@2": 0.32689283956692883,
546
+ "pass@4": 0.4437324613798277,
547
+ "pass@8": 0.5529708859896567,
548
+ "pass@16": 0.6443344673707929,
549
+ "pass@32": 0.7174125768438164,
550
+ "pass@64": 0.7771474691103983,
551
+ "pass@128": 0.827
552
+ },
553
+ "17": {
554
+ "pass@1": 0.1809140625,
555
+ "pass@2": 0.27425541338582665,
556
+ "pass@4": 0.3767013890138727,
557
+ "pass@8": 0.4734960870046542,
558
+ "pass@16": 0.5576667050582922,
559
+ "pass@32": 0.629289835766627,
560
+ "pass@64": 0.689218158435304,
561
+ "pass@128": 0.739
562
+ },
563
+ "18": {
564
+ "pass@1": 0.159875,
565
+ "pass@2": 0.24176119586614178,
566
+ "pass@4": 0.331536096550431,
567
+ "pass@8": 0.4179432667184478,
568
+ "pass@16": 0.49631175901573166,
569
+ "pass@32": 0.5675651014563933,
570
+ "pass@64": 0.6314278423785766,
571
+ "pass@128": 0.688
572
+ },
573
+ "19": {
574
+ "pass@1": 0.155921875,
575
+ "pass@2": 0.23286380413385835,
576
+ "pass@4": 0.31412745463067143,
577
+ "pass@8": 0.3882530608677067,
578
+ "pass@16": 0.45322203345971057,
579
+ "pass@32": 0.5123397833251867,
580
+ "pass@64": 0.5666400236612369,
581
+ "pass@128": 0.617
582
+ },
583
+ "20": {
584
+ "pass@1": 0.14534375,
585
+ "pass@2": 0.22177042322834656,
586
+ "pass@4": 0.3041310620547426,
587
+ "pass@8": 0.37912283991548595,
588
+ "pass@16": 0.44355611922790866,
589
+ "pass@32": 0.5013377809615996,
590
+ "pass@64": 0.5550020881883593,
591
+ "pass@128": 0.605
592
+ }
593
+ },
594
+ "per_op_accuracy": {
595
+ "11": 0.5468203125,
596
+ "12": 0.5172734375,
597
+ "13": 0.40490625,
598
+ "14": 0.33403125,
599
+ "15": 0.2540703125,
600
+ "16": 0.2188359375,
601
+ "17": 0.1809140625,
602
+ "18": 0.159875,
603
+ "19": 0.155921875,
604
+ "20": 0.14534375
605
+ },
606
+ "per_op_avg_response_len": {
607
+ "11": 296.6561015625,
608
+ "12": 303.042890625,
609
+ "13": 315.422875,
610
+ "14": 333.3207109375,
611
+ "15": 338.5395546875,
612
+ "16": 343.201203125,
613
+ "17": 346.66403125,
614
+ "18": 348.8883203125,
615
+ "19": 346.0546640625,
616
+ "20": 353.26475
617
+ },
618
+ "per_op_avg_loss": {},
619
+ "per_op_length": {
620
+ "11": 296.6561015625,
621
+ "12": 303.042890625,
622
+ "13": 315.422875,
623
+ "14": 333.3207109375,
624
+ "15": 338.5395546875,
625
+ "16": 343.201203125,
626
+ "17": 346.66403125,
627
+ "18": 348.8883203125,
628
+ "19": 346.0546640625,
629
+ "20": 353.26475
630
+ },
631
+ "per_op_loss": {},
632
+ "per_template": {
633
+ "crazy_zootopia": {
634
+ "count": 430720,
635
+ "correct": 127184,
636
+ "answer_accuracy": 0.29528231797919763,
637
+ "avg_response_len": 318.2857680163447,
638
+ "resp_tokens_sum": 137092046,
639
+ "pass_at_k": {
640
+ "pass@1": 0.29528231797919763,
641
+ "pass@2": 0.4018862391922413,
642
+ "pass@4": 0.5033818059703934,
643
+ "pass@8": 0.589573539916414,
644
+ "pass@16": 0.6599079704112036,
645
+ "pass@32": 0.718119954486833,
646
+ "pass@64": 0.7665420924306448,
647
+ "pass@128": 0.8074294205052006
648
+ },
649
+ "per_op_pass_at_k": {
650
+ "11": {
651
+ "pass@1": 0.5223309135977338,
652
+ "pass@2": 0.6289980008253216,
653
+ "pass@4": 0.7091189907918726,
654
+ "pass@8": 0.7677548796207677,
655
+ "pass@16": 0.810923847471511,
656
+ "pass@32": 0.8425591447235117,
657
+ "pass@64": 0.8667520481945041,
658
+ "pass@128": 0.8895184135977338
659
+ },
660
+ "12": {
661
+ "pass@1": 0.5407618290960452,
662
+ "pass@2": 0.6592635765603452,
663
+ "pass@4": 0.7460638715181781,
664
+ "pass@8": 0.8079141082622625,
665
+ "pass@16": 0.852190879714896,
666
+ "pass@32": 0.8835107730092441,
667
+ "pass@64": 0.9053088531949328,
668
+ "pass@128": 0.9209039548022598
669
+ },
670
+ "13": {
671
+ "pass@1": 0.41783405172413796,
672
+ "pass@2": 0.5587110927356649,
673
+ "pass@4": 0.6785095712957503,
674
+ "pass@8": 0.7668253027465036,
675
+ "pass@16": 0.8288992955389046,
676
+ "pass@32": 0.8726845941313368,
677
+ "pass@64": 0.9048733323087187,
678
+ "pass@128": 0.9310344827586207
679
+ },
680
+ "14": {
681
+ "pass@1": 0.3425330996884735,
682
+ "pass@2": 0.47973429796158645,
683
+ "pass@4": 0.6109350025756123,
684
+ "pass@8": 0.7166362098428933,
685
+ "pass@16": 0.7922397712359336,
686
+ "pass@32": 0.8449829905372319,
687
+ "pass@64": 0.8827357163562379,
688
+ "pass@128": 0.9127725856697819
689
+ },
690
+ "15": {
691
+ "pass@1": 0.24970646676300579,
692
+ "pass@2": 0.36728884097219044,
693
+ "pass@4": 0.4872996991705517,
694
+ "pass@8": 0.590706698102598,
695
+ "pass@16": 0.6738243586965353,
696
+ "pass@32": 0.7419453886625642,
697
+ "pass@64": 0.796950252939072,
698
+ "pass@128": 0.838150289017341
699
+ },
700
+ "16": {
701
+ "pass@1": 0.22497479838709677,
702
+ "pass@2": 0.33648915735331486,
703
+ "pass@4": 0.4564946569178856,
704
+ "pass@8": 0.5695374743650482,
705
+ "pass@16": 0.6659607330734459,
706
+ "pass@32": 0.7420235555671716,
707
+ "pass@64": 0.8008246573666877,
708
+ "pass@128": 0.8516129032258064
709
+ },
710
+ "17": {
711
+ "pass@1": 0.1771134393063584,
712
+ "pass@2": 0.27040438248145277,
713
+ "pass@4": 0.3734399091551416,
714
+ "pass@8": 0.4705561750269978,
715
+ "pass@16": 0.556979762253697,
716
+ "pass@32": 0.6330896125700303,
717
+ "pass@64": 0.6962373325698672,
718
+ "pass@128": 0.7456647398843931
719
+ },
720
+ "18": {
721
+ "pass@1": 0.16613839285714285,
722
+ "pass@2": 0.25231404668166474,
723
+ "pass@4": 0.3428713514382129,
724
+ "pass@8": 0.4269419637345666,
725
+ "pass@16": 0.5030587296667712,
726
+ "pass@32": 0.5724071144846937,
727
+ "pass@64": 0.6359387733746433,
728
+ "pass@128": 0.6971428571428572
729
+ },
730
+ "19": {
731
+ "pass@1": 0.15615615615615616,
732
+ "pass@2": 0.2335178436310326,
733
+ "pass@4": 0.3141585730837702,
734
+ "pass@8": 0.3884583356444021,
735
+ "pass@16": 0.45846532735472434,
736
+ "pass@32": 0.5290161554991337,
737
+ "pass@64": 0.5971115408381422,
738
+ "pass@128": 0.6606606606606606
739
+ },
740
+ "20": {
741
+ "pass@1": 0.14114114114114115,
742
+ "pass@2": 0.22127085254545897,
743
+ "pass@4": 0.3091605038671966,
744
+ "pass@8": 0.38948503249060373,
745
+ "pass@16": 0.4591147943528322,
746
+ "pass@32": 0.5236253667328297,
747
+ "pass@64": 0.5845144560920016,
748
+ "pass@128": 0.6336336336336337
749
+ }
750
+ }
751
+ },
752
+ "teachers_in_school": {
753
+ "count": 430464,
754
+ "correct": 127571,
755
+ "answer_accuracy": 0.29635695435622955,
756
+ "avg_response_len": 328.95371738403213,
757
+ "resp_tokens_sum": 141602733,
758
+ "pass_at_k": {
759
+ "pass@1": 0.29635695435622955,
760
+ "pass@2": 0.3994659635542887,
761
+ "pass@4": 0.4983539200218205,
762
+ "pass@8": 0.5834962063768169,
763
+ "pass@16": 0.6522365336911832,
764
+ "pass@32": 0.7075035234484064,
765
+ "pass@64": 0.752666990208834,
766
+ "pass@128": 0.7891763306571513
767
+ },
768
+ "per_op_pass_at_k": {
769
+ "11": {
770
+ "pass@1": 0.5852315512048193,
771
+ "pass@2": 0.6793565749217343,
772
+ "pass@4": 0.7481086757039406,
773
+ "pass@8": 0.79819997407497,
774
+ "pass@16": 0.8376718939360224,
775
+ "pass@32": 0.8684909661065356,
776
+ "pass@64": 0.8906779124327036,
777
+ "pass@128": 0.9036144578313253
778
+ },
779
+ "12": {
780
+ "pass@1": 0.48897923197492166,
781
+ "pass@2": 0.6015723348061118,
782
+ "pass@4": 0.6915262334958906,
783
+ "pass@8": 0.7628699479020743,
784
+ "pass@16": 0.8170783827895058,
785
+ "pass@32": 0.8559537327526784,
786
+ "pass@64": 0.8835455838809168,
787
+ "pass@128": 0.9028213166144201
788
+ },
789
+ "13": {
790
+ "pass@1": 0.3866421568627451,
791
+ "pass@2": 0.5096691727872251,
792
+ "pass@4": 0.6137185039370071,
793
+ "pass@8": 0.698094258047033,
794
+ "pass@16": 0.7655805416707796,
795
+ "pass@32": 0.8172702345082308,
796
+ "pass@64": 0.8567427013588318,
797
+ "pass@128": 0.8879551820728291
798
+ },
799
+ "14": {
800
+ "pass@1": 0.34528051900584794,
801
+ "pass@2": 0.4769977868720359,
802
+ "pass@4": 0.5996480609879901,
803
+ "pass@8": 0.6973310194018792,
804
+ "pass@16": 0.7674308590346286,
805
+ "pass@32": 0.8171571009241003,
806
+ "pass@64": 0.85288334433611,
807
+ "pass@128": 0.8771929824561403
808
+ },
809
+ "15": {
810
+ "pass@1": 0.28020833333333334,
811
+ "pass@2": 0.405685173586256,
812
+ "pass@4": 0.5307156491802157,
813
+ "pass@8": 0.6359535624159368,
814
+ "pass@16": 0.7152338549121703,
815
+ "pass@32": 0.775609713345843,
816
+ "pass@64": 0.822731933165433,
817
+ "pass@128": 0.8575757575757575
818
+ },
819
+ "16": {
820
+ "pass@1": 0.22115384615384615,
821
+ "pass@2": 0.3294303256163492,
822
+ "pass@4": 0.4457514172907874,
823
+ "pass@8": 0.5532684120879664,
824
+ "pass@16": 0.6416775711504988,
825
+ "pass@32": 0.7124958371733157,
826
+ "pass@64": 0.771953988942229,
827
+ "pass@128": 0.8205128205128205
828
+ },
829
+ "17": {
830
+ "pass@1": 0.2061820652173913,
831
+ "pass@2": 0.3083883791020687,
832
+ "pass@4": 0.41528318888322185,
833
+ "pass@8": 0.5125658863922051,
834
+ "pass@16": 0.5953014311232198,
835
+ "pass@32": 0.6644794210402157,
836
+ "pass@64": 0.7232154857023478,
837
+ "pass@128": 0.7732919254658385
838
+ },
839
+ "18": {
840
+ "pass@1": 0.15284926470588237,
841
+ "pass@2": 0.23485265169059735,
842
+ "pass@4": 0.32879796367365843,
843
+ "pass@8": 0.42114827587863835,
844
+ "pass@16": 0.5031462412182746,
845
+ "pass@32": 0.5751373282485435,
846
+ "pass@64": 0.6395195992057614,
847
+ "pass@128": 0.6941176470588235
848
+ },
849
+ "19": {
850
+ "pass@1": 0.15843563988095238,
851
+ "pass@2": 0.23614211309523814,
852
+ "pass@4": 0.31818266717776345,
853
+ "pass@8": 0.39120863355325985,
854
+ "pass@16": 0.451844936121816,
855
+ "pass@32": 0.5046091647650751,
856
+ "pass@64": 0.5519295462717415,
857
+ "pass@128": 0.5982142857142857
858
+ },
859
+ "20": {
860
+ "pass@1": 0.14539670658682635,
861
+ "pass@2": 0.21817388844358518,
862
+ "pass@4": 0.2958124524292249,
863
+ "pass@8": 0.36673161731226755,
864
+ "pass@16": 0.42847335860140157,
865
+ "pass@32": 0.4838859785436273,
866
+ "pass@64": 0.532640616772042,
867
+ "pass@128": 0.5748502994011976
868
+ }
869
+ }
870
+ },
871
+ "movie_festival_awards": {
872
+ "count": 418816,
873
+ "correct": 118748,
874
+ "answer_accuracy": 0.28353262530562345,
875
+ "avg_response_len": 350.77999407854526,
876
+ "resp_tokens_sum": 146912274,
877
+ "pass_at_k": {
878
+ "pass@1": 0.28353262530562345,
879
+ "pass@2": 0.3841836694790448,
880
+ "pass@4": 0.48139252662373644,
881
+ "pass@8": 0.5666045170074969,
882
+ "pass@16": 0.636885319894913,
883
+ "pass@32": 0.6942662192065767,
884
+ "pass@64": 0.741034527225611,
885
+ "pass@128": 0.7799511002444988
886
+ },
887
+ "per_op_pass_at_k": {
888
+ "11": {
889
+ "pass@1": 0.533779761904762,
890
+ "pass@2": 0.637331271091113,
891
+ "pass@4": 0.7189003565030557,
892
+ "pass@8": 0.7825376796716175,
893
+ "pass@16": 0.8307175370206826,
894
+ "pass@32": 0.8640638671817454,
895
+ "pass@64": 0.8861393273804471,
896
+ "pass@128": 0.9047619047619048
897
+ },
898
+ "12": {
899
+ "pass@1": 0.519447629969419,
900
+ "pass@2": 0.6325109411495584,
901
+ "pass@4": 0.716240447868328,
902
+ "pass@8": 0.7778574011864423,
903
+ "pass@16": 0.8239268041149975,
904
+ "pass@32": 0.8611130957104958,
905
+ "pass@64": 0.8914035041714263,
906
+ "pass@128": 0.9113149847094801
907
+ },
908
+ "13": {
909
+ "pass@1": 0.4123022762345679,
910
+ "pass@2": 0.5496697895401961,
911
+ "pass@4": 0.6635212849551213,
912
+ "pass@8": 0.7482771208219928,
913
+ "pass@16": 0.8095717661613954,
914
+ "pass@32": 0.8546775092104458,
915
+ "pass@64": 0.8861848341058806,
916
+ "pass@128": 0.9074074074074074
917
+ },
918
+ "14": {
919
+ "pass@1": 0.3145168768545994,
920
+ "pass@2": 0.4401428041543027,
921
+ "pass@4": 0.5588902007612545,
922
+ "pass@8": 0.6583031703543211,
923
+ "pass@16": 0.7376897670615584,
924
+ "pass@32": 0.7999922261319541,
925
+ "pass@64": 0.8459428217294271,
926
+ "pass@128": 0.8813056379821959
927
+ },
928
+ "15": {
929
+ "pass@1": 0.23210841049382716,
930
+ "pass@2": 0.3398600782541072,
931
+ "pass@4": 0.4533102031227575,
932
+ "pass@8": 0.5593127461017794,
933
+ "pass@16": 0.6496853742117522,
934
+ "pass@32": 0.7226145759565928,
935
+ "pass@64": 0.778493587814901,
936
+ "pass@128": 0.8179012345679012
937
+ },
938
+ "16": {
939
+ "pass@1": 0.21082227138643067,
940
+ "pass@2": 0.3154901371565279,
941
+ "pass@4": 0.4299715937056539,
942
+ "pass@8": 0.5375134403941474,
943
+ "pass@16": 0.6273091818412996,
944
+ "pass@32": 0.6999977456346922,
945
+ "pass@64": 0.760873083209445,
946
+ "pass@128": 0.8112094395280236
947
+ },
948
+ "17": {
949
+ "pass@1": 0.16036803463855423,
950
+ "pass@2": 0.24516397282041513,
951
+ "pass@4": 0.34268070369215886,
952
+ "pass@8": 0.43866697297296425,
953
+ "pass@16": 0.5218814653519248,
954
+ "pass@32": 0.591200169404961,
955
+ "pass@64": 0.6489296836444958,
956
+ "pass@128": 0.6987951807228916
957
+ },
958
+ "18": {
959
+ "pass@1": 0.16050907258064517,
960
+ "pass@2": 0.23742379984759968,
961
+ "pass@4": 0.32174134160649276,
962
+ "pass@8": 0.4042682761697166,
963
+ "pass@16": 0.48119832780048005,
964
+ "pass@32": 0.5537932896201498,
965
+ "pass@64": 0.6174600257015876,
966
+ "pass@128": 0.6709677419354839
967
+ },
968
+ "19": {
969
+ "pass@1": 0.1531344410876133,
970
+ "pass@2": 0.22887798249161445,
971
+ "pass@4": 0.3099796786164572,
972
+ "pass@8": 0.38504632696140717,
973
+ "pass@16": 0.44934495744307373,
974
+ "pass@32": 0.503410042848671,
975
+ "pass@64": 0.5509170785946527,
976
+ "pass@128": 0.5921450151057401
977
+ },
978
+ "20": {
979
+ "pass@1": 0.14949324324324326,
980
+ "pass@2": 0.2258773291007543,
981
+ "pass@4": 0.30744521067749414,
982
+ "pass@8": 0.3811890808222759,
983
+ "pass@16": 0.4431254983049504,
984
+ "pass@32": 0.49655440542341023,
985
+ "pass@64": 0.5479183432668487,
986
+ "pass@128": 0.6066066066066066
987
+ }
988
+ }
989
+ }
990
+ }
991
+ },
992
+ "total": {
993
+ "answer_accuracy": 0.4945670230263158,
994
+ "avg_loss": 0.0,
995
+ "count": 2432000,
996
+ "validation_metrics": {
997
+ "id": {},
998
+ "ood": {}
999
+ },
1000
+ "pass_at_k": {
1001
+ "pass@1": 0.4945670230263158,
1002
+ "pass@2": 0.5761442252900684,
1003
+ "pass@4": 0.6483423776517793,
1004
+ "pass@8": 0.707801672744191,
1005
+ "pass@16": 0.755014521390677,
1006
+ "pass@32": 0.793115610405612,
1007
+ "pass@64": 0.824534994508942,
1008
+ "pass@128": 0.8510526315789474
1009
+ },
1010
+ "per_op_pass_at_k": {
1011
+ "10": {
1012
+ "pass@1": 0.6137109375,
1013
+ "pass@2": 0.6928655265748027,
1014
+ "pass@4": 0.7514061629171355,
1015
+ "pass@8": 0.7965425200019911,
1016
+ "pass@16": 0.8332050121400162,
1017
+ "pass@32": 0.863816716890956,
1018
+ "pass@64": 0.8889742483433577,
1019
+ "pass@128": 0.91
1020
+ },
1021
+ "2": {
1022
+ "pass@1": 0.92753125,
1023
+ "pass@2": 0.9454394685039371,
1024
+ "pass@4": 0.9582661204536929,
1025
+ "pass@8": 0.9675005822154368,
1026
+ "pass@16": 0.9738152511151082,
1027
+ "pass@32": 0.978352349440396,
1028
+ "pass@64": 0.982220783659935,
1029
+ "pass@128": 0.986
1030
+ },
1031
+ "3": {
1032
+ "pass@1": 0.87525,
1033
+ "pass@2": 0.8933171751968509,
1034
+ "pass@4": 0.903821142950881,
1035
+ "pass@8": 0.9113115359504261,
1036
+ "pass@16": 0.9173621733495384,
1037
+ "pass@32": 0.9227675643594974,
1038
+ "pass@64": 0.9282661218073456,
1039
+ "pass@128": 0.934
1040
+ },
1041
+ "4": {
1042
+ "pass@1": 0.7896640625,
1043
+ "pass@2": 0.8281980807086614,
1044
+ "pass@4": 0.8554755267154103,
1045
+ "pass@8": 0.8741056836959573,
1046
+ "pass@16": 0.8877460253022831,
1047
+ "pass@32": 0.8997405892622435,
1048
+ "pass@64": 0.911213861081669,
1049
+ "pass@128": 0.923
1050
+ },
1051
+ "5": {
1052
+ "pass@1": 0.698640625,
1053
+ "pass@2": 0.7646503444881888,
1054
+ "pass@4": 0.8178164796587929,
1055
+ "pass@8": 0.8575505477175563,
1056
+ "pass@16": 0.8860211077579768,
1057
+ "pass@32": 0.906900117552899,
1058
+ "pass@64": 0.9237245923609523,
1059
+ "pass@128": 0.94
1060
+ },
1061
+ "6": {
1062
+ "pass@1": 0.672421875,
1063
+ "pass@2": 0.7338943159448824,
1064
+ "pass@4": 0.7815458341769774,
1065
+ "pass@8": 0.8147635720348121,
1066
+ "pass@16": 0.8370253152434817,
1067
+ "pass@32": 0.8540255270767936,
1068
+ "pass@64": 0.8681434397988514,
1069
+ "pass@128": 0.879
1070
+ },
1071
+ "7": {
1072
+ "pass@1": 0.66359375,
1073
+ "pass@2": 0.7303299704724414,
1074
+ "pass@4": 0.7794709802212226,
1075
+ "pass@8": 0.8147819203326597,
1076
+ "pass@16": 0.8397578705147839,
1077
+ "pass@32": 0.8577877159512644,
1078
+ "pass@64": 0.872366823689421,
1079
+ "pass@128": 0.885
1080
+ },
1081
+ "8": {
1082
+ "pass@1": 0.6255859375,
1083
+ "pass@2": 0.7037237942913391,
1084
+ "pass@4": 0.7589305926134238,
1085
+ "pass@8": 0.7973030953082081,
1086
+ "pass@16": 0.8247642585556798,
1087
+ "pass@32": 0.8463344723730156,
1088
+ "pass@64": 0.8662206995650144,
1089
+ "pass@128": 0.888
1090
+ },
1091
+ "9": {
1092
+ "pass@1": 0.6123828125,
1093
+ "pass@2": 0.7015214074803167,
1094
+ "pass@4": 0.7668119784401946,
1095
+ "pass@8": 0.8142296413699427,
1096
+ "pass@16": 0.8476283425102903,
1097
+ "pass@32": 0.8720244793500149,
1098
+ "pass@64": 0.8897361231796059,
1099
+ "pass@128": 0.902
1100
+ },
1101
+ "11": {
1102
+ "pass@1": 0.5468203125,
1103
+ "pass@2": 0.648342027559055,
1104
+ "pass@4": 0.7251446963817036,
1105
+ "pass@8": 0.7825192329955811,
1106
+ "pass@16": 0.8260392111057165,
1107
+ "pass@32": 0.8579424969970187,
1108
+ "pass@64": 0.8808024280651586,
1109
+ "pass@128": 0.899
1110
+ },
1111
+ "12": {
1112
+ "pass@1": 0.5172734375,
1113
+ "pass@2": 0.6321119586614163,
1114
+ "pass@4": 0.718914105455568,
1115
+ "pass@8": 0.7837164778935699,
1116
+ "pass@16": 0.831747640474528,
1117
+ "pass@32": 0.867396036690709,
1118
+ "pass@64": 0.8938193211530763,
1119
+ "pass@128": 0.912
1120
+ },
1121
+ "13": {
1122
+ "pass@1": 0.40490625,
1123
+ "pass@2": 0.5382737450787406,
1124
+ "pass@4": 0.6505229554743155,
1125
+ "pass@8": 0.7362787088452506,
1126
+ "pass@16": 0.8000323808896687,
1127
+ "pass@32": 0.84706737223152,
1128
+ "pass@64": 0.8816356236418902,
1129
+ "pass@128": 0.908
1130
+ },
1131
+ "14": {
1132
+ "pass@1": 0.33403125,
1133
+ "pass@2": 0.4654560777559057,
1134
+ "pass@4": 0.5895357703412076,
1135
+ "pass@8": 0.6903756004044173,
1136
+ "pass@16": 0.7653717718563208,
1137
+ "pass@32": 0.8203046486849622,
1138
+ "pass@64": 0.8601269996361208,
1139
+ "pass@128": 0.89
1140
+ },
1141
+ "15": {
1142
+ "pass@1": 0.2540703125,
1143
+ "pass@2": 0.37107271161417366,
1144
+ "pass@4": 0.4906143659542563,
1145
+ "pass@8": 0.5954665228777352,
1146
+ "pass@16": 0.6796684614746239,
1147
+ "pass@32": 0.7467914324913115,
1148
+ "pass@64": 0.7994782479135412,
1149
+ "pass@128": 0.838
1150
+ },
1151
+ "16": {
1152
+ "pass@1": 0.2188359375,
1153
+ "pass@2": 0.32689283956692883,
1154
+ "pass@4": 0.4437324613798277,
1155
+ "pass@8": 0.5529708859896567,
1156
+ "pass@16": 0.6443344673707929,
1157
+ "pass@32": 0.7174125768438164,
1158
+ "pass@64": 0.7771474691103983,
1159
+ "pass@128": 0.827
1160
+ },
1161
+ "17": {
1162
+ "pass@1": 0.1809140625,
1163
+ "pass@2": 0.27425541338582665,
1164
+ "pass@4": 0.3767013890138727,
1165
+ "pass@8": 0.4734960870046542,
1166
+ "pass@16": 0.5576667050582922,
1167
+ "pass@32": 0.629289835766627,
1168
+ "pass@64": 0.689218158435304,
1169
+ "pass@128": 0.739
1170
+ },
1171
+ "18": {
1172
+ "pass@1": 0.159875,
1173
+ "pass@2": 0.24176119586614178,
1174
+ "pass@4": 0.331536096550431,
1175
+ "pass@8": 0.4179432667184478,
1176
+ "pass@16": 0.49631175901573166,
1177
+ "pass@32": 0.5675651014563933,
1178
+ "pass@64": 0.6314278423785766,
1179
+ "pass@128": 0.688
1180
+ },
1181
+ "19": {
1182
+ "pass@1": 0.155921875,
1183
+ "pass@2": 0.23286380413385835,
1184
+ "pass@4": 0.31412745463067143,
1185
+ "pass@8": 0.3882530608677067,
1186
+ "pass@16": 0.45322203345971057,
1187
+ "pass@32": 0.5123397833251867,
1188
+ "pass@64": 0.5666400236612369,
1189
+ "pass@128": 0.617
1190
+ },
1191
+ "20": {
1192
+ "pass@1": 0.14534375,
1193
+ "pass@2": 0.22177042322834656,
1194
+ "pass@4": 0.3041310620547426,
1195
+ "pass@8": 0.37912283991548595,
1196
+ "pass@16": 0.44355611922790866,
1197
+ "pass@32": 0.5013377809615996,
1198
+ "pass@64": 0.5550020881883593,
1199
+ "pass@128": 0.605
1200
+ }
1201
+ },
1202
+ "per_op_accuracy": {
1203
+ "10": 0.6137109375,
1204
+ "2": 0.92753125,
1205
+ "3": 0.87525,
1206
+ "4": 0.7896640625,
1207
+ "5": 0.698640625,
1208
+ "6": 0.672421875,
1209
+ "7": 0.66359375,
1210
+ "8": 0.6255859375,
1211
+ "9": 0.6123828125,
1212
+ "11": 0.5468203125,
1213
+ "12": 0.5172734375,
1214
+ "13": 0.40490625,
1215
+ "14": 0.33403125,
1216
+ "15": 0.2540703125,
1217
+ "16": 0.2188359375,
1218
+ "17": 0.1809140625,
1219
+ "18": 0.159875,
1220
+ "19": 0.155921875,
1221
+ "20": 0.14534375
1222
+ },
1223
+ "per_op_avg_response_len": {
1224
+ "10": 281.0300546875,
1225
+ "2": 123.2331015625,
1226
+ "3": 148.998984375,
1227
+ "4": 162.370984375,
1228
+ "5": 179.260140625,
1229
+ "6": 203.594375,
1230
+ "7": 226.4866328125,
1231
+ "8": 247.12684375,
1232
+ "9": 262.1648359375,
1233
+ "11": 296.6561015625,
1234
+ "12": 303.042890625,
1235
+ "13": 315.422875,
1236
+ "14": 333.3207109375,
1237
+ "15": 338.5395546875,
1238
+ "16": 343.201203125,
1239
+ "17": 346.66403125,
1240
+ "18": 348.8883203125,
1241
+ "19": 346.0546640625,
1242
+ "20": 353.26475
1243
+ },
1244
+ "per_op_avg_loss": {},
1245
+ "per_op_length": {
1246
+ "10": 281.0300546875,
1247
+ "2": 123.2331015625,
1248
+ "3": 148.998984375,
1249
+ "4": 162.370984375,
1250
+ "5": 179.260140625,
1251
+ "6": 203.594375,
1252
+ "7": 226.4866328125,
1253
+ "8": 247.12684375,
1254
+ "9": 262.1648359375,
1255
+ "11": 296.6561015625,
1256
+ "12": 303.042890625,
1257
+ "13": 315.422875,
1258
+ "14": 333.3207109375,
1259
+ "15": 338.5395546875,
1260
+ "16": 343.201203125,
1261
+ "17": 346.66403125,
1262
+ "18": 348.8883203125,
1263
+ "19": 346.0546640625,
1264
+ "20": 353.26475
1265
+ },
1266
+ "per_op_loss": {},
1267
+ "per_template": {
1268
+ "crazy_zootopia": {
1269
+ "count": 809728,
1270
+ "correct": 400116,
1271
+ "answer_accuracy": 0.49413630256085994,
1272
+ "avg_response_len": 260.3990833959058,
1273
+ "resp_tokens_sum": 210852429,
1274
+ "pass_at_k": {
1275
+ "pass@1": 0.49413630256085994,
1276
+ "pass@2": 0.5783138453725597,
1277
+ "pass@4": 0.6522844111813358,
1278
+ "pass@8": 0.7126406720979045,
1279
+ "pass@16": 0.7607958557092384,
1280
+ "pass@32": 0.8002622415970753,
1281
+ "pass@64": 0.8330792580630666,
1282
+ "pass@128": 0.8612077141953841
1283
+ },
1284
+ "per_op_pass_at_k": {
1285
+ "10": {
1286
+ "pass@1": 0.5904775073746312,
1287
+ "pass@2": 0.6759274179499686,
1288
+ "pass@4": 0.739327368647503,
1289
+ "pass@8": 0.7888088892182306,
1290
+ "pass@16": 0.8301737990227551,
1291
+ "pass@32": 0.8653796021525334,
1292
+ "pass@64": 0.8950191494801747,
1293
+ "pass@128": 0.9203539823008849
1294
+ },
1295
+ "2": {
1296
+ "pass@1": 0.9306853582554517,
1297
+ "pass@2": 0.951853291264994,
1298
+ "pass@4": 0.9672763547664015,
1299
+ "pass@8": 0.9777387432602255,
1300
+ "pass@16": 0.9849406600666408,
1301
+ "pass@32": 0.9906709430385632,
1302
+ "pass@64": 0.9946393301197776,
1303
+ "pass@128": 0.9968847352024922
1304
+ },
1305
+ "3": {
1306
+ "pass@1": 0.8729692278287462,
1307
+ "pass@2": 0.8916826193744132,
1308
+ "pass@4": 0.9012887589166031,
1309
+ "pass@8": 0.9069931487274436,
1310
+ "pass@16": 0.9113667010227782,
1311
+ "pass@32": 0.9158878607999068,
1312
+ "pass@64": 0.921283309008342,
1313
+ "pass@128": 0.926605504587156
1314
+ },
1315
+ "4": {
1316
+ "pass@1": 0.79365234375,
1317
+ "pass@2": 0.8259854053887796,
1318
+ "pass@4": 0.8509953552680912,
1319
+ "pass@8": 0.8705874877072775,
1320
+ "pass@16": 0.8861866677459165,
1321
+ "pass@32": 0.9001222965392037,
1322
+ "pass@64": 0.9150975754710673,
1323
+ "pass@128": 0.934375
1324
+ },
1325
+ "5": {
1326
+ "pass@1": 0.711954365079365,
1327
+ "pass@2": 0.7870863798275217,
1328
+ "pass@4": 0.8438272906362893,
1329
+ "pass@8": 0.8820148031784064,
1330
+ "pass@16": 0.9061775402294971,
1331
+ "pass@32": 0.9233827284096477,
1332
+ "pass@64": 0.939203360121733,
1333
+ "pass@128": 0.9587301587301588
1334
+ },
1335
+ "6": {
1336
+ "pass@1": 0.6886695359281437,
1337
+ "pass@2": 0.7508096503842708,
1338
+ "pass@4": 0.7981536092606685,
1339
+ "pass@8": 0.8300540071307995,
1340
+ "pass@16": 0.8515233346236275,
1341
+ "pass@32": 0.867618886093707,
1342
+ "pass@64": 0.8782963968718454,
1343
+ "pass@128": 0.8832335329341318
1344
+ },
1345
+ "7": {
1346
+ "pass@1": 0.6789478550295858,
1347
+ "pass@2": 0.7444577587010199,
1348
+ "pass@4": 0.7924073042090151,
1349
+ "pass@8": 0.8294529774774302,
1350
+ "pass@16": 0.85758601479046,
1351
+ "pass@32": 0.8766141160571631,
1352
+ "pass@64": 0.8892577006853214,
1353
+ "pass@128": 0.8994082840236687
1354
+ },
1355
+ "8": {
1356
+ "pass@1": 0.6259515224358975,
1357
+ "pass@2": 0.6997999949525543,
1358
+ "pass@4": 0.7484703330352931,
1359
+ "pass@8": 0.7817076118380484,
1360
+ "pass@16": 0.8062330519425672,
1361
+ "pass@32": 0.8273824576064478,
1362
+ "pass@64": 0.8497525386355358,
1363
+ "pass@128": 0.875
1364
+ },
1365
+ "9": {
1366
+ "pass@1": 0.6052596830985916,
1367
+ "pass@2": 0.6952658866585334,
1368
+ "pass@4": 0.7621365283564899,
1369
+ "pass@8": 0.8126598351468055,
1370
+ "pass@16": 0.8499978703599235,
1371
+ "pass@32": 0.8788524080867729,
1372
+ "pass@64": 0.8985118669285895,
1373
+ "pass@128": 0.9098591549295775
1374
+ },
1375
+ "11": {
1376
+ "pass@1": 0.5223309135977338,
1377
+ "pass@2": 0.6289980008253216,
1378
+ "pass@4": 0.7091189907918726,
1379
+ "pass@8": 0.7677548796207677,
1380
+ "pass@16": 0.810923847471511,
1381
+ "pass@32": 0.8425591447235117,
1382
+ "pass@64": 0.8667520481945041,
1383
+ "pass@128": 0.8895184135977338
1384
+ },
1385
+ "12": {
1386
+ "pass@1": 0.5407618290960452,
1387
+ "pass@2": 0.6592635765603452,
1388
+ "pass@4": 0.7460638715181781,
1389
+ "pass@8": 0.8079141082622625,
1390
+ "pass@16": 0.852190879714896,
1391
+ "pass@32": 0.8835107730092441,
1392
+ "pass@64": 0.9053088531949328,
1393
+ "pass@128": 0.9209039548022598
1394
+ },
1395
+ "13": {
1396
+ "pass@1": 0.41783405172413796,
1397
+ "pass@2": 0.5587110927356649,
1398
+ "pass@4": 0.6785095712957503,
1399
+ "pass@8": 0.7668253027465036,
1400
+ "pass@16": 0.8288992955389046,
1401
+ "pass@32": 0.8726845941313368,
1402
+ "pass@64": 0.9048733323087187,
1403
+ "pass@128": 0.9310344827586207
1404
+ },
1405
+ "14": {
1406
+ "pass@1": 0.3425330996884735,
1407
+ "pass@2": 0.47973429796158645,
1408
+ "pass@4": 0.6109350025756123,
1409
+ "pass@8": 0.7166362098428933,
1410
+ "pass@16": 0.7922397712359336,
1411
+ "pass@32": 0.8449829905372319,
1412
+ "pass@64": 0.8827357163562379,
1413
+ "pass@128": 0.9127725856697819
1414
+ },
1415
+ "15": {
1416
+ "pass@1": 0.24970646676300579,
1417
+ "pass@2": 0.36728884097219044,
1418
+ "pass@4": 0.4872996991705517,
1419
+ "pass@8": 0.590706698102598,
1420
+ "pass@16": 0.6738243586965353,
1421
+ "pass@32": 0.7419453886625642,
1422
+ "pass@64": 0.796950252939072,
1423
+ "pass@128": 0.838150289017341
1424
+ },
1425
+ "16": {
1426
+ "pass@1": 0.22497479838709677,
1427
+ "pass@2": 0.33648915735331486,
1428
+ "pass@4": 0.4564946569178856,
1429
+ "pass@8": 0.5695374743650482,
1430
+ "pass@16": 0.6659607330734459,
1431
+ "pass@32": 0.7420235555671716,
1432
+ "pass@64": 0.8008246573666877,
1433
+ "pass@128": 0.8516129032258064
1434
+ },
1435
+ "17": {
1436
+ "pass@1": 0.1771134393063584,
1437
+ "pass@2": 0.27040438248145277,
1438
+ "pass@4": 0.3734399091551416,
1439
+ "pass@8": 0.4705561750269978,
1440
+ "pass@16": 0.556979762253697,
1441
+ "pass@32": 0.6330896125700303,
1442
+ "pass@64": 0.6962373325698672,
1443
+ "pass@128": 0.7456647398843931
1444
+ },
1445
+ "18": {
1446
+ "pass@1": 0.16613839285714285,
1447
+ "pass@2": 0.25231404668166474,
1448
+ "pass@4": 0.3428713514382129,
1449
+ "pass@8": 0.4269419637345666,
1450
+ "pass@16": 0.5030587296667712,
1451
+ "pass@32": 0.5724071144846937,
1452
+ "pass@64": 0.6359387733746433,
1453
+ "pass@128": 0.6971428571428572
1454
+ },
1455
+ "19": {
1456
+ "pass@1": 0.15615615615615616,
1457
+ "pass@2": 0.2335178436310326,
1458
+ "pass@4": 0.3141585730837702,
1459
+ "pass@8": 0.3884583356444021,
1460
+ "pass@16": 0.45846532735472434,
1461
+ "pass@32": 0.5290161554991337,
1462
+ "pass@64": 0.5971115408381422,
1463
+ "pass@128": 0.6606606606606606
1464
+ },
1465
+ "20": {
1466
+ "pass@1": 0.14114114114114115,
1467
+ "pass@2": 0.22127085254545897,
1468
+ "pass@4": 0.3091605038671966,
1469
+ "pass@8": 0.38948503249060373,
1470
+ "pass@16": 0.4591147943528322,
1471
+ "pass@32": 0.5236253667328297,
1472
+ "pass@64": 0.5845144560920016,
1473
+ "pass@128": 0.6336336336336337
1474
+ }
1475
+ }
1476
+ },
1477
+ "teachers_in_school": {
1478
+ "count": 821120,
1479
+ "correct": 410116,
1480
+ "answer_accuracy": 0.49945927513639904,
1481
+ "avg_response_len": 268.1280945537802,
1482
+ "resp_tokens_sum": 220165341,
1483
+ "pass_at_k": {
1484
+ "pass@1": 0.49945927513639904,
1485
+ "pass@2": 0.5797713589581576,
1486
+ "pass@4": 0.6508590268139766,
1487
+ "pass@8": 0.7092176995513156,
1488
+ "pass@16": 0.7549870987709693,
1489
+ "pass@32": 0.7915088244120045,
1490
+ "pass@64": 0.8217816179895945,
1491
+ "pass@128": 0.8470771628994544
1492
+ },
1493
+ "per_op_pass_at_k": {
1494
+ "10": {
1495
+ "pass@1": 0.6425421779141104,
1496
+ "pass@2": 0.7180774781411522,
1497
+ "pass@4": 0.7713729633489064,
1498
+ "pass@8": 0.8115314019701301,
1499
+ "pass@16": 0.8422782157793477,
1500
+ "pass@32": 0.8665394346696862,
1501
+ "pass@64": 0.8875838773290864,
1502
+ "pass@128": 0.9079754601226994
1503
+ },
1504
+ "2": {
1505
+ "pass@1": 0.9346267097264438,
1506
+ "pass@2": 0.9484187648086544,
1507
+ "pass@4": 0.9577753805660308,
1508
+ "pass@8": 0.9642355108332087,
1509
+ "pass@16": 0.9687406308251917,
1510
+ "pass@32": 0.9722132661616432,
1511
+ "pass@64": 0.976268636177634,
1512
+ "pass@128": 0.9817629179331308
1513
+ },
1514
+ "3": {
1515
+ "pass@1": 0.8785274621212121,
1516
+ "pass@2": 0.8940225334049151,
1517
+ "pass@4": 0.9043032615241278,
1518
+ "pass@8": 0.9116606528207251,
1519
+ "pass@16": 0.9163981804252643,
1520
+ "pass@32": 0.919534935837756,
1521
+ "pass@64": 0.9226355633253187,
1522
+ "pass@128": 0.9272727272727272
1523
+ },
1524
+ "4": {
1525
+ "pass@1": 0.791350776627219,
1526
+ "pass@2": 0.8298940327540418,
1527
+ "pass@4": 0.8566257274423535,
1528
+ "pass@8": 0.8740505626426824,
1529
+ "pass@16": 0.8863419432111916,
1530
+ "pass@32": 0.8973476695044501,
1531
+ "pass@64": 0.9076491744197636,
1532
+ "pass@128": 0.9171597633136095
1533
+ },
1534
+ "5": {
1535
+ "pass@1": 0.7036231884057971,
1536
+ "pass@2": 0.760960144927536,
1537
+ "pass@4": 0.8089746806105759,
1538
+ "pass@8": 0.848232289495532,
1539
+ "pass@16": 0.8788940387283235,
1540
+ "pass@32": 0.903333779760882,
1541
+ "pass@64": 0.9245269760364313,
1542
+ "pass@128": 0.9449275362318841
1543
+ },
1544
+ "6": {
1545
+ "pass@1": 0.6609733166189111,
1546
+ "pass@2": 0.7211168580646615,
1547
+ "pass@4": 0.7662969808752413,
1548
+ "pass@8": 0.7963462083773434,
1549
+ "pass@16": 0.8159017311926985,
1550
+ "pass@32": 0.8326346080497815,
1551
+ "pass@64": 0.8482729333475885,
1552
+ "pass@128": 0.8595988538681948
1553
+ },
1554
+ "7": {
1555
+ "pass@1": 0.669549313583815,
1556
+ "pass@2": 0.7348724597196304,
1557
+ "pass@4": 0.7796674157601687,
1558
+ "pass@8": 0.8099202435632497,
1559
+ "pass@16": 0.8312183082099309,
1560
+ "pass@32": 0.8471171620508142,
1561
+ "pass@64": 0.8607780565256175,
1562
+ "pass@128": 0.8728323699421965
1563
+ },
1564
+ "8": {
1565
+ "pass@1": 0.6231073943661972,
1566
+ "pass@2": 0.7033488826660749,
1567
+ "pass@4": 0.7635277449473741,
1568
+ "pass@8": 0.8050630470213901,
1569
+ "pass@16": 0.8335760394037561,
1570
+ "pass@32": 0.8552420543936154,
1571
+ "pass@64": 0.8752369946603544,
1572
+ "pass@128": 0.8985915492957747
1573
+ },
1574
+ "9": {
1575
+ "pass@1": 0.6189651946107785,
1576
+ "pass@2": 0.7066431850865202,
1577
+ "pass@4": 0.7706885601687509,
1578
+ "pass@8": 0.8163749170683283,
1579
+ "pass@16": 0.8472702317498962,
1580
+ "pass@32": 0.8686874199142173,
1581
+ "pass@64": 0.883720682738893,
1582
+ "pass@128": 0.8922155688622755
1583
+ },
1584
+ "11": {
1585
+ "pass@1": 0.5852315512048193,
1586
+ "pass@2": 0.6793565749217343,
1587
+ "pass@4": 0.7481086757039406,
1588
+ "pass@8": 0.79819997407497,
1589
+ "pass@16": 0.8376718939360224,
1590
+ "pass@32": 0.8684909661065356,
1591
+ "pass@64": 0.8906779124327036,
1592
+ "pass@128": 0.9036144578313253
1593
+ },
1594
+ "12": {
1595
+ "pass@1": 0.48897923197492166,
1596
+ "pass@2": 0.6015723348061118,
1597
+ "pass@4": 0.6915262334958906,
1598
+ "pass@8": 0.7628699479020743,
1599
+ "pass@16": 0.8170783827895058,
1600
+ "pass@32": 0.8559537327526784,
1601
+ "pass@64": 0.8835455838809168,
1602
+ "pass@128": 0.9028213166144201
1603
+ },
1604
+ "13": {
1605
+ "pass@1": 0.3866421568627451,
1606
+ "pass@2": 0.5096691727872251,
1607
+ "pass@4": 0.6137185039370071,
1608
+ "pass@8": 0.698094258047033,
1609
+ "pass@16": 0.7655805416707796,
1610
+ "pass@32": 0.8172702345082308,
1611
+ "pass@64": 0.8567427013588318,
1612
+ "pass@128": 0.8879551820728291
1613
+ },
1614
+ "14": {
1615
+ "pass@1": 0.34528051900584794,
1616
+ "pass@2": 0.4769977868720359,
1617
+ "pass@4": 0.5996480609879901,
1618
+ "pass@8": 0.6973310194018792,
1619
+ "pass@16": 0.7674308590346286,
1620
+ "pass@32": 0.8171571009241003,
1621
+ "pass@64": 0.85288334433611,
1622
+ "pass@128": 0.8771929824561403
1623
+ },
1624
+ "15": {
1625
+ "pass@1": 0.28020833333333334,
1626
+ "pass@2": 0.405685173586256,
1627
+ "pass@4": 0.5307156491802157,
1628
+ "pass@8": 0.6359535624159368,
1629
+ "pass@16": 0.7152338549121703,
1630
+ "pass@32": 0.775609713345843,
1631
+ "pass@64": 0.822731933165433,
1632
+ "pass@128": 0.8575757575757575
1633
+ },
1634
+ "16": {
1635
+ "pass@1": 0.22115384615384615,
1636
+ "pass@2": 0.3294303256163492,
1637
+ "pass@4": 0.4457514172907874,
1638
+ "pass@8": 0.5532684120879664,
1639
+ "pass@16": 0.6416775711504988,
1640
+ "pass@32": 0.7124958371733157,
1641
+ "pass@64": 0.771953988942229,
1642
+ "pass@128": 0.8205128205128205
1643
+ },
1644
+ "17": {
1645
+ "pass@1": 0.2061820652173913,
1646
+ "pass@2": 0.3083883791020687,
1647
+ "pass@4": 0.41528318888322185,
1648
+ "pass@8": 0.5125658863922051,
1649
+ "pass@16": 0.5953014311232198,
1650
+ "pass@32": 0.6644794210402157,
1651
+ "pass@64": 0.7232154857023478,
1652
+ "pass@128": 0.7732919254658385
1653
+ },
1654
+ "18": {
1655
+ "pass@1": 0.15284926470588237,
1656
+ "pass@2": 0.23485265169059735,
1657
+ "pass@4": 0.32879796367365843,
1658
+ "pass@8": 0.42114827587863835,
1659
+ "pass@16": 0.5031462412182746,
1660
+ "pass@32": 0.5751373282485435,
1661
+ "pass@64": 0.6395195992057614,
1662
+ "pass@128": 0.6941176470588235
1663
+ },
1664
+ "19": {
1665
+ "pass@1": 0.15843563988095238,
1666
+ "pass@2": 0.23614211309523814,
1667
+ "pass@4": 0.31818266717776345,
1668
+ "pass@8": 0.39120863355325985,
1669
+ "pass@16": 0.451844936121816,
1670
+ "pass@32": 0.5046091647650751,
1671
+ "pass@64": 0.5519295462717415,
1672
+ "pass@128": 0.5982142857142857
1673
+ },
1674
+ "20": {
1675
+ "pass@1": 0.14539670658682635,
1676
+ "pass@2": 0.21817388844358518,
1677
+ "pass@4": 0.2958124524292249,
1678
+ "pass@8": 0.36673161731226755,
1679
+ "pass@16": 0.42847335860140157,
1680
+ "pass@32": 0.4838859785436273,
1681
+ "pass@64": 0.532640616772042,
1682
+ "pass@128": 0.5748502994011976
1683
+ }
1684
+ }
1685
+ },
1686
+ "movie_festival_awards": {
1687
+ "count": 801152,
1688
+ "correct": 392555,
1689
+ "answer_accuracy": 0.4899881670394632,
1690
+ "avg_response_len": 286.30687435093466,
1691
+ "resp_tokens_sum": 229375325,
1692
+ "pass_at_k": {
1693
+ "pass@1": 0.4899881670394632,
1694
+ "pass@2": 0.5702338435802126,
1695
+ "pass@4": 0.6417787718868243,
1696
+ "pass@8": 0.7014595538946482,
1697
+ "pass@16": 0.74919940638925,
1698
+ "pass@32": 0.7875393111935014,
1699
+ "pass@64": 0.8187212701325537,
1700
+ "pass@128": 0.8448633967087394
1701
+ },
1702
+ "per_op_pass_at_k": {
1703
+ "10": {
1704
+ "pass@1": 0.6091651119402985,
1705
+ "pass@2": 0.6854712657186507,
1706
+ "pass@4": 0.7441988026683222,
1707
+ "pass@8": 0.7897822970589528,
1708
+ "pass@16": 0.8274429729171232,
1709
+ "pass@32": 0.8595856001759086,
1710
+ "pass@64": 0.8842101870456582,
1711
+ "pass@128": 0.9014925373134328
1712
+ },
1713
+ "2": {
1714
+ "pass@1": 0.91796875,
1715
+ "pass@2": 0.9367565382452198,
1716
+ "pass@4": 0.9504637439070117,
1717
+ "pass@8": 0.9611798930422266,
1718
+ "pass@16": 0.9683818048349383,
1719
+ "pass@32": 0.9728251775938187,
1720
+ "pass@64": 0.9764262211115565,
1721
+ "pass@128": 0.98
1722
+ },
1723
+ "3": {
1724
+ "pass@1": 0.874271137026239,
1725
+ "pass@2": 0.8941968590023187,
1726
+ "pass@4": 0.9057715524262092,
1727
+ "pass@8": 0.915092597334439,
1728
+ "pass@16": 0.9240054302470929,
1729
+ "pass@32": 0.9324364578760008,
1730
+ "pass@64": 0.9403403611202984,
1731
+ "pass@128": 0.9475218658892128
1732
+ },
1733
+ "4": {
1734
+ "pass@1": 0.784265350877193,
1735
+ "pass@2": 0.8285923038403095,
1736
+ "pass@4": 0.8585307519125894,
1737
+ "pass@8": 0.8774520393462032,
1738
+ "pass@16": 0.8905927333865701,
1739
+ "pass@32": 0.9017483686467666,
1740
+ "pass@64": 0.9111029706931209,
1741
+ "pass@128": 0.9181286549707602
1742
+ },
1743
+ "5": {
1744
+ "pass@1": 0.68125,
1745
+ "pass@2": 0.7476084848309403,
1746
+ "pass@4": 0.8026900538168023,
1747
+ "pass@8": 0.8443404260011751,
1748
+ "pass@16": 0.8745786447776861,
1749
+ "pass@32": 0.8952482473128105,
1750
+ "pass@64": 0.9085697858530525,
1751
+ "pass@128": 0.9176470588235294
1752
+ },
1753
+ "6": {
1754
+ "pass@1": 0.6679071372239748,
1755
+ "pass@2": 0.7301391459052637,
1756
+ "pass@4": 0.7808355910361361,
1757
+ "pass@8": 0.818929674856254,
1758
+ "pass@16": 0.8450057327852958,
1759
+ "pass@32": 0.8632534413631587,
1760
+ "pass@64": 0.8793223643701782,
1761
+ "pass@128": 0.8958990536277602
1762
+ },
1763
+ "7": {
1764
+ "pass@1": 0.6406497231012658,
1765
+ "pass@2": 0.71024486382438,
1766
+ "pass@4": 0.7654189415998627,
1767
+ "pass@8": 0.8044126888367252,
1768
+ "pass@16": 0.8300388065663724,
1769
+ "pass@32": 0.8493341982733007,
1770
+ "pass@64": 0.8669889661389816,
1771
+ "pass@128": 0.8829113924050633
1772
+ },
1773
+ "8": {
1774
+ "pass@1": 0.6278856981981982,
1775
+ "pass@2": 0.7077998273864413,
1776
+ "pass@4": 0.7638303280783592,
1777
+ "pass@8": 0.8036424586250549,
1778
+ "pass@16": 0.8327328899737727,
1779
+ "pass@32": 0.8545952441143289,
1780
+ "pass@64": 0.8720383615804834,
1781
+ "pass@128": 0.8888888888888888
1782
+ },
1783
+ "9": {
1784
+ "pass@1": 0.6134445337620579,
1785
+ "pass@2": 0.7031613951692531,
1786
+ "pass@4": 0.7679856327243657,
1787
+ "pass@8": 0.8137176128360283,
1788
+ "pass@16": 0.8453081708297483,
1789
+ "pass@32": 0.8678143930156336,
1790
+ "pass@64": 0.8861791073478019,
1791
+ "pass@128": 0.9035369774919614
1792
+ },
1793
+ "11": {
1794
+ "pass@1": 0.533779761904762,
1795
+ "pass@2": 0.637331271091113,
1796
+ "pass@4": 0.7189003565030557,
1797
+ "pass@8": 0.7825376796716175,
1798
+ "pass@16": 0.8307175370206826,
1799
+ "pass@32": 0.8640638671817454,
1800
+ "pass@64": 0.8861393273804471,
1801
+ "pass@128": 0.9047619047619048
1802
+ },
1803
+ "12": {
1804
+ "pass@1": 0.519447629969419,
1805
+ "pass@2": 0.6325109411495584,
1806
+ "pass@4": 0.716240447868328,
1807
+ "pass@8": 0.7778574011864423,
1808
+ "pass@16": 0.8239268041149975,
1809
+ "pass@32": 0.8611130957104958,
1810
+ "pass@64": 0.8914035041714263,
1811
+ "pass@128": 0.9113149847094801
1812
+ },
1813
+ "13": {
1814
+ "pass@1": 0.4123022762345679,
1815
+ "pass@2": 0.5496697895401961,
1816
+ "pass@4": 0.6635212849551213,
1817
+ "pass@8": 0.7482771208219928,
1818
+ "pass@16": 0.8095717661613954,
1819
+ "pass@32": 0.8546775092104458,
1820
+ "pass@64": 0.8861848341058806,
1821
+ "pass@128": 0.9074074074074074
1822
+ },
1823
+ "14": {
1824
+ "pass@1": 0.3145168768545994,
1825
+ "pass@2": 0.4401428041543027,
1826
+ "pass@4": 0.5588902007612545,
1827
+ "pass@8": 0.6583031703543211,
1828
+ "pass@16": 0.7376897670615584,
1829
+ "pass@32": 0.7999922261319541,
1830
+ "pass@64": 0.8459428217294271,
1831
+ "pass@128": 0.8813056379821959
1832
+ },
1833
+ "15": {
1834
+ "pass@1": 0.23210841049382716,
1835
+ "pass@2": 0.3398600782541072,
1836
+ "pass@4": 0.4533102031227575,
1837
+ "pass@8": 0.5593127461017794,
1838
+ "pass@16": 0.6496853742117522,
1839
+ "pass@32": 0.7226145759565928,
1840
+ "pass@64": 0.778493587814901,
1841
+ "pass@128": 0.8179012345679012
1842
+ },
1843
+ "16": {
1844
+ "pass@1": 0.21082227138643067,
1845
+ "pass@2": 0.3154901371565279,
1846
+ "pass@4": 0.4299715937056539,
1847
+ "pass@8": 0.5375134403941474,
1848
+ "pass@16": 0.6273091818412996,
1849
+ "pass@32": 0.6999977456346922,
1850
+ "pass@64": 0.760873083209445,
1851
+ "pass@128": 0.8112094395280236
1852
+ },
1853
+ "17": {
1854
+ "pass@1": 0.16036803463855423,
1855
+ "pass@2": 0.24516397282041513,
1856
+ "pass@4": 0.34268070369215886,
1857
+ "pass@8": 0.43866697297296425,
1858
+ "pass@16": 0.5218814653519248,
1859
+ "pass@32": 0.591200169404961,
1860
+ "pass@64": 0.6489296836444958,
1861
+ "pass@128": 0.6987951807228916
1862
+ },
1863
+ "18": {
1864
+ "pass@1": 0.16050907258064517,
1865
+ "pass@2": 0.23742379984759968,
1866
+ "pass@4": 0.32174134160649276,
1867
+ "pass@8": 0.4042682761697166,
1868
+ "pass@16": 0.48119832780048005,
1869
+ "pass@32": 0.5537932896201498,
1870
+ "pass@64": 0.6174600257015876,
1871
+ "pass@128": 0.6709677419354839
1872
+ },
1873
+ "19": {
1874
+ "pass@1": 0.1531344410876133,
1875
+ "pass@2": 0.22887798249161445,
1876
+ "pass@4": 0.3099796786164572,
1877
+ "pass@8": 0.38504632696140717,
1878
+ "pass@16": 0.44934495744307373,
1879
+ "pass@32": 0.503410042848671,
1880
+ "pass@64": 0.5509170785946527,
1881
+ "pass@128": 0.5921450151057401
1882
+ },
1883
+ "20": {
1884
+ "pass@1": 0.14949324324324326,
1885
+ "pass@2": 0.2258773291007543,
1886
+ "pass@4": 0.30744521067749414,
1887
+ "pass@8": 0.3811890808222759,
1888
+ "pass@16": 0.4431254983049504,
1889
+ "pass@32": 0.49655440542341023,
1890
+ "pass@64": 0.5479183432668487,
1891
+ "pass@128": 0.6066066066066066
1892
+ }
1893
+ }
1894
+ }
1895
+ }
1896
+ }
1897
+ }