ZachW commited on
Commit
ce791f5
·
verified ·
1 Parent(s): 4f37c12

Upload extrapolation artifact: 020305 RL OP13-20 step200 metrics

Browse files
020305/evals/rl_ckpt18809_op13_20_step200_merged_hf/rl_ckpt18809_op13_20_step200_merged_hf_metrics.json ADDED
@@ -0,0 +1,1897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "rl_ckpt18809_op13_20_step200_merged_hf",
3
+ "id": {
4
+ "answer_accuracy": 0.6831440972222222,
5
+ "avg_response_len": 199.82384895833334,
6
+ "avg_loss": 0.0,
7
+ "count": 1152000,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {
10
+ "pass@1": 0.6831440972222222,
11
+ "pass@2": 0.7378728811242216,
12
+ "pass@4": 0.7780918725575918,
13
+ "pass@8": 0.8082755057779146,
14
+ "pass@16": 0.8315360346541011,
15
+ "pass@32": 0.8505286915169447,
16
+ "pass@64": 0.8671238823872478,
17
+ "pass@128": 0.8822222222222222
18
+ },
19
+ "per_op_pass_at_k": {
20
+ "10": {
21
+ "pass@1": 0.59,
22
+ "pass@2": 0.6525396161417322,
23
+ "pass@4": 0.7017073478627673,
24
+ "pass@8": 0.7441872508821505,
25
+ "pass@16": 0.7811828115511297,
26
+ "pass@32": 0.8124735924334449,
27
+ "pass@64": 0.8389039204254533,
28
+ "pass@128": 0.861
29
+ },
30
+ "2": {
31
+ "pass@1": 0.9444296875,
32
+ "pass@2": 0.9486743356299212,
33
+ "pass@4": 0.9518330838957629,
34
+ "pass@8": 0.9547528882426679,
35
+ "pass@16": 0.9577601684119486,
36
+ "pass@32": 0.9610236105812119,
37
+ "pass@64": 0.9647495568226766,
38
+ "pass@128": 0.969
39
+ },
40
+ "3": {
41
+ "pass@1": 0.9078359375,
42
+ "pass@2": 0.9258110236220475,
43
+ "pass@4": 0.9359801903824524,
44
+ "pass@8": 0.942389475610652,
45
+ "pass@16": 0.9468214696981089,
46
+ "pass@32": 0.9505577533809155,
47
+ "pass@64": 0.9545659426082578,
48
+ "pass@128": 0.959
49
+ },
50
+ "4": {
51
+ "pass@1": 0.77075,
52
+ "pass@2": 0.8123546998031498,
53
+ "pass@4": 0.8433460299025114,
54
+ "pass@8": 0.8672563836341415,
55
+ "pass@16": 0.8841498615548093,
56
+ "pass@32": 0.8960136944828668,
57
+ "pass@64": 0.9053688586941233,
58
+ "pass@128": 0.913
59
+ },
60
+ "5": {
61
+ "pass@1": 0.613390625,
62
+ "pass@2": 0.6642330216535437,
63
+ "pass@4": 0.7034845461192348,
64
+ "pass@8": 0.7352774295302196,
65
+ "pass@16": 0.7615447053080375,
66
+ "pass@32": 0.7830960903484209,
67
+ "pass@64": 0.8015853923107826,
68
+ "pass@128": 0.821
69
+ },
70
+ "6": {
71
+ "pass@1": 0.618515625,
72
+ "pass@2": 0.693633243110236,
73
+ "pass@4": 0.7454748279902509,
74
+ "pass@8": 0.7795075821092589,
75
+ "pass@16": 0.8025615674056414,
76
+ "pass@32": 0.8210030197709425,
77
+ "pass@64": 0.8383984335616581,
78
+ "pass@128": 0.856
79
+ },
80
+ "7": {
81
+ "pass@1": 0.6011484375,
82
+ "pass@2": 0.6669009596456699,
83
+ "pass@4": 0.7120814229471315,
84
+ "pass@8": 0.7453713718484262,
85
+ "pass@16": 0.7730127547498042,
86
+ "pass@32": 0.7972900605991194,
87
+ "pass@64": 0.8194086339543623,
88
+ "pass@128": 0.84
89
+ },
90
+ "8": {
91
+ "pass@1": 0.5643671875,
92
+ "pass@2": 0.644838336614174,
93
+ "pass@4": 0.705028264998125,
94
+ "pass@8": 0.7487015677656571,
95
+ "pass@16": 0.7801481836227231,
96
+ "pass@32": 0.8052043841885894,
97
+ "pass@64": 0.8275029223931598,
98
+ "pass@128": 0.846
99
+ },
100
+ "9": {
101
+ "pass@1": 0.537859375,
102
+ "pass@2": 0.6318706938976384,
103
+ "pass@4": 0.7038911389201337,
104
+ "pass@8": 0.7570356023780919,
105
+ "pass@16": 0.7966427895846567,
106
+ "pass@32": 0.8280960178669846,
107
+ "pass@64": 0.8536312807147408,
108
+ "pass@128": 0.875
109
+ }
110
+ },
111
+ "per_op_accuracy": {
112
+ "10": 0.59,
113
+ "2": 0.9444296875,
114
+ "3": 0.9078359375,
115
+ "4": 0.77075,
116
+ "5": 0.613390625,
117
+ "6": 0.618515625,
118
+ "7": 0.6011484375,
119
+ "8": 0.5643671875,
120
+ "9": 0.537859375
121
+ },
122
+ "per_op_avg_response_len": {
123
+ "10": 272.9003203125,
124
+ "2": 117.993890625,
125
+ "3": 149.4579921875,
126
+ "4": 161.6555234375,
127
+ "5": 172.9878203125,
128
+ "6": 202.7284140625,
129
+ "7": 223.0876328125,
130
+ "8": 241.644890625,
131
+ "9": 255.95815625
132
+ },
133
+ "per_op_avg_loss": {},
134
+ "per_op_length": {
135
+ "10": 272.9003203125,
136
+ "2": 117.993890625,
137
+ "3": 149.4579921875,
138
+ "4": 161.6555234375,
139
+ "5": 172.9878203125,
140
+ "6": 202.7284140625,
141
+ "7": 223.0876328125,
142
+ "8": 241.644890625,
143
+ "9": 255.95815625
144
+ },
145
+ "per_op_loss": {},
146
+ "per_template": {
147
+ "crazy_zootopia": {
148
+ "count": 379008,
149
+ "correct": 256789,
150
+ "answer_accuracy": 0.6775292342114151,
151
+ "avg_response_len": 190.6563951156704,
152
+ "resp_tokens_sum": 72260299,
153
+ "pass_at_k": {
154
+ "pass@1": 0.6775292342114151,
155
+ "pass@2": 0.7369279970322881,
156
+ "pass@4": 0.779272400980275,
157
+ "pass@8": 0.8100290305523985,
158
+ "pass@16": 0.8335809339351438,
159
+ "pass@32": 0.8530711337098859,
160
+ "pass@64": 0.8708661645407495,
161
+ "pass@128": 0.8872002701789936
162
+ },
163
+ "per_op_pass_at_k": {
164
+ "10": {
165
+ "pass@1": 0.5741380899705014,
166
+ "pass@2": 0.6374345283720061,
167
+ "pass@4": 0.6877484625815574,
168
+ "pass@8": 0.7339820506256319,
169
+ "pass@16": 0.7768820473639336,
170
+ "pass@32": 0.8134513625208626,
171
+ "pass@64": 0.8442222274966321,
172
+ "pass@128": 0.8672566371681416
173
+ },
174
+ "2": {
175
+ "pass@1": 0.9447770638629284,
176
+ "pass@2": 0.9496820344886795,
177
+ "pass@4": 0.9534258314089241,
178
+ "pass@8": 0.9564988072518206,
179
+ "pass@16": 0.9596533365254536,
180
+ "pass@32": 0.9640128309589705,
181
+ "pass@64": 0.9696471858423932,
182
+ "pass@128": 0.9750778816199377
183
+ },
184
+ "3": {
185
+ "pass@1": 0.890075496941896,
186
+ "pass@2": 0.9153731428640232,
187
+ "pass@4": 0.9288290752646745,
188
+ "pass@8": 0.9363709567903885,
189
+ "pass@16": 0.9418395773507783,
190
+ "pass@32": 0.9473595562005767,
191
+ "pass@64": 0.9545467492115126,
192
+ "pass@128": 0.963302752293578
193
+ },
194
+ "4": {
195
+ "pass@1": 0.7752197265625,
196
+ "pass@2": 0.8222317913385826,
197
+ "pass@4": 0.8546887019005436,
198
+ "pass@8": 0.8798078050264433,
199
+ "pass@16": 0.8987486358045697,
200
+ "pass@32": 0.9141457607856986,
201
+ "pass@64": 0.9275849795438009,
202
+ "pass@128": 0.940625
203
+ },
204
+ "5": {
205
+ "pass@1": 0.5919642857142857,
206
+ "pass@2": 0.6483459098862642,
207
+ "pass@4": 0.6903727599526247,
208
+ "pass@8": 0.721324081648959,
209
+ "pass@16": 0.7452422640151924,
210
+ "pass@32": 0.7649371701462218,
211
+ "pass@64": 0.784241095420447,
212
+ "pass@128": 0.8063492063492064
213
+ },
214
+ "6": {
215
+ "pass@1": 0.6101936751497006,
216
+ "pass@2": 0.6852625948889624,
217
+ "pass@4": 0.737672784633208,
218
+ "pass@8": 0.772938922756174,
219
+ "pass@16": 0.798958840272014,
220
+ "pass@32": 0.8201741899475089,
221
+ "pass@64": 0.8383901021445377,
222
+ "pass@128": 0.8532934131736527
223
+ },
224
+ "7": {
225
+ "pass@1": 0.6123104659763313,
226
+ "pass@2": 0.6833205691189485,
227
+ "pass@4": 0.7293244864362364,
228
+ "pass@8": 0.7616322923990702,
229
+ "pass@16": 0.7878927447570453,
230
+ "pass@32": 0.8115178358283376,
231
+ "pass@64": 0.8346118917446502,
232
+ "pass@128": 0.8579881656804734
233
+ },
234
+ "8": {
235
+ "pass@1": 0.5609475160256411,
236
+ "pass@2": 0.6446952919947505,
237
+ "pass@4": 0.7031014176112595,
238
+ "pass@8": 0.7419600517349403,
239
+ "pass@16": 0.7673220278771473,
240
+ "pass@32": 0.7887273838182889,
241
+ "pass@64": 0.8099674098488452,
242
+ "pass@128": 0.8269230769230769
243
+ },
244
+ "9": {
245
+ "pass@1": 0.5545994718309859,
246
+ "pass@2": 0.6576068121326386,
247
+ "pass@4": 0.7359780795006251,
248
+ "pass@8": 0.790439971280655,
249
+ "pass@16": 0.8279563910100225,
250
+ "pass@32": 0.8539518807147877,
251
+ "pass@64": 0.8742415789463832,
252
+ "pass@128": 0.8929577464788733
253
+ }
254
+ }
255
+ },
256
+ "teachers_in_school": {
257
+ "count": 390656,
258
+ "correct": 263955,
259
+ "answer_accuracy": 0.6756711787352556,
260
+ "avg_response_len": 197.56889693233944,
261
+ "resp_tokens_sum": 77181475,
262
+ "pass_at_k": {
263
+ "pass@1": 0.6756711787352556,
264
+ "pass@2": 0.7305195428194783,
265
+ "pass@4": 0.7712560914834049,
266
+ "pass@8": 0.8021417833557318,
267
+ "pass@16": 0.8261166997131413,
268
+ "pass@32": 0.8455816770569933,
269
+ "pass@64": 0.8619696745039069,
270
+ "pass@128": 0.8764744429882044
271
+ },
272
+ "per_op_pass_at_k": {
273
+ "10": {
274
+ "pass@1": 0.6025929831288344,
275
+ "pass@2": 0.6655788820588381,
276
+ "pass@4": 0.712638290708293,
277
+ "pass@8": 0.7520765879868826,
278
+ "pass@16": 0.7863821934536588,
279
+ "pass@32": 0.8153207950273516,
280
+ "pass@64": 0.8387030964952171,
281
+ "pass@128": 0.8588957055214724
282
+ },
283
+ "2": {
284
+ "pass@1": 0.9505841565349544,
285
+ "pass@2": 0.9532843052198261,
286
+ "pass@4": 0.9558882856208325,
287
+ "pass@8": 0.9587953989382784,
288
+ "pass@16": 0.961782530755302,
289
+ "pass@32": 0.9643643890269245,
290
+ "pass@64": 0.9665359573547208,
291
+ "pass@128": 0.9696048632218845
292
+ },
293
+ "3": {
294
+ "pass@1": 0.9034327651515152,
295
+ "pass@2": 0.9192156555714627,
296
+ "pass@4": 0.9302861139516647,
297
+ "pass@8": 0.9386739269294071,
298
+ "pass@16": 0.9438559355563287,
299
+ "pass@32": 0.9468557193564537,
300
+ "pass@64": 0.9492044725881807,
301
+ "pass@128": 0.9515151515151515
302
+ },
303
+ "4": {
304
+ "pass@1": 0.7489367603550295,
305
+ "pass@2": 0.7910608518147509,
306
+ "pass@4": 0.8231998353201413,
307
+ "pass@8": 0.8477757288644923,
308
+ "pass@16": 0.8653097885880157,
309
+ "pass@32": 0.877466942076976,
310
+ "pass@64": 0.8870172256269219,
311
+ "pass@128": 0.893491124260355
312
+ },
313
+ "5": {
314
+ "pass@1": 0.6153532608695652,
315
+ "pass@2": 0.6656110920917495,
316
+ "pass@4": 0.7047145457361306,
317
+ "pass@8": 0.734960864574528,
318
+ "pass@16": 0.7579012844461602,
319
+ "pass@32": 0.7753412883276696,
320
+ "pass@64": 0.7880631576774196,
321
+ "pass@128": 0.8
322
+ },
323
+ "6": {
324
+ "pass@1": 0.6086138968481375,
325
+ "pass@2": 0.6848056454887982,
326
+ "pass@4": 0.7375910427027563,
327
+ "pass@8": 0.7741764834993599,
328
+ "pass@16": 0.8002319395512839,
329
+ "pass@32": 0.8218286879890876,
330
+ "pass@64": 0.8432342376670118,
331
+ "pass@128": 0.8653295128939829
332
+ },
333
+ "7": {
334
+ "pass@1": 0.603888186416185,
335
+ "pass@2": 0.6690694551909333,
336
+ "pass@4": 0.7135883215103886,
337
+ "pass@8": 0.7469629249550308,
338
+ "pass@16": 0.7763369658056625,
339
+ "pass@32": 0.8021818464035171,
340
+ "pass@64": 0.8246229458789898,
341
+ "pass@128": 0.8439306358381503
342
+ },
343
+ "8": {
344
+ "pass@1": 0.5481294014084507,
345
+ "pass@2": 0.6296779693911498,
346
+ "pass@4": 0.6941487652951831,
347
+ "pass@8": 0.7432917081970493,
348
+ "pass@16": 0.7800826825934518,
349
+ "pass@32": 0.8083199455882679,
350
+ "pass@64": 0.8310643932196761,
351
+ "pass@128": 0.8507042253521127
352
+ },
353
+ "9": {
354
+ "pass@1": 0.5193207335329342,
355
+ "pass@2": 0.6124250023574895,
356
+ "pass@4": 0.6825153814979706,
357
+ "pass@8": 0.7339480753308065,
358
+ "pass@16": 0.7732788025060358,
359
+ "pass@32": 0.8077219918528098,
360
+ "pass@64": 0.8375948255424386,
361
+ "pass@128": 0.8622754491017964
362
+ }
363
+ }
364
+ },
365
+ "movie_festival_awards": {
366
+ "count": 382336,
367
+ "correct": 266238,
368
+ "answer_accuracy": 0.6963456226983595,
369
+ "avg_response_len": 211.21552770338133,
370
+ "resp_tokens_sum": 80755300,
371
+ "pass_at_k": {
372
+ "pass@1": 0.6963456226983595,
373
+ "pass@2": 0.7463228946168333,
374
+ "pass@4": 0.7839061541708885,
375
+ "pass@8": 0.8128044424954579,
376
+ "pass@16": 0.8350461998595287,
377
+ "pass@32": 0.8530630459857931,
378
+ "pass@64": 0.8686805429173466,
379
+ "pass@128": 0.8831603615667895
380
+ },
381
+ "per_op_pass_at_k": {
382
+ "10": {
383
+ "pass@1": 0.5937966417910447,
384
+ "pass@2": 0.6551361058878838,
385
+ "pass@4": 0.7051956306767619,
386
+ "pass@8": 0.7468369195114555,
387
+ "pass@16": 0.7804752311309393,
388
+ "pass@32": 0.8087134368953917,
389
+ "pass@64": 0.8337175398407523,
390
+ "pass@128": 0.8567164179104477
391
+ },
392
+ "2": {
393
+ "pass@1": 0.9383258928571429,
394
+ "pass@2": 0.9434167604049495,
395
+ "pass@4": 0.94656041726927,
396
+ "pass@8": 0.9493516710403994,
397
+ "pass@16": 0.9522428421965253,
398
+ "pass@32": 0.9551417367243559,
399
+ "pass@64": 0.9585785148501862,
400
+ "pass@128": 0.9628571428571429
401
+ },
402
+ "3": {
403
+ "pass@1": 0.9290041909620991,
404
+ "pass@2": 0.9421074039852161,
405
+ "pass@4": 0.9482759917400994,
406
+ "pass@8": 0.9517019733337934,
407
+ "pass@16": 0.9544241083697254,
408
+ "pass@32": 0.9571684872177765,
409
+ "pass@64": 0.9597425063031881,
410
+ "pass@128": 0.9620991253644315
411
+ },
412
+ "4": {
413
+ "pass@1": 0.78812591374269,
414
+ "pass@2": 0.8241577738637934,
415
+ "pass@4": 0.852643570047165,
416
+ "pass@8": 0.8747651744721673,
417
+ "pass@16": 0.8891099109783571,
418
+ "pass@32": 0.8973778497351624,
419
+ "pass@64": 0.9027188391175657,
420
+ "pass@128": 0.9064327485380117
421
+ },
422
+ "5": {
423
+ "pass@1": 0.63125,
424
+ "pass@2": 0.6775536272579896,
425
+ "pass@4": 0.7143841425152739,
426
+ "pass@8": 0.7485260162723093,
427
+ "pass@16": 0.7803454382627247,
428
+ "pass@32": 0.807788639056809,
429
+ "pass@64": 0.8313754642783288,
430
+ "pass@128": 0.8558823529411764
431
+ },
432
+ "6": {
433
+ "pass@1": 0.6381851340694006,
434
+ "pass@2": 0.7121715020243922,
435
+ "pass@4": 0.762374902269708,
436
+ "pass@8": 0.7922977576259308,
437
+ "pass@16": 0.8089222960611702,
438
+ "pass@32": 0.8209672814519965,
439
+ "pass@64": 0.8330832507873658,
440
+ "pass@128": 0.8485804416403786
441
+ },
442
+ "7": {
443
+ "pass@1": 0.586209454113924,
444
+ "pass@2": 0.6469638474284858,
445
+ "pass@4": 0.6919879344591413,
446
+ "pass@8": 0.7262357119718345,
447
+ "pass@16": 0.7534570153581185,
448
+ "pass@32": 0.7767155482073544,
449
+ "pass@64": 0.7974375799700597,
450
+ "pass@128": 0.8164556962025317
451
+ },
452
+ "8": {
453
+ "pass@1": 0.5848817567567568,
454
+ "pass@2": 0.6611343134473057,
455
+ "pass@4": 0.7184318649358015,
456
+ "pass@8": 0.760785210854063,
457
+ "pass@16": 0.792235311124261,
458
+ "pass@32": 0.817320900160503,
459
+ "pass@64": 0.840135828610614,
460
+ "pass@128": 0.8588588588588588
461
+ },
462
+ "9": {
463
+ "pass@1": 0.5386605707395499,
464
+ "pass@2": 0.623377250170899,
465
+ "pass@4": 0.690221168093538,
466
+ "pass@8": 0.7437001781767524,
467
+ "pass@16": 0.785990838389334,
468
+ "pass@32": 0.8204628776025612,
469
+ "pass@64": 0.8473274870019281,
470
+ "pass@128": 0.8681672025723473
471
+ }
472
+ }
473
+ }
474
+ }
475
+ },
476
+ "ood": {
477
+ "answer_accuracy": 0.28934609375,
478
+ "avg_response_len": 313.83307109375,
479
+ "avg_loss": 0.0,
480
+ "count": 1280000,
481
+ "validation_metrics": {},
482
+ "pass_at_k": {
483
+ "pass@1": 0.28934609375,
484
+ "pass@2": 0.37076151574803684,
485
+ "pass@4": 0.44599904675665564,
486
+ "pass@8": 0.5111028938533587,
487
+ "pass@16": 0.5651620745137119,
488
+ "pass@32": 0.6105293104827964,
489
+ "pass@64": 0.6501892863425639,
490
+ "pass@128": 0.6856
491
+ },
492
+ "per_op_pass_at_k": {
493
+ "11": {
494
+ "pass@1": 0.55396875,
495
+ "pass@2": 0.6387955216535438,
496
+ "pass@4": 0.7043307344394445,
497
+ "pass@8": 0.7546983339977341,
498
+ "pass@16": 0.7929470653592272,
499
+ "pass@32": 0.8232224510293635,
500
+ "pass@64": 0.8489851562729089,
501
+ "pass@128": 0.87
502
+ },
503
+ "12": {
504
+ "pass@1": 0.548125,
505
+ "pass@2": 0.6337396653543307,
506
+ "pass@4": 0.6963535999250092,
507
+ "pass@8": 0.7450771226331848,
508
+ "pass@16": 0.7842876873708078,
509
+ "pass@32": 0.8163879483066379,
510
+ "pass@64": 0.8437814099411024,
511
+ "pass@128": 0.867
512
+ },
513
+ "13": {
514
+ "pass@1": 0.4446015625,
515
+ "pass@2": 0.5573118848425194,
516
+ "pass@4": 0.6469788156167974,
517
+ "pass@8": 0.7153580123659391,
518
+ "pass@16": 0.7666650305593797,
519
+ "pass@32": 0.8060510867048352,
520
+ "pass@64": 0.8375774928020564,
521
+ "pass@128": 0.861
522
+ },
523
+ "14": {
524
+ "pass@1": 0.337375,
525
+ "pass@2": 0.44927989665354384,
526
+ "pass@4": 0.5528005280277467,
527
+ "pass@8": 0.639973979765343,
528
+ "pass@16": 0.7076267967485103,
529
+ "pass@32": 0.7591571835994034,
530
+ "pass@64": 0.7992540700872042,
531
+ "pass@128": 0.831
532
+ },
533
+ "15": {
534
+ "pass@1": 0.23525,
535
+ "pass@2": 0.3350001230314964,
536
+ "pass@4": 0.43675789576302876,
537
+ "pass@8": 0.5272547912266787,
538
+ "pass@16": 0.6012428286892009,
539
+ "pass@32": 0.6610689530008375,
540
+ "pass@64": 0.7108299522498246,
541
+ "pass@128": 0.753
542
+ },
543
+ "16": {
544
+ "pass@1": 0.17765625,
545
+ "pass@2": 0.2467701771653543,
546
+ "pass@4": 0.31910015944881853,
547
+ "pass@8": 0.38902820548262157,
548
+ "pass@16": 0.45373992733889545,
549
+ "pass@32": 0.5124913954015271,
550
+ "pass@64": 0.5660727186562804,
551
+ "pass@128": 0.614
552
+ },
553
+ "17": {
554
+ "pass@1": 0.1604609375,
555
+ "pass@2": 0.2241515748031496,
556
+ "pass@4": 0.29066624925009377,
557
+ "pass@8": 0.3551817123763024,
558
+ "pass@16": 0.41555541132674007,
559
+ "pass@32": 0.47172652564446815,
560
+ "pass@64": 0.5251010037440795,
561
+ "pass@128": 0.578
562
+ },
563
+ "18": {
564
+ "pass@1": 0.15584375,
565
+ "pass@2": 0.22099729330708662,
566
+ "pass@4": 0.2880469713160855,
567
+ "pass@8": 0.34985024066375137,
568
+ "pass@16": 0.4038654648619413,
569
+ "pass@32": 0.4512529431204043,
570
+ "pass@64": 0.494365624612651,
571
+ "pass@128": 0.539
572
+ },
573
+ "19": {
574
+ "pass@1": 0.137875,
575
+ "pass@2": 0.1946306594488188,
576
+ "pass@4": 0.2525709119797523,
577
+ "pass@8": 0.30590235229761004,
578
+ "pass@16": 0.3528854382823781,
579
+ "pass@32": 0.39543239658733015,
580
+ "pass@64": 0.43542615681957186,
581
+ "pass@128": 0.472
582
+ },
583
+ "20": {
584
+ "pass@1": 0.1423046875,
585
+ "pass@2": 0.2069383612204721,
586
+ "pass@4": 0.27238460179977464,
587
+ "pass@8": 0.3287041877243397,
588
+ "pass@16": 0.37280509459990285,
589
+ "pass@32": 0.4085022214331564,
590
+ "pass@64": 0.4404992782399348,
591
+ "pass@128": 0.471
592
+ }
593
+ },
594
+ "per_op_accuracy": {
595
+ "11": 0.55396875,
596
+ "12": 0.548125,
597
+ "13": 0.4446015625,
598
+ "14": 0.337375,
599
+ "15": 0.23525,
600
+ "16": 0.17765625,
601
+ "17": 0.1604609375,
602
+ "18": 0.15584375,
603
+ "19": 0.137875,
604
+ "20": 0.1423046875
605
+ },
606
+ "per_op_avg_response_len": {
607
+ "11": 290.2597421875,
608
+ "12": 296.8439921875,
609
+ "13": 307.613546875,
610
+ "14": 320.121765625,
611
+ "15": 321.7593515625,
612
+ "16": 320.8959921875,
613
+ "17": 322.09896875,
614
+ "18": 320.365109375,
615
+ "19": 314.7880625,
616
+ "20": 323.5841796875
617
+ },
618
+ "per_op_avg_loss": {},
619
+ "per_op_length": {
620
+ "11": 290.2597421875,
621
+ "12": 296.8439921875,
622
+ "13": 307.613546875,
623
+ "14": 320.121765625,
624
+ "15": 321.7593515625,
625
+ "16": 320.8959921875,
626
+ "17": 322.09896875,
627
+ "18": 320.365109375,
628
+ "19": 314.7880625,
629
+ "20": 323.5841796875
630
+ },
631
+ "per_op_loss": {},
632
+ "per_template": {
633
+ "crazy_zootopia": {
634
+ "count": 430720,
635
+ "correct": 124452,
636
+ "answer_accuracy": 0.2889394502228826,
637
+ "avg_response_len": 299.6257243684993,
638
+ "resp_tokens_sum": 129054792,
639
+ "pass_at_k": {
640
+ "pass@1": 0.2889394502228826,
641
+ "pass@2": 0.3705931324659835,
642
+ "pass@4": 0.4461128146221719,
643
+ "pass@8": 0.5112165872753024,
644
+ "pass@16": 0.5652476971118939,
645
+ "pass@32": 0.6111237933147466,
646
+ "pass@64": 0.6515956181551907,
647
+ "pass@128": 0.687369985141159
648
+ },
649
+ "per_op_pass_at_k": {
650
+ "11": {
651
+ "pass@1": 0.5319360835694051,
652
+ "pass@2": 0.6215784696973076,
653
+ "pass@4": 0.6907830577480926,
654
+ "pass@8": 0.7432213329841176,
655
+ "pass@16": 0.7828037516289943,
656
+ "pass@32": 0.8150725689730364,
657
+ "pass@64": 0.8445493457746914,
658
+ "pass@128": 0.8725212464589235
659
+ },
660
+ "12": {
661
+ "pass@1": 0.519332627118644,
662
+ "pass@2": 0.6027417256105698,
663
+ "pass@4": 0.6637228021285474,
664
+ "pass@8": 0.7114096130410679,
665
+ "pass@16": 0.7524092343759183,
666
+ "pass@32": 0.7890026535707286,
667
+ "pass@64": 0.8212713564382537,
668
+ "pass@128": 0.847457627118644
669
+ },
670
+ "13": {
671
+ "pass@1": 0.4448961598746082,
672
+ "pass@2": 0.5568930034803646,
673
+ "pass@4": 0.6451671588661135,
674
+ "pass@8": 0.7110854193688476,
675
+ "pass@16": 0.7608013473487977,
676
+ "pass@32": 0.8014909701103228,
677
+ "pass@64": 0.8384952651641907,
678
+ "pass@128": 0.8683385579937304
679
+ },
680
+ "14": {
681
+ "pass@1": 0.34197332554517135,
682
+ "pass@2": 0.4501220349792721,
683
+ "pass@4": 0.5494017024157026,
684
+ "pass@8": 0.6321989976492106,
685
+ "pass@16": 0.6955289358403063,
686
+ "pass@32": 0.7459415528358541,
687
+ "pass@64": 0.7887735797886692,
688
+ "pass@128": 0.822429906542056
689
+ },
690
+ "15": {
691
+ "pass@1": 0.23803287572254336,
692
+ "pass@2": 0.33669275692958894,
693
+ "pass@4": 0.44012880236068747,
694
+ "pass@8": 0.5341238336228396,
695
+ "pass@16": 0.6110083221314118,
696
+ "pass@32": 0.6745370889577138,
697
+ "pass@64": 0.7290513406818961,
698
+ "pass@128": 0.7745664739884393
699
+ },
700
+ "16": {
701
+ "pass@1": 0.20163810483870967,
702
+ "pass@2": 0.2788862077724155,
703
+ "pass@4": 0.3585027432054865,
704
+ "pass@8": 0.43076881842449627,
705
+ "pass@16": 0.4919829537899595,
706
+ "pass@32": 0.5441670943646694,
707
+ "pass@64": 0.5902012441068575,
708
+ "pass@128": 0.6290322580645161
709
+ },
710
+ "17": {
711
+ "pass@1": 0.156882225433526,
712
+ "pass@2": 0.21556575997906322,
713
+ "pass@4": 0.2771691409780425,
714
+ "pass@8": 0.341674925379861,
715
+ "pass@16": 0.4072087355948067,
716
+ "pass@32": 0.4670126441556912,
717
+ "pass@64": 0.5169342900561144,
718
+ "pass@128": 0.5606936416184971
719
+ },
720
+ "18": {
721
+ "pass@1": 0.14910714285714285,
722
+ "pass@2": 0.21308176321709787,
723
+ "pass@4": 0.280601617654936,
724
+ "pass@8": 0.34355741540217244,
725
+ "pass@16": 0.3985258710428493,
726
+ "pass@32": 0.4461388543379518,
727
+ "pass@64": 0.4872909465164243,
728
+ "pass@128": 0.5285714285714286
729
+ },
730
+ "19": {
731
+ "pass@1": 0.14057807807807807,
732
+ "pass@2": 0.20221685760563707,
733
+ "pass@4": 0.2649557811467261,
734
+ "pass@8": 0.3198520926064229,
735
+ "pass@16": 0.365459672897077,
736
+ "pass@32": 0.4061949642563818,
737
+ "pass@64": 0.44616338763451124,
738
+ "pass@128": 0.4864864864864865
739
+ },
740
+ "20": {
741
+ "pass@1": 0.15261355105105104,
742
+ "pass@2": 0.2182194497647254,
743
+ "pass@4": 0.2835954342981902,
744
+ "pass@8": 0.33920423135788597,
745
+ "pass@16": 0.3826292663379262,
746
+ "pass@32": 0.41802288691041456,
747
+ "pass@64": 0.4501180597103666,
748
+ "pass@128": 0.4804804804804805
749
+ }
750
+ }
751
+ },
752
+ "teachers_in_school": {
753
+ "count": 430464,
754
+ "correct": 121340,
755
+ "answer_accuracy": 0.2818818763009218,
756
+ "avg_response_len": 309.77385565343445,
757
+ "resp_tokens_sum": 133346493,
758
+ "pass_at_k": {
759
+ "pass@1": 0.2818818763009218,
760
+ "pass@2": 0.3627959194663561,
761
+ "pass@4": 0.4387880045881881,
762
+ "pass@8": 0.5053501116728005,
763
+ "pass@16": 0.5605613713414009,
764
+ "pass@32": 0.6061452941983058,
765
+ "pass@64": 0.6453566116191353,
766
+ "pass@128": 0.679155515908415
767
+ },
768
+ "per_op_pass_at_k": {
769
+ "11": {
770
+ "pass@1": 0.5750894201807228,
771
+ "pass@2": 0.6621766346883593,
772
+ "pass@4": 0.7284196917026932,
773
+ "pass@8": 0.7780922038504625,
774
+ "pass@16": 0.813405098694823,
775
+ "pass@32": 0.8392815462756775,
776
+ "pass@64": 0.8604137867079064,
777
+ "pass@128": 0.8765060240963856
778
+ },
779
+ "12": {
780
+ "pass@1": 0.5449647335423198,
781
+ "pass@2": 0.6343681349196552,
782
+ "pass@4": 0.7027254673103162,
783
+ "pass@8": 0.7573112125137871,
784
+ "pass@16": 0.7986891262756064,
785
+ "pass@32": 0.8283058253225973,
786
+ "pass@64": 0.8514306286603033,
787
+ "pass@128": 0.8714733542319749
788
+ },
789
+ "13": {
790
+ "pass@1": 0.41474089635854344,
791
+ "pass@2": 0.5216235608416595,
792
+ "pass@4": 0.609166683208716,
793
+ "pass@8": 0.6811120686409082,
794
+ "pass@16": 0.7386794715073752,
795
+ "pass@32": 0.7834822105451329,
796
+ "pass@64": 0.8176940029791813,
797
+ "pass@128": 0.8403361344537815
798
+ },
799
+ "14": {
800
+ "pass@1": 0.31992415935672514,
801
+ "pass@2": 0.4319049448588662,
802
+ "pass@4": 0.5384300460249487,
803
+ "pass@8": 0.6301912311847001,
804
+ "pass@16": 0.7053491688109808,
805
+ "pass@32": 0.7645782124873614,
806
+ "pass@64": 0.8085707748135609,
807
+ "pass@128": 0.8391812865497076
808
+ },
809
+ "15": {
810
+ "pass@1": 0.2126657196969697,
811
+ "pass@2": 0.30765442316869457,
812
+ "pass@4": 0.40781750747065687,
813
+ "pass@8": 0.49642400822735633,
814
+ "pass@16": 0.5680396411962745,
815
+ "pass@32": 0.626610399902073,
816
+ "pass@64": 0.676446573867005,
817
+ "pass@128": 0.7212121212121212
818
+ },
819
+ "16": {
820
+ "pass@1": 0.15422453703703703,
821
+ "pass@2": 0.2157730584157749,
822
+ "pass@4": 0.2825862648579184,
823
+ "pass@8": 0.350729859385697,
824
+ "pass@16": 0.41576288513575577,
825
+ "pass@32": 0.4756244133047699,
826
+ "pass@64": 0.5334865753499802,
827
+ "pass@128": 0.5868945868945868
828
+ },
829
+ "17": {
830
+ "pass@1": 0.1774310947204969,
831
+ "pass@2": 0.24994039467892606,
832
+ "pass@4": 0.32356567764045,
833
+ "pass@8": 0.39138325498864096,
834
+ "pass@16": 0.4512251459100479,
835
+ "pass@32": 0.5049210552230611,
836
+ "pass@64": 0.5575395956365649,
837
+ "pass@128": 0.6086956521739131
838
+ },
839
+ "18": {
840
+ "pass@1": 0.1604779411764706,
841
+ "pass@2": 0.2271639069013432,
842
+ "pass@4": 0.29349490863274436,
843
+ "pass@8": 0.35195918625723943,
844
+ "pass@16": 0.4028748773367787,
845
+ "pass@32": 0.44786918936674186,
846
+ "pass@64": 0.4874247416201932,
847
+ "pass@128": 0.5264705882352941
848
+ },
849
+ "19": {
850
+ "pass@1": 0.13846261160714285,
851
+ "pass@2": 0.19192473987626552,
852
+ "pass@4": 0.24761419108325744,
853
+ "pass@8": 0.3012326120456533,
854
+ "pass@16": 0.3485993171013056,
855
+ "pass@32": 0.38909940621048456,
856
+ "pass@64": 0.4246445731793099,
857
+ "pass@128": 0.45535714285714285
858
+ },
859
+ "20": {
860
+ "pass@1": 0.12930389221556887,
861
+ "pass@2": 0.1930537301381488,
862
+ "pass@4": 0.26072267753581696,
863
+ "pass@8": 0.32056687148160085,
864
+ "pass@16": 0.36709700781450916,
865
+ "pass@32": 0.4044418838960831,
866
+ "pass@64": 0.4377505306516329,
867
+ "pass@128": 0.46706586826347307
868
+ }
869
+ }
870
+ },
871
+ "movie_festival_awards": {
872
+ "count": 418816,
873
+ "correct": 124571,
874
+ "answer_accuracy": 0.29743610559290956,
875
+ "avg_response_len": 332.61634226008556,
876
+ "resp_tokens_sum": 139305046,
877
+ "pass_at_k": {
878
+ "pass@1": 0.29743610559290956,
879
+ "pass@2": 0.3791218183273016,
880
+ "pass@4": 0.4532936390381691,
881
+ "pass@8": 0.516898745964696,
882
+ "pass@16": 0.5698026749799292,
883
+ "pass@32": 0.6144238740632357,
884
+ "pass@64": 0.6537100621816129,
885
+ "pass@128": 0.690403422982885
886
+ },
887
+ "per_op_pass_at_k": {
888
+ "11": {
889
+ "pass@1": 0.5563988095238095,
890
+ "pass@2": 0.6334466004249469,
891
+ "pass@4": 0.6941237378065835,
892
+ "pass@8": 0.7429034659555144,
893
+ "pass@16": 0.7827518992619408,
894
+ "pass@32": 0.8154297168836708,
895
+ "pass@64": 0.8419106667537043,
896
+ "pass@128": 0.8603174603174604
897
+ },
898
+ "12": {
899
+ "pass@1": 0.5823776758409785,
900
+ "pass@2": 0.6666839738495992,
901
+ "pass@4": 0.7254627030566131,
902
+ "pass@8": 0.7695897334090223,
903
+ "pass@16": 0.8047491960850638,
904
+ "pass@32": 0.834408106008233,
905
+ "pass@64": 0.8606880098450249,
906
+ "pass@128": 0.8837920489296636
907
+ },
908
+ "13": {
909
+ "pass@1": 0.4772135416666667,
910
+ "pass@2": 0.5970475478759603,
911
+ "pass@4": 0.6904258828178885,
912
+ "pass@8": 0.7572986267977532,
913
+ "pass@16": 0.8032741648980943,
914
+ "pass@32": 0.8354083891204626,
915
+ "pass@64": 0.8585825436762666,
916
+ "pass@128": 0.8765432098765432
917
+ },
918
+ "14": {
919
+ "pass@1": 0.3507047477744807,
920
+ "pass@2": 0.4661104815533072,
921
+ "pass@4": 0.5706216789666868,
922
+ "pass@8": 0.6573077164830237,
923
+ "pass@16": 0.7214616991407049,
924
+ "pass@32": 0.7662439064344672,
925
+ "pass@64": 0.799782035575084,
926
+ "pass@128": 0.8308605341246291
927
+ },
928
+ "15": {
929
+ "pass@1": 0.2552806712962963,
930
+ "pass@2": 0.36104465891416354,
931
+ "pass@4": 0.4626344218546755,
932
+ "pass@8": 0.5513210557967552,
933
+ "pass@16": 0.6246323075214912,
934
+ "pass@32": 0.6817829884376061,
935
+ "pass@64": 0.7263914166598042,
936
+ "pass@128": 0.7623456790123457
937
+ },
938
+ "16": {
939
+ "pass@1": 0.17998709439528024,
940
+ "pass@2": 0.24949589749843215,
941
+ "pass@4": 0.3208747200294653,
942
+ "pass@8": 0.3905123632054533,
943
+ "pass@16": 0.45808979050548,
944
+ "pass@32": 0.5216974250103424,
945
+ "pass@64": 0.5777479204581456,
946
+ "pass@128": 0.6283185840707964
947
+ },
948
+ "17": {
949
+ "pass@1": 0.1477315512048193,
950
+ "pass@2": 0.20808739386680566,
951
+ "pass@4": 0.2728240309381508,
952
+ "pass@8": 0.33414692797749634,
953
+ "pass@16": 0.3896587103250072,
954
+ "pass@32": 0.4444444909179926,
955
+ "pass@64": 0.5021505710532824,
956
+ "pass@128": 0.5662650602409639
957
+ },
958
+ "18": {
959
+ "pass@1": 0.15836693548387096,
960
+ "pass@2": 0.22317079946659896,
961
+ "pass@4": 0.29047785871524134,
962
+ "pass@8": 0.3546420062759032,
963
+ "pass@16": 0.4109804893627079,
964
+ "pass@32": 0.4607381281207391,
965
+ "pass@64": 0.509965745745279,
966
+ "pass@128": 0.5645161290322581
967
+ },
968
+ "19": {
969
+ "pass@1": 0.1345591012084592,
970
+ "pass@2": 0.18974541772724032,
971
+ "pass@4": 0.24514280560096152,
972
+ "pass@8": 0.29660860366263314,
973
+ "pass@16": 0.3445860926332715,
974
+ "pass@32": 0.3910334531819703,
975
+ "pass@64": 0.4355684959185234,
976
+ "pass@128": 0.4743202416918429
977
+ },
978
+ "20": {
979
+ "pass@1": 0.14503566066066065,
980
+ "pass@2": 0.20958359934737886,
981
+ "pass@4": 0.27287071435890325,
982
+ "pass@8": 0.3263658967186459,
983
+ "pass@16": 0.3687061510490307,
984
+ "pass@32": 0.4030540866987888,
985
+ "pass@64": 0.43363749885506614,
986
+ "pass@128": 0.46546546546546547
987
+ }
988
+ }
989
+ }
990
+ }
991
+ },
992
+ "total": {
993
+ "answer_accuracy": 0.47588199013157895,
994
+ "avg_loss": 0.0,
995
+ "count": 2432000,
996
+ "validation_metrics": {
997
+ "id": {},
998
+ "ood": {}
999
+ },
1000
+ "pass_at_k": {
1001
+ "pass@1": 0.47588199013157895,
1002
+ "pass@2": 0.5446563730314754,
1003
+ "pass@4": 0.6033061747676053,
1004
+ "pass@8": 0.651868867922887,
1005
+ "pass@16": 0.6913392135275901,
1006
+ "pass@32": 0.7242132278147758,
1007
+ "pass@64": 0.7529477792058367,
1008
+ "pass@128": 0.7787368421052632
1009
+ },
1010
+ "per_op_pass_at_k": {
1011
+ "10": {
1012
+ "pass@1": 0.59,
1013
+ "pass@2": 0.6525396161417322,
1014
+ "pass@4": 0.7017073478627673,
1015
+ "pass@8": 0.7441872508821505,
1016
+ "pass@16": 0.7811828115511297,
1017
+ "pass@32": 0.8124735924334449,
1018
+ "pass@64": 0.8389039204254533,
1019
+ "pass@128": 0.861
1020
+ },
1021
+ "2": {
1022
+ "pass@1": 0.9444296875,
1023
+ "pass@2": 0.9486743356299212,
1024
+ "pass@4": 0.9518330838957629,
1025
+ "pass@8": 0.9547528882426679,
1026
+ "pass@16": 0.9577601684119486,
1027
+ "pass@32": 0.9610236105812119,
1028
+ "pass@64": 0.9647495568226766,
1029
+ "pass@128": 0.969
1030
+ },
1031
+ "3": {
1032
+ "pass@1": 0.9078359375,
1033
+ "pass@2": 0.9258110236220475,
1034
+ "pass@4": 0.9359801903824524,
1035
+ "pass@8": 0.942389475610652,
1036
+ "pass@16": 0.9468214696981089,
1037
+ "pass@32": 0.9505577533809155,
1038
+ "pass@64": 0.9545659426082578,
1039
+ "pass@128": 0.959
1040
+ },
1041
+ "4": {
1042
+ "pass@1": 0.77075,
1043
+ "pass@2": 0.8123546998031498,
1044
+ "pass@4": 0.8433460299025114,
1045
+ "pass@8": 0.8672563836341415,
1046
+ "pass@16": 0.8841498615548093,
1047
+ "pass@32": 0.8960136944828668,
1048
+ "pass@64": 0.9053688586941233,
1049
+ "pass@128": 0.913
1050
+ },
1051
+ "5": {
1052
+ "pass@1": 0.613390625,
1053
+ "pass@2": 0.6642330216535437,
1054
+ "pass@4": 0.7034845461192348,
1055
+ "pass@8": 0.7352774295302196,
1056
+ "pass@16": 0.7615447053080375,
1057
+ "pass@32": 0.7830960903484209,
1058
+ "pass@64": 0.8015853923107826,
1059
+ "pass@128": 0.821
1060
+ },
1061
+ "6": {
1062
+ "pass@1": 0.618515625,
1063
+ "pass@2": 0.693633243110236,
1064
+ "pass@4": 0.7454748279902509,
1065
+ "pass@8": 0.7795075821092589,
1066
+ "pass@16": 0.8025615674056414,
1067
+ "pass@32": 0.8210030197709425,
1068
+ "pass@64": 0.8383984335616581,
1069
+ "pass@128": 0.856
1070
+ },
1071
+ "7": {
1072
+ "pass@1": 0.6011484375,
1073
+ "pass@2": 0.6669009596456699,
1074
+ "pass@4": 0.7120814229471315,
1075
+ "pass@8": 0.7453713718484262,
1076
+ "pass@16": 0.7730127547498042,
1077
+ "pass@32": 0.7972900605991194,
1078
+ "pass@64": 0.8194086339543623,
1079
+ "pass@128": 0.84
1080
+ },
1081
+ "8": {
1082
+ "pass@1": 0.5643671875,
1083
+ "pass@2": 0.644838336614174,
1084
+ "pass@4": 0.705028264998125,
1085
+ "pass@8": 0.7487015677656571,
1086
+ "pass@16": 0.7801481836227231,
1087
+ "pass@32": 0.8052043841885894,
1088
+ "pass@64": 0.8275029223931598,
1089
+ "pass@128": 0.846
1090
+ },
1091
+ "9": {
1092
+ "pass@1": 0.537859375,
1093
+ "pass@2": 0.6318706938976384,
1094
+ "pass@4": 0.7038911389201337,
1095
+ "pass@8": 0.7570356023780919,
1096
+ "pass@16": 0.7966427895846567,
1097
+ "pass@32": 0.8280960178669846,
1098
+ "pass@64": 0.8536312807147408,
1099
+ "pass@128": 0.875
1100
+ },
1101
+ "11": {
1102
+ "pass@1": 0.55396875,
1103
+ "pass@2": 0.6387955216535438,
1104
+ "pass@4": 0.7043307344394445,
1105
+ "pass@8": 0.7546983339977341,
1106
+ "pass@16": 0.7929470653592272,
1107
+ "pass@32": 0.8232224510293635,
1108
+ "pass@64": 0.8489851562729089,
1109
+ "pass@128": 0.87
1110
+ },
1111
+ "12": {
1112
+ "pass@1": 0.548125,
1113
+ "pass@2": 0.6337396653543307,
1114
+ "pass@4": 0.6963535999250092,
1115
+ "pass@8": 0.7450771226331848,
1116
+ "pass@16": 0.7842876873708078,
1117
+ "pass@32": 0.8163879483066379,
1118
+ "pass@64": 0.8437814099411024,
1119
+ "pass@128": 0.867
1120
+ },
1121
+ "13": {
1122
+ "pass@1": 0.4446015625,
1123
+ "pass@2": 0.5573118848425194,
1124
+ "pass@4": 0.6469788156167974,
1125
+ "pass@8": 0.7153580123659391,
1126
+ "pass@16": 0.7666650305593797,
1127
+ "pass@32": 0.8060510867048352,
1128
+ "pass@64": 0.8375774928020564,
1129
+ "pass@128": 0.861
1130
+ },
1131
+ "14": {
1132
+ "pass@1": 0.337375,
1133
+ "pass@2": 0.44927989665354384,
1134
+ "pass@4": 0.5528005280277467,
1135
+ "pass@8": 0.639973979765343,
1136
+ "pass@16": 0.7076267967485103,
1137
+ "pass@32": 0.7591571835994034,
1138
+ "pass@64": 0.7992540700872042,
1139
+ "pass@128": 0.831
1140
+ },
1141
+ "15": {
1142
+ "pass@1": 0.23525,
1143
+ "pass@2": 0.3350001230314964,
1144
+ "pass@4": 0.43675789576302876,
1145
+ "pass@8": 0.5272547912266787,
1146
+ "pass@16": 0.6012428286892009,
1147
+ "pass@32": 0.6610689530008375,
1148
+ "pass@64": 0.7108299522498246,
1149
+ "pass@128": 0.753
1150
+ },
1151
+ "16": {
1152
+ "pass@1": 0.17765625,
1153
+ "pass@2": 0.2467701771653543,
1154
+ "pass@4": 0.31910015944881853,
1155
+ "pass@8": 0.38902820548262157,
1156
+ "pass@16": 0.45373992733889545,
1157
+ "pass@32": 0.5124913954015271,
1158
+ "pass@64": 0.5660727186562804,
1159
+ "pass@128": 0.614
1160
+ },
1161
+ "17": {
1162
+ "pass@1": 0.1604609375,
1163
+ "pass@2": 0.2241515748031496,
1164
+ "pass@4": 0.29066624925009377,
1165
+ "pass@8": 0.3551817123763024,
1166
+ "pass@16": 0.41555541132674007,
1167
+ "pass@32": 0.47172652564446815,
1168
+ "pass@64": 0.5251010037440795,
1169
+ "pass@128": 0.578
1170
+ },
1171
+ "18": {
1172
+ "pass@1": 0.15584375,
1173
+ "pass@2": 0.22099729330708662,
1174
+ "pass@4": 0.2880469713160855,
1175
+ "pass@8": 0.34985024066375137,
1176
+ "pass@16": 0.4038654648619413,
1177
+ "pass@32": 0.4512529431204043,
1178
+ "pass@64": 0.494365624612651,
1179
+ "pass@128": 0.539
1180
+ },
1181
+ "19": {
1182
+ "pass@1": 0.137875,
1183
+ "pass@2": 0.1946306594488188,
1184
+ "pass@4": 0.2525709119797523,
1185
+ "pass@8": 0.30590235229761004,
1186
+ "pass@16": 0.3528854382823781,
1187
+ "pass@32": 0.39543239658733015,
1188
+ "pass@64": 0.43542615681957186,
1189
+ "pass@128": 0.472
1190
+ },
1191
+ "20": {
1192
+ "pass@1": 0.1423046875,
1193
+ "pass@2": 0.2069383612204721,
1194
+ "pass@4": 0.27238460179977464,
1195
+ "pass@8": 0.3287041877243397,
1196
+ "pass@16": 0.37280509459990285,
1197
+ "pass@32": 0.4085022214331564,
1198
+ "pass@64": 0.4404992782399348,
1199
+ "pass@128": 0.471
1200
+ }
1201
+ },
1202
+ "per_op_accuracy": {
1203
+ "10": 0.59,
1204
+ "2": 0.9444296875,
1205
+ "3": 0.9078359375,
1206
+ "4": 0.77075,
1207
+ "5": 0.613390625,
1208
+ "6": 0.618515625,
1209
+ "7": 0.6011484375,
1210
+ "8": 0.5643671875,
1211
+ "9": 0.537859375,
1212
+ "11": 0.55396875,
1213
+ "12": 0.548125,
1214
+ "13": 0.4446015625,
1215
+ "14": 0.337375,
1216
+ "15": 0.23525,
1217
+ "16": 0.17765625,
1218
+ "17": 0.1604609375,
1219
+ "18": 0.15584375,
1220
+ "19": 0.137875,
1221
+ "20": 0.1423046875
1222
+ },
1223
+ "per_op_avg_response_len": {
1224
+ "10": 272.9003203125,
1225
+ "2": 117.993890625,
1226
+ "3": 149.4579921875,
1227
+ "4": 161.6555234375,
1228
+ "5": 172.9878203125,
1229
+ "6": 202.7284140625,
1230
+ "7": 223.0876328125,
1231
+ "8": 241.644890625,
1232
+ "9": 255.95815625,
1233
+ "11": 290.2597421875,
1234
+ "12": 296.8439921875,
1235
+ "13": 307.613546875,
1236
+ "14": 320.121765625,
1237
+ "15": 321.7593515625,
1238
+ "16": 320.8959921875,
1239
+ "17": 322.09896875,
1240
+ "18": 320.365109375,
1241
+ "19": 314.7880625,
1242
+ "20": 323.5841796875
1243
+ },
1244
+ "per_op_avg_loss": {},
1245
+ "per_op_length": {
1246
+ "10": 272.9003203125,
1247
+ "2": 117.993890625,
1248
+ "3": 149.4579921875,
1249
+ "4": 161.6555234375,
1250
+ "5": 172.9878203125,
1251
+ "6": 202.7284140625,
1252
+ "7": 223.0876328125,
1253
+ "8": 241.644890625,
1254
+ "9": 255.95815625,
1255
+ "11": 290.2597421875,
1256
+ "12": 296.8439921875,
1257
+ "13": 307.613546875,
1258
+ "14": 320.121765625,
1259
+ "15": 321.7593515625,
1260
+ "16": 320.8959921875,
1261
+ "17": 322.09896875,
1262
+ "18": 320.365109375,
1263
+ "19": 314.7880625,
1264
+ "20": 323.5841796875
1265
+ },
1266
+ "per_op_loss": {},
1267
+ "per_template": {
1268
+ "crazy_zootopia": {
1269
+ "count": 809728,
1270
+ "correct": 381241,
1271
+ "answer_accuracy": 0.47082600576983874,
1272
+ "avg_response_len": 248.62063680643377,
1273
+ "resp_tokens_sum": 201315091,
1274
+ "pass_at_k": {
1275
+ "pass@1": 0.47082600576983874,
1276
+ "pass@2": 0.5420628659438301,
1277
+ "pass@4": 0.60205425237215,
1278
+ "pass@8": 0.651081216510764,
1279
+ "pass@16": 0.6908459763141861,
1280
+ "pass@32": 0.7243716711064014,
1281
+ "pass@64": 0.7542292077612036,
1282
+ "pass@128": 0.7809042048687954
1283
+ },
1284
+ "per_op_pass_at_k": {
1285
+ "10": {
1286
+ "pass@1": 0.5741380899705014,
1287
+ "pass@2": 0.6374345283720061,
1288
+ "pass@4": 0.6877484625815574,
1289
+ "pass@8": 0.7339820506256319,
1290
+ "pass@16": 0.7768820473639336,
1291
+ "pass@32": 0.8134513625208626,
1292
+ "pass@64": 0.8442222274966321,
1293
+ "pass@128": 0.8672566371681416
1294
+ },
1295
+ "2": {
1296
+ "pass@1": 0.9447770638629284,
1297
+ "pass@2": 0.9496820344886795,
1298
+ "pass@4": 0.9534258314089241,
1299
+ "pass@8": 0.9564988072518206,
1300
+ "pass@16": 0.9596533365254536,
1301
+ "pass@32": 0.9640128309589705,
1302
+ "pass@64": 0.9696471858423932,
1303
+ "pass@128": 0.9750778816199377
1304
+ },
1305
+ "3": {
1306
+ "pass@1": 0.890075496941896,
1307
+ "pass@2": 0.9153731428640232,
1308
+ "pass@4": 0.9288290752646745,
1309
+ "pass@8": 0.9363709567903885,
1310
+ "pass@16": 0.9418395773507783,
1311
+ "pass@32": 0.9473595562005767,
1312
+ "pass@64": 0.9545467492115126,
1313
+ "pass@128": 0.963302752293578
1314
+ },
1315
+ "4": {
1316
+ "pass@1": 0.7752197265625,
1317
+ "pass@2": 0.8222317913385826,
1318
+ "pass@4": 0.8546887019005436,
1319
+ "pass@8": 0.8798078050264433,
1320
+ "pass@16": 0.8987486358045697,
1321
+ "pass@32": 0.9141457607856986,
1322
+ "pass@64": 0.9275849795438009,
1323
+ "pass@128": 0.940625
1324
+ },
1325
+ "5": {
1326
+ "pass@1": 0.5919642857142857,
1327
+ "pass@2": 0.6483459098862642,
1328
+ "pass@4": 0.6903727599526247,
1329
+ "pass@8": 0.721324081648959,
1330
+ "pass@16": 0.7452422640151924,
1331
+ "pass@32": 0.7649371701462218,
1332
+ "pass@64": 0.784241095420447,
1333
+ "pass@128": 0.8063492063492064
1334
+ },
1335
+ "6": {
1336
+ "pass@1": 0.6101936751497006,
1337
+ "pass@2": 0.6852625948889624,
1338
+ "pass@4": 0.737672784633208,
1339
+ "pass@8": 0.772938922756174,
1340
+ "pass@16": 0.798958840272014,
1341
+ "pass@32": 0.8201741899475089,
1342
+ "pass@64": 0.8383901021445377,
1343
+ "pass@128": 0.8532934131736527
1344
+ },
1345
+ "7": {
1346
+ "pass@1": 0.6123104659763313,
1347
+ "pass@2": 0.6833205691189485,
1348
+ "pass@4": 0.7293244864362364,
1349
+ "pass@8": 0.7616322923990702,
1350
+ "pass@16": 0.7878927447570453,
1351
+ "pass@32": 0.8115178358283376,
1352
+ "pass@64": 0.8346118917446502,
1353
+ "pass@128": 0.8579881656804734
1354
+ },
1355
+ "8": {
1356
+ "pass@1": 0.5609475160256411,
1357
+ "pass@2": 0.6446952919947505,
1358
+ "pass@4": 0.7031014176112595,
1359
+ "pass@8": 0.7419600517349403,
1360
+ "pass@16": 0.7673220278771473,
1361
+ "pass@32": 0.7887273838182889,
1362
+ "pass@64": 0.8099674098488452,
1363
+ "pass@128": 0.8269230769230769
1364
+ },
1365
+ "9": {
1366
+ "pass@1": 0.5545994718309859,
1367
+ "pass@2": 0.6576068121326386,
1368
+ "pass@4": 0.7359780795006251,
1369
+ "pass@8": 0.790439971280655,
1370
+ "pass@16": 0.8279563910100225,
1371
+ "pass@32": 0.8539518807147877,
1372
+ "pass@64": 0.8742415789463832,
1373
+ "pass@128": 0.8929577464788733
1374
+ },
1375
+ "11": {
1376
+ "pass@1": 0.5319360835694051,
1377
+ "pass@2": 0.6215784696973076,
1378
+ "pass@4": 0.6907830577480926,
1379
+ "pass@8": 0.7432213329841176,
1380
+ "pass@16": 0.7828037516289943,
1381
+ "pass@32": 0.8150725689730364,
1382
+ "pass@64": 0.8445493457746914,
1383
+ "pass@128": 0.8725212464589235
1384
+ },
1385
+ "12": {
1386
+ "pass@1": 0.519332627118644,
1387
+ "pass@2": 0.6027417256105698,
1388
+ "pass@4": 0.6637228021285474,
1389
+ "pass@8": 0.7114096130410679,
1390
+ "pass@16": 0.7524092343759183,
1391
+ "pass@32": 0.7890026535707286,
1392
+ "pass@64": 0.8212713564382537,
1393
+ "pass@128": 0.847457627118644
1394
+ },
1395
+ "13": {
1396
+ "pass@1": 0.4448961598746082,
1397
+ "pass@2": 0.5568930034803646,
1398
+ "pass@4": 0.6451671588661135,
1399
+ "pass@8": 0.7110854193688476,
1400
+ "pass@16": 0.7608013473487977,
1401
+ "pass@32": 0.8014909701103228,
1402
+ "pass@64": 0.8384952651641907,
1403
+ "pass@128": 0.8683385579937304
1404
+ },
1405
+ "14": {
1406
+ "pass@1": 0.34197332554517135,
1407
+ "pass@2": 0.4501220349792721,
1408
+ "pass@4": 0.5494017024157026,
1409
+ "pass@8": 0.6321989976492106,
1410
+ "pass@16": 0.6955289358403063,
1411
+ "pass@32": 0.7459415528358541,
1412
+ "pass@64": 0.7887735797886692,
1413
+ "pass@128": 0.822429906542056
1414
+ },
1415
+ "15": {
1416
+ "pass@1": 0.23803287572254336,
1417
+ "pass@2": 0.33669275692958894,
1418
+ "pass@4": 0.44012880236068747,
1419
+ "pass@8": 0.5341238336228396,
1420
+ "pass@16": 0.6110083221314118,
1421
+ "pass@32": 0.6745370889577138,
1422
+ "pass@64": 0.7290513406818961,
1423
+ "pass@128": 0.7745664739884393
1424
+ },
1425
+ "16": {
1426
+ "pass@1": 0.20163810483870967,
1427
+ "pass@2": 0.2788862077724155,
1428
+ "pass@4": 0.3585027432054865,
1429
+ "pass@8": 0.43076881842449627,
1430
+ "pass@16": 0.4919829537899595,
1431
+ "pass@32": 0.5441670943646694,
1432
+ "pass@64": 0.5902012441068575,
1433
+ "pass@128": 0.6290322580645161
1434
+ },
1435
+ "17": {
1436
+ "pass@1": 0.156882225433526,
1437
+ "pass@2": 0.21556575997906322,
1438
+ "pass@4": 0.2771691409780425,
1439
+ "pass@8": 0.341674925379861,
1440
+ "pass@16": 0.4072087355948067,
1441
+ "pass@32": 0.4670126441556912,
1442
+ "pass@64": 0.5169342900561144,
1443
+ "pass@128": 0.5606936416184971
1444
+ },
1445
+ "18": {
1446
+ "pass@1": 0.14910714285714285,
1447
+ "pass@2": 0.21308176321709787,
1448
+ "pass@4": 0.280601617654936,
1449
+ "pass@8": 0.34355741540217244,
1450
+ "pass@16": 0.3985258710428493,
1451
+ "pass@32": 0.4461388543379518,
1452
+ "pass@64": 0.4872909465164243,
1453
+ "pass@128": 0.5285714285714286
1454
+ },
1455
+ "19": {
1456
+ "pass@1": 0.14057807807807807,
1457
+ "pass@2": 0.20221685760563707,
1458
+ "pass@4": 0.2649557811467261,
1459
+ "pass@8": 0.3198520926064229,
1460
+ "pass@16": 0.365459672897077,
1461
+ "pass@32": 0.4061949642563818,
1462
+ "pass@64": 0.44616338763451124,
1463
+ "pass@128": 0.4864864864864865
1464
+ },
1465
+ "20": {
1466
+ "pass@1": 0.15261355105105104,
1467
+ "pass@2": 0.2182194497647254,
1468
+ "pass@4": 0.2835954342981902,
1469
+ "pass@8": 0.33920423135788597,
1470
+ "pass@16": 0.3826292663379262,
1471
+ "pass@32": 0.41802288691041456,
1472
+ "pass@64": 0.4501180597103666,
1473
+ "pass@128": 0.4804804804804805
1474
+ }
1475
+ }
1476
+ },
1477
+ "teachers_in_school": {
1478
+ "count": 821120,
1479
+ "correct": 385295,
1480
+ "answer_accuracy": 0.46923105027279816,
1481
+ "avg_response_len": 256.39123148869834,
1482
+ "resp_tokens_sum": 210527968,
1483
+ "pass_at_k": {
1484
+ "pass@1": 0.46923105027279816,
1485
+ "pass@2": 0.537744087583857,
1486
+ "pass@4": 0.5969630008787917,
1487
+ "pass@8": 0.646551698886574,
1488
+ "pass@16": 0.6869019578091474,
1489
+ "pass@32": 0.720059532777375,
1490
+ "pass@64": 0.7484124289105339,
1491
+ "pass@128": 0.7730319563522993
1492
+ },
1493
+ "per_op_pass_at_k": {
1494
+ "10": {
1495
+ "pass@1": 0.6025929831288344,
1496
+ "pass@2": 0.6655788820588381,
1497
+ "pass@4": 0.712638290708293,
1498
+ "pass@8": 0.7520765879868826,
1499
+ "pass@16": 0.7863821934536588,
1500
+ "pass@32": 0.8153207950273516,
1501
+ "pass@64": 0.8387030964952171,
1502
+ "pass@128": 0.8588957055214724
1503
+ },
1504
+ "2": {
1505
+ "pass@1": 0.9505841565349544,
1506
+ "pass@2": 0.9532843052198261,
1507
+ "pass@4": 0.9558882856208325,
1508
+ "pass@8": 0.9587953989382784,
1509
+ "pass@16": 0.961782530755302,
1510
+ "pass@32": 0.9643643890269245,
1511
+ "pass@64": 0.9665359573547208,
1512
+ "pass@128": 0.9696048632218845
1513
+ },
1514
+ "3": {
1515
+ "pass@1": 0.9034327651515152,
1516
+ "pass@2": 0.9192156555714627,
1517
+ "pass@4": 0.9302861139516647,
1518
+ "pass@8": 0.9386739269294071,
1519
+ "pass@16": 0.9438559355563287,
1520
+ "pass@32": 0.9468557193564537,
1521
+ "pass@64": 0.9492044725881807,
1522
+ "pass@128": 0.9515151515151515
1523
+ },
1524
+ "4": {
1525
+ "pass@1": 0.7489367603550295,
1526
+ "pass@2": 0.7910608518147509,
1527
+ "pass@4": 0.8231998353201413,
1528
+ "pass@8": 0.8477757288644923,
1529
+ "pass@16": 0.8653097885880157,
1530
+ "pass@32": 0.877466942076976,
1531
+ "pass@64": 0.8870172256269219,
1532
+ "pass@128": 0.893491124260355
1533
+ },
1534
+ "5": {
1535
+ "pass@1": 0.6153532608695652,
1536
+ "pass@2": 0.6656110920917495,
1537
+ "pass@4": 0.7047145457361306,
1538
+ "pass@8": 0.734960864574528,
1539
+ "pass@16": 0.7579012844461602,
1540
+ "pass@32": 0.7753412883276696,
1541
+ "pass@64": 0.7880631576774196,
1542
+ "pass@128": 0.8
1543
+ },
1544
+ "6": {
1545
+ "pass@1": 0.6086138968481375,
1546
+ "pass@2": 0.6848056454887982,
1547
+ "pass@4": 0.7375910427027563,
1548
+ "pass@8": 0.7741764834993599,
1549
+ "pass@16": 0.8002319395512839,
1550
+ "pass@32": 0.8218286879890876,
1551
+ "pass@64": 0.8432342376670118,
1552
+ "pass@128": 0.8653295128939829
1553
+ },
1554
+ "7": {
1555
+ "pass@1": 0.603888186416185,
1556
+ "pass@2": 0.6690694551909333,
1557
+ "pass@4": 0.7135883215103886,
1558
+ "pass@8": 0.7469629249550308,
1559
+ "pass@16": 0.7763369658056625,
1560
+ "pass@32": 0.8021818464035171,
1561
+ "pass@64": 0.8246229458789898,
1562
+ "pass@128": 0.8439306358381503
1563
+ },
1564
+ "8": {
1565
+ "pass@1": 0.5481294014084507,
1566
+ "pass@2": 0.6296779693911498,
1567
+ "pass@4": 0.6941487652951831,
1568
+ "pass@8": 0.7432917081970493,
1569
+ "pass@16": 0.7800826825934518,
1570
+ "pass@32": 0.8083199455882679,
1571
+ "pass@64": 0.8310643932196761,
1572
+ "pass@128": 0.8507042253521127
1573
+ },
1574
+ "9": {
1575
+ "pass@1": 0.5193207335329342,
1576
+ "pass@2": 0.6124250023574895,
1577
+ "pass@4": 0.6825153814979706,
1578
+ "pass@8": 0.7339480753308065,
1579
+ "pass@16": 0.7732788025060358,
1580
+ "pass@32": 0.8077219918528098,
1581
+ "pass@64": 0.8375948255424386,
1582
+ "pass@128": 0.8622754491017964
1583
+ },
1584
+ "11": {
1585
+ "pass@1": 0.5750894201807228,
1586
+ "pass@2": 0.6621766346883593,
1587
+ "pass@4": 0.7284196917026932,
1588
+ "pass@8": 0.7780922038504625,
1589
+ "pass@16": 0.813405098694823,
1590
+ "pass@32": 0.8392815462756775,
1591
+ "pass@64": 0.8604137867079064,
1592
+ "pass@128": 0.8765060240963856
1593
+ },
1594
+ "12": {
1595
+ "pass@1": 0.5449647335423198,
1596
+ "pass@2": 0.6343681349196552,
1597
+ "pass@4": 0.7027254673103162,
1598
+ "pass@8": 0.7573112125137871,
1599
+ "pass@16": 0.7986891262756064,
1600
+ "pass@32": 0.8283058253225973,
1601
+ "pass@64": 0.8514306286603033,
1602
+ "pass@128": 0.8714733542319749
1603
+ },
1604
+ "13": {
1605
+ "pass@1": 0.41474089635854344,
1606
+ "pass@2": 0.5216235608416595,
1607
+ "pass@4": 0.609166683208716,
1608
+ "pass@8": 0.6811120686409082,
1609
+ "pass@16": 0.7386794715073752,
1610
+ "pass@32": 0.7834822105451329,
1611
+ "pass@64": 0.8176940029791813,
1612
+ "pass@128": 0.8403361344537815
1613
+ },
1614
+ "14": {
1615
+ "pass@1": 0.31992415935672514,
1616
+ "pass@2": 0.4319049448588662,
1617
+ "pass@4": 0.5384300460249487,
1618
+ "pass@8": 0.6301912311847001,
1619
+ "pass@16": 0.7053491688109808,
1620
+ "pass@32": 0.7645782124873614,
1621
+ "pass@64": 0.8085707748135609,
1622
+ "pass@128": 0.8391812865497076
1623
+ },
1624
+ "15": {
1625
+ "pass@1": 0.2126657196969697,
1626
+ "pass@2": 0.30765442316869457,
1627
+ "pass@4": 0.40781750747065687,
1628
+ "pass@8": 0.49642400822735633,
1629
+ "pass@16": 0.5680396411962745,
1630
+ "pass@32": 0.626610399902073,
1631
+ "pass@64": 0.676446573867005,
1632
+ "pass@128": 0.7212121212121212
1633
+ },
1634
+ "16": {
1635
+ "pass@1": 0.15422453703703703,
1636
+ "pass@2": 0.2157730584157749,
1637
+ "pass@4": 0.2825862648579184,
1638
+ "pass@8": 0.350729859385697,
1639
+ "pass@16": 0.41576288513575577,
1640
+ "pass@32": 0.4756244133047699,
1641
+ "pass@64": 0.5334865753499802,
1642
+ "pass@128": 0.5868945868945868
1643
+ },
1644
+ "17": {
1645
+ "pass@1": 0.1774310947204969,
1646
+ "pass@2": 0.24994039467892606,
1647
+ "pass@4": 0.32356567764045,
1648
+ "pass@8": 0.39138325498864096,
1649
+ "pass@16": 0.4512251459100479,
1650
+ "pass@32": 0.5049210552230611,
1651
+ "pass@64": 0.5575395956365649,
1652
+ "pass@128": 0.6086956521739131
1653
+ },
1654
+ "18": {
1655
+ "pass@1": 0.1604779411764706,
1656
+ "pass@2": 0.2271639069013432,
1657
+ "pass@4": 0.29349490863274436,
1658
+ "pass@8": 0.35195918625723943,
1659
+ "pass@16": 0.4028748773367787,
1660
+ "pass@32": 0.44786918936674186,
1661
+ "pass@64": 0.4874247416201932,
1662
+ "pass@128": 0.5264705882352941
1663
+ },
1664
+ "19": {
1665
+ "pass@1": 0.13846261160714285,
1666
+ "pass@2": 0.19192473987626552,
1667
+ "pass@4": 0.24761419108325744,
1668
+ "pass@8": 0.3012326120456533,
1669
+ "pass@16": 0.3485993171013056,
1670
+ "pass@32": 0.38909940621048456,
1671
+ "pass@64": 0.4246445731793099,
1672
+ "pass@128": 0.45535714285714285
1673
+ },
1674
+ "20": {
1675
+ "pass@1": 0.12930389221556887,
1676
+ "pass@2": 0.1930537301381488,
1677
+ "pass@4": 0.26072267753581696,
1678
+ "pass@8": 0.32056687148160085,
1679
+ "pass@16": 0.36709700781450916,
1680
+ "pass@32": 0.4044418838960831,
1681
+ "pass@64": 0.4377505306516329,
1682
+ "pass@128": 0.46706586826347307
1683
+ }
1684
+ }
1685
+ },
1686
+ "movie_festival_awards": {
1687
+ "count": 801152,
1688
+ "correct": 390809,
1689
+ "answer_accuracy": 0.4878088053203387,
1690
+ "avg_response_len": 274.67989345342704,
1691
+ "resp_tokens_sum": 220060346,
1692
+ "pass_at_k": {
1693
+ "pass@1": 0.4878088053203387,
1694
+ "pass@2": 0.5543622105428079,
1695
+ "pass@4": 0.6110727703213531,
1696
+ "pass@8": 0.6581146455552732,
1697
+ "pass@16": 0.6963855810057189,
1698
+ "pass@32": 0.728310310639796,
1699
+ "pass@64": 0.7563010233507497,
1700
+ "pass@128": 0.7823933535708579
1701
+ },
1702
+ "per_op_pass_at_k": {
1703
+ "10": {
1704
+ "pass@1": 0.5937966417910447,
1705
+ "pass@2": 0.6551361058878838,
1706
+ "pass@4": 0.7051956306767619,
1707
+ "pass@8": 0.7468369195114555,
1708
+ "pass@16": 0.7804752311309393,
1709
+ "pass@32": 0.8087134368953917,
1710
+ "pass@64": 0.8337175398407523,
1711
+ "pass@128": 0.8567164179104477
1712
+ },
1713
+ "2": {
1714
+ "pass@1": 0.9383258928571429,
1715
+ "pass@2": 0.9434167604049495,
1716
+ "pass@4": 0.94656041726927,
1717
+ "pass@8": 0.9493516710403994,
1718
+ "pass@16": 0.9522428421965253,
1719
+ "pass@32": 0.9551417367243559,
1720
+ "pass@64": 0.9585785148501862,
1721
+ "pass@128": 0.9628571428571429
1722
+ },
1723
+ "3": {
1724
+ "pass@1": 0.9290041909620991,
1725
+ "pass@2": 0.9421074039852161,
1726
+ "pass@4": 0.9482759917400994,
1727
+ "pass@8": 0.9517019733337934,
1728
+ "pass@16": 0.9544241083697254,
1729
+ "pass@32": 0.9571684872177765,
1730
+ "pass@64": 0.9597425063031881,
1731
+ "pass@128": 0.9620991253644315
1732
+ },
1733
+ "4": {
1734
+ "pass@1": 0.78812591374269,
1735
+ "pass@2": 0.8241577738637934,
1736
+ "pass@4": 0.852643570047165,
1737
+ "pass@8": 0.8747651744721673,
1738
+ "pass@16": 0.8891099109783571,
1739
+ "pass@32": 0.8973778497351624,
1740
+ "pass@64": 0.9027188391175657,
1741
+ "pass@128": 0.9064327485380117
1742
+ },
1743
+ "5": {
1744
+ "pass@1": 0.63125,
1745
+ "pass@2": 0.6775536272579896,
1746
+ "pass@4": 0.7143841425152739,
1747
+ "pass@8": 0.7485260162723093,
1748
+ "pass@16": 0.7803454382627247,
1749
+ "pass@32": 0.807788639056809,
1750
+ "pass@64": 0.8313754642783288,
1751
+ "pass@128": 0.8558823529411764
1752
+ },
1753
+ "6": {
1754
+ "pass@1": 0.6381851340694006,
1755
+ "pass@2": 0.7121715020243922,
1756
+ "pass@4": 0.762374902269708,
1757
+ "pass@8": 0.7922977576259308,
1758
+ "pass@16": 0.8089222960611702,
1759
+ "pass@32": 0.8209672814519965,
1760
+ "pass@64": 0.8330832507873658,
1761
+ "pass@128": 0.8485804416403786
1762
+ },
1763
+ "7": {
1764
+ "pass@1": 0.586209454113924,
1765
+ "pass@2": 0.6469638474284858,
1766
+ "pass@4": 0.6919879344591413,
1767
+ "pass@8": 0.7262357119718345,
1768
+ "pass@16": 0.7534570153581185,
1769
+ "pass@32": 0.7767155482073544,
1770
+ "pass@64": 0.7974375799700597,
1771
+ "pass@128": 0.8164556962025317
1772
+ },
1773
+ "8": {
1774
+ "pass@1": 0.5848817567567568,
1775
+ "pass@2": 0.6611343134473057,
1776
+ "pass@4": 0.7184318649358015,
1777
+ "pass@8": 0.760785210854063,
1778
+ "pass@16": 0.792235311124261,
1779
+ "pass@32": 0.817320900160503,
1780
+ "pass@64": 0.840135828610614,
1781
+ "pass@128": 0.8588588588588588
1782
+ },
1783
+ "9": {
1784
+ "pass@1": 0.5386605707395499,
1785
+ "pass@2": 0.623377250170899,
1786
+ "pass@4": 0.690221168093538,
1787
+ "pass@8": 0.7437001781767524,
1788
+ "pass@16": 0.785990838389334,
1789
+ "pass@32": 0.8204628776025612,
1790
+ "pass@64": 0.8473274870019281,
1791
+ "pass@128": 0.8681672025723473
1792
+ },
1793
+ "11": {
1794
+ "pass@1": 0.5563988095238095,
1795
+ "pass@2": 0.6334466004249469,
1796
+ "pass@4": 0.6941237378065835,
1797
+ "pass@8": 0.7429034659555144,
1798
+ "pass@16": 0.7827518992619408,
1799
+ "pass@32": 0.8154297168836708,
1800
+ "pass@64": 0.8419106667537043,
1801
+ "pass@128": 0.8603174603174604
1802
+ },
1803
+ "12": {
1804
+ "pass@1": 0.5823776758409785,
1805
+ "pass@2": 0.6666839738495992,
1806
+ "pass@4": 0.7254627030566131,
1807
+ "pass@8": 0.7695897334090223,
1808
+ "pass@16": 0.8047491960850638,
1809
+ "pass@32": 0.834408106008233,
1810
+ "pass@64": 0.8606880098450249,
1811
+ "pass@128": 0.8837920489296636
1812
+ },
1813
+ "13": {
1814
+ "pass@1": 0.4772135416666667,
1815
+ "pass@2": 0.5970475478759603,
1816
+ "pass@4": 0.6904258828178885,
1817
+ "pass@8": 0.7572986267977532,
1818
+ "pass@16": 0.8032741648980943,
1819
+ "pass@32": 0.8354083891204626,
1820
+ "pass@64": 0.8585825436762666,
1821
+ "pass@128": 0.8765432098765432
1822
+ },
1823
+ "14": {
1824
+ "pass@1": 0.3507047477744807,
1825
+ "pass@2": 0.4661104815533072,
1826
+ "pass@4": 0.5706216789666868,
1827
+ "pass@8": 0.6573077164830237,
1828
+ "pass@16": 0.7214616991407049,
1829
+ "pass@32": 0.7662439064344672,
1830
+ "pass@64": 0.799782035575084,
1831
+ "pass@128": 0.8308605341246291
1832
+ },
1833
+ "15": {
1834
+ "pass@1": 0.2552806712962963,
1835
+ "pass@2": 0.36104465891416354,
1836
+ "pass@4": 0.4626344218546755,
1837
+ "pass@8": 0.5513210557967552,
1838
+ "pass@16": 0.6246323075214912,
1839
+ "pass@32": 0.6817829884376061,
1840
+ "pass@64": 0.7263914166598042,
1841
+ "pass@128": 0.7623456790123457
1842
+ },
1843
+ "16": {
1844
+ "pass@1": 0.17998709439528024,
1845
+ "pass@2": 0.24949589749843215,
1846
+ "pass@4": 0.3208747200294653,
1847
+ "pass@8": 0.3905123632054533,
1848
+ "pass@16": 0.45808979050548,
1849
+ "pass@32": 0.5216974250103424,
1850
+ "pass@64": 0.5777479204581456,
1851
+ "pass@128": 0.6283185840707964
1852
+ },
1853
+ "17": {
1854
+ "pass@1": 0.1477315512048193,
1855
+ "pass@2": 0.20808739386680566,
1856
+ "pass@4": 0.2728240309381508,
1857
+ "pass@8": 0.33414692797749634,
1858
+ "pass@16": 0.3896587103250072,
1859
+ "pass@32": 0.4444444909179926,
1860
+ "pass@64": 0.5021505710532824,
1861
+ "pass@128": 0.5662650602409639
1862
+ },
1863
+ "18": {
1864
+ "pass@1": 0.15836693548387096,
1865
+ "pass@2": 0.22317079946659896,
1866
+ "pass@4": 0.29047785871524134,
1867
+ "pass@8": 0.3546420062759032,
1868
+ "pass@16": 0.4109804893627079,
1869
+ "pass@32": 0.4607381281207391,
1870
+ "pass@64": 0.509965745745279,
1871
+ "pass@128": 0.5645161290322581
1872
+ },
1873
+ "19": {
1874
+ "pass@1": 0.1345591012084592,
1875
+ "pass@2": 0.18974541772724032,
1876
+ "pass@4": 0.24514280560096152,
1877
+ "pass@8": 0.29660860366263314,
1878
+ "pass@16": 0.3445860926332715,
1879
+ "pass@32": 0.3910334531819703,
1880
+ "pass@64": 0.4355684959185234,
1881
+ "pass@128": 0.4743202416918429
1882
+ },
1883
+ "20": {
1884
+ "pass@1": 0.14503566066066065,
1885
+ "pass@2": 0.20958359934737886,
1886
+ "pass@4": 0.27287071435890325,
1887
+ "pass@8": 0.3263658967186459,
1888
+ "pass@16": 0.3687061510490307,
1889
+ "pass@32": 0.4030540866987888,
1890
+ "pass@64": 0.43363749885506614,
1891
+ "pass@128": 0.46546546546546547
1892
+ }
1893
+ }
1894
+ }
1895
+ }
1896
+ }
1897
+ }