ZachW commited on
Commit
e32aac8
·
verified ·
1 Parent(s): 6402b41

Upload extrapolation artifact: opd_ckpt18809_refkl_op11_14_s200_merged_hf far-OOD metrics

Browse files
020305/evals/far_ood/opd_ckpt18809_refkl_op11_14_s200_merged_hf/opd_ckpt18809_refkl_op11_14_s200_merged_hf_metrics.json ADDED
@@ -0,0 +1,904 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "opd_ckpt18809_refkl_op11_14_s200_merged_hf",
3
+ "id": {
4
+ "answer_accuracy": 0.0,
5
+ "avg_response_len": 0.0,
6
+ "avg_loss": 0.0,
7
+ "count": 0,
8
+ "validation_metrics": {},
9
+ "pass_at_k": {},
10
+ "per_op_pass_at_k": {},
11
+ "per_op_accuracy": {},
12
+ "per_op_avg_response_len": {},
13
+ "per_op_avg_loss": {},
14
+ "per_op_length": {},
15
+ "per_op_loss": {},
16
+ "per_template": {
17
+ "crazy_zootopia": {
18
+ "count": 0,
19
+ "correct": 0,
20
+ "answer_accuracy": 0.0,
21
+ "avg_response_len": 0.0,
22
+ "resp_tokens_sum": 0,
23
+ "pass_at_k": {},
24
+ "per_op_pass_at_k": {}
25
+ },
26
+ "teachers_in_school": {
27
+ "count": 0,
28
+ "correct": 0,
29
+ "answer_accuracy": 0.0,
30
+ "avg_response_len": 0.0,
31
+ "resp_tokens_sum": 0,
32
+ "pass_at_k": {},
33
+ "per_op_pass_at_k": {}
34
+ },
35
+ "movie_festival_awards": {
36
+ "count": 0,
37
+ "correct": 0,
38
+ "answer_accuracy": 0.0,
39
+ "avg_response_len": 0.0,
40
+ "resp_tokens_sum": 0,
41
+ "pass_at_k": {},
42
+ "per_op_pass_at_k": {}
43
+ }
44
+ }
45
+ },
46
+ "ood": {
47
+ "answer_accuracy": 0.002484375,
48
+ "avg_response_len": 297.913984375,
49
+ "avg_loss": 0.0,
50
+ "count": 64000,
51
+ "validation_metrics": {},
52
+ "pass_at_k": {
53
+ "pass@1": 0.002484375,
54
+ "pass@2": 0.004610887096774193,
55
+ "pass@4": 0.008140781423804228,
56
+ "pass@8": 0.013616341851820162,
57
+ "pass@16": 0.022010931899641574,
58
+ "pass@32": 0.0355
59
+ },
60
+ "per_op_pass_at_k": {
61
+ "21": {
62
+ "pass@1": 0.0034375,
63
+ "pass@2": 0.00658266129032258,
64
+ "pass@4": 0.0120952447163515,
65
+ "pass@8": 0.020621523915461622,
66
+ "pass@16": 0.03146551724137931,
67
+ "pass@32": 0.045
68
+ },
69
+ "22": {
70
+ "pass@1": 0.00375,
71
+ "pass@2": 0.007298387096774193,
72
+ "pass@4": 0.013849972191323692,
73
+ "pass@8": 0.02514460511679644,
74
+ "pass@16": 0.0428642936596218,
75
+ "pass@32": 0.07
76
+ },
77
+ "23": {
78
+ "pass@1": 0.00328125,
79
+ "pass@2": 0.005756048387096774,
80
+ "pass@4": 0.009340100111234706,
81
+ "pass@8": 0.014399555061179087,
82
+ "pass@16": 0.02258063710080444,
83
+ "pass@32": 0.035
84
+ },
85
+ "24": {
86
+ "pass@1": 0.00375,
87
+ "pass@2": 0.007298387096774193,
88
+ "pass@4": 0.013872914349276975,
89
+ "pass@8": 0.025402515140279327,
90
+ "pass@16": 0.04447194413545916,
91
+ "pass@32": 0.075
92
+ },
93
+ "25": {
94
+ "pass@1": 0.004375,
95
+ "pass@2": 0.007600806451612902,
96
+ "pass@4": 0.011979838709677419,
97
+ "pass@8": 0.017230994552351615,
98
+ "pass@16": 0.024997952636584934,
99
+ "pass@32": 0.04
100
+ },
101
+ "26": {
102
+ "pass@1": 0.00203125,
103
+ "pass@2": 0.0035080645161290323,
104
+ "pass@4": 0.005417825361512791,
105
+ "pass@8": 0.007403268589030548,
106
+ "pass@16": 0.009999830729796393,
107
+ "pass@32": 0.015
108
+ },
109
+ "27": {
110
+ "pass@1": 0.0015625,
111
+ "pass@2": 0.002973790322580645,
112
+ "pass@4": 0.005421301446051168,
113
+ "pass@8": 0.009257353849956742,
114
+ "pass@16": 0.014955815103201089,
115
+ "pass@32": 0.025
116
+ },
117
+ "28": {
118
+ "pass@1": 0.00125,
119
+ "pass@2": 0.0022883064516129033,
120
+ "pass@4": 0.003866101223581757,
121
+ "pass@8": 0.00573586035766236,
122
+ "pass@16": 0.007483005808923495,
123
+ "pass@32": 0.01
124
+ },
125
+ "29": {
126
+ "pass@1": 0.0009375,
127
+ "pass@2": 0.0018649193548387095,
128
+ "pass@4": 0.0036895161290322583,
129
+ "pass@8": 0.0072177419354838715,
130
+ "pass@16": 0.01379032258064516,
131
+ "pass@32": 0.025
132
+ },
133
+ "30": {
134
+ "pass@1": 0.00046875,
135
+ "pass@2": 0.0009375,
136
+ "pass@4": 0.001875,
137
+ "pass@8": 0.00375,
138
+ "pass@16": 0.0075,
139
+ "pass@32": 0.015
140
+ }
141
+ },
142
+ "per_op_accuracy": {
143
+ "21": 0.0034375,
144
+ "22": 0.00375,
145
+ "23": 0.00328125,
146
+ "24": 0.00375,
147
+ "25": 0.004375,
148
+ "26": 0.00203125,
149
+ "27": 0.0015625,
150
+ "28": 0.00125,
151
+ "29": 0.0009375,
152
+ "30": 0.00046875
153
+ },
154
+ "per_op_avg_response_len": {
155
+ "21": 291.64796875,
156
+ "22": 295.7371875,
157
+ "23": 293.74296875,
158
+ "24": 300.16828125,
159
+ "25": 297.91453125,
160
+ "26": 304.6490625,
161
+ "27": 304.83328125,
162
+ "28": 294.66640625,
163
+ "29": 299.793125,
164
+ "30": 295.98703125
165
+ },
166
+ "per_op_avg_loss": {},
167
+ "per_op_length": {
168
+ "21": 291.64796875,
169
+ "22": 295.7371875,
170
+ "23": 293.74296875,
171
+ "24": 300.16828125,
172
+ "25": 297.91453125,
173
+ "26": 304.6490625,
174
+ "27": 304.83328125,
175
+ "28": 294.66640625,
176
+ "29": 299.793125,
177
+ "30": 295.98703125
178
+ },
179
+ "per_op_loss": {},
180
+ "per_template": {
181
+ "crazy_zootopia": {
182
+ "count": 23072,
183
+ "correct": 61,
184
+ "answer_accuracy": 0.0026438973647711513,
185
+ "avg_response_len": 276.247572815534,
186
+ "resp_tokens_sum": 6373584,
187
+ "pass_at_k": {
188
+ "pass@1": 0.0026438973647711513,
189
+ "pass@2": 0.004840387454700014,
190
+ "pass@4": 0.008337704553834667,
191
+ "pass@8": 0.013432497404332345,
192
+ "pass@16": 0.020918036351159127,
193
+ "pass@32": 0.033287101248266296
194
+ },
195
+ "per_op_pass_at_k": {
196
+ "21": {
197
+ "pass@1": 0.0011867088607594937,
198
+ "pass@2": 0.0023734177215189874,
199
+ "pass@4": 0.004746835443037975,
200
+ "pass@8": 0.00949367088607595,
201
+ "pass@16": 0.0189873417721519,
202
+ "pass@32": 0.0379746835443038
203
+ },
204
+ "22": {
205
+ "pass@1": 0.00390625,
206
+ "pass@2": 0.007560483870967741,
207
+ "pass@4": 0.014168520578420465,
208
+ "pass@8": 0.024950562353231984,
209
+ "pass@16": 0.03939562476826103,
210
+ "pass@32": 0.05555555555555555
211
+ },
212
+ "23": {
213
+ "pass@1": 0.0004734848484848485,
214
+ "pass@2": 0.000946969696969697,
215
+ "pass@4": 0.001893939393939394,
216
+ "pass@8": 0.003787878787878788,
217
+ "pass@16": 0.007575757575757576,
218
+ "pass@32": 0.015151515151515152
219
+ },
220
+ "24": {
221
+ "pass@1": 0.006161971830985915,
222
+ "pass@2": 0.011869604725124942,
223
+ "pass@4": 0.02214314809882655,
224
+ "pass@8": 0.03936646700124117,
225
+ "pass@16": 0.06689052868689084,
226
+ "pass@32": 0.11267605633802817
227
+ },
228
+ "25": {
229
+ "pass@1": 0.011408730158730158,
230
+ "pass@2": 0.019169226830517153,
231
+ "pass@4": 0.028110599078341014,
232
+ "pass@8": 0.0348603001661956,
233
+ "pass@16": 0.03967604011614265,
234
+ "pass@32": 0.047619047619047616
235
+ },
236
+ "26": {
237
+ "pass@1": 0.0,
238
+ "pass@2": 0.0,
239
+ "pass@4": 0.0,
240
+ "pass@8": 0.0,
241
+ "pass@16": 0.0,
242
+ "pass@32": 0.0
243
+ },
244
+ "27": {
245
+ "pass@1": 0.00043402777777777775,
246
+ "pass@2": 0.0008680555555555555,
247
+ "pass@4": 0.001736111111111111,
248
+ "pass@8": 0.003472222222222222,
249
+ "pass@16": 0.006944444444444444,
250
+ "pass@32": 0.013888888888888888
251
+ },
252
+ "28": {
253
+ "pass@1": 0.0028735632183908046,
254
+ "pass@2": 0.005260474601408972,
255
+ "pass@4": 0.008887589019728177,
256
+ "pass@8": 0.013185885879683586,
257
+ "pass@16": 0.017202312204421828,
258
+ "pass@32": 0.022988505747126436
259
+ },
260
+ "29": {
261
+ "pass@1": 0.0004166666666666667,
262
+ "pass@2": 0.0008333333333333334,
263
+ "pass@4": 0.0016666666666666668,
264
+ "pass@8": 0.0033333333333333335,
265
+ "pass@16": 0.006666666666666667,
266
+ "pass@32": 0.013333333333333334
267
+ },
268
+ "30": {
269
+ "pass@1": 0.0004464285714285714,
270
+ "pass@2": 0.0008928571428571428,
271
+ "pass@4": 0.0017857142857142857,
272
+ "pass@8": 0.0035714285714285713,
273
+ "pass@16": 0.007142857142857143,
274
+ "pass@32": 0.014285714285714285
275
+ }
276
+ }
277
+ },
278
+ "teachers_in_school": {
279
+ "count": 19936,
280
+ "correct": 29,
281
+ "answer_accuracy": 0.0014546548956661316,
282
+ "avg_response_len": 290.34595706260035,
283
+ "resp_tokens_sum": 5788337,
284
+ "pass_at_k": {
285
+ "pass@1": 0.0014546548956661316,
286
+ "pass@2": 0.0028607673587738823,
287
+ "pass@4": 0.005535087139803991,
288
+ "pass@8": 0.0103836615322536,
289
+ "pass@16": 0.018455498083299257,
290
+ "pass@32": 0.030497592295345103
291
+ },
292
+ "per_op_pass_at_k": {
293
+ "21": {
294
+ "pass@1": 0.004326923076923077,
295
+ "pass@2": 0.008374689826302729,
296
+ "pass@4": 0.015694361256096516,
297
+ "pass@8": 0.027637545991272352,
298
+ "pass@16": 0.0436382305125353,
299
+ "pass@32": 0.06153846153846154
300
+ },
301
+ "22": {
302
+ "pass@1": 0.0030737704918032786,
303
+ "pass@2": 0.006114489687995769,
304
+ "pass@4": 0.012096774193548387,
305
+ "pass@8": 0.023664727657324167,
306
+ "pass@16": 0.0452141723955579,
307
+ "pass@32": 0.08196721311475409
308
+ },
309
+ "23": {
310
+ "pass@1": 0.0013392857142857143,
311
+ "pass@2": 0.0026497695852534552,
312
+ "pass@4": 0.005184331797235023,
313
+ "pass@8": 0.00990783410138249,
314
+ "pass@16": 0.017972350230414748,
315
+ "pass@32": 0.02857142857142857
316
+ },
317
+ "24": {
318
+ "pass@1": 0.0027901785714285715,
319
+ "pass@2": 0.005472350230414745,
320
+ "pass@4": 0.010527073732718894,
321
+ "pass@8": 0.019498847926267278,
322
+ "pass@16": 0.03369815668202765,
323
+ "pass@32": 0.05357142857142857
324
+ },
325
+ "25": {
326
+ "pass@1": 0.0004280821917808219,
327
+ "pass@2": 0.0008561643835616438,
328
+ "pass@4": 0.0017123287671232876,
329
+ "pass@8": 0.003424657534246575,
330
+ "pass@16": 0.00684931506849315,
331
+ "pass@32": 0.0136986301369863
332
+ },
333
+ "26": {
334
+ "pass@1": 0.0006127450980392157,
335
+ "pass@2": 0.0012254901960784314,
336
+ "pass@4": 0.0024509803921568627,
337
+ "pass@8": 0.004901960784313725,
338
+ "pass@16": 0.00980392156862745,
339
+ "pass@32": 0.0196078431372549
340
+ },
341
+ "27": {
342
+ "pass@1": 0.0005387931034482759,
343
+ "pass@2": 0.0010775862068965517,
344
+ "pass@4": 0.0021551724137931034,
345
+ "pass@8": 0.004310344827586207,
346
+ "pass@16": 0.008620689655172414,
347
+ "pass@32": 0.017241379310344827
348
+ },
349
+ "28": {
350
+ "pass@1": 0.0,
351
+ "pass@2": 0.0,
352
+ "pass@4": 0.0,
353
+ "pass@8": 0.0,
354
+ "pass@16": 0.0,
355
+ "pass@32": 0.0
356
+ },
357
+ "29": {
358
+ "pass@1": 0.0014423076923076924,
359
+ "pass@2": 0.002853598014888337,
360
+ "pass@4": 0.005583126550868486,
361
+ "pass@8": 0.010669975186104219,
362
+ "pass@16": 0.01935483870967742,
363
+ "pass@32": 0.03076923076923077
364
+ },
365
+ "30": {
366
+ "pass@1": 0.0,
367
+ "pass@2": 0.0,
368
+ "pass@4": 0.0,
369
+ "pass@8": 0.0,
370
+ "pass@16": 0.0,
371
+ "pass@32": 0.0
372
+ }
373
+ }
374
+ },
375
+ "movie_festival_awards": {
376
+ "count": 20992,
377
+ "correct": 69,
378
+ "answer_accuracy": 0.003286966463414634,
379
+ "avg_response_len": 328.9145388719512,
380
+ "resp_tokens_sum": 6904574,
381
+ "pass_at_k": {
382
+ "pass@1": 0.003286966463414634,
383
+ "pass@2": 0.0060207267899291895,
384
+ "pass@4": 0.010398961244200873,
385
+ "pass@8": 0.016888463323967543,
386
+ "pass@16": 0.026588695555185955,
387
+ "pass@32": 0.042682926829268296
388
+ },
389
+ "per_op_pass_at_k": {
390
+ "21": {
391
+ "pass@1": 0.005580357142857143,
392
+ "pass@2": 0.010440668202764977,
393
+ "pass@4": 0.01828420467185762,
394
+ "pass@8": 0.028176148101064676,
395
+ "pass@16": 0.03493961544573335,
396
+ "pass@32": 0.03571428571428571
397
+ },
398
+ "22": {
399
+ "pass@1": 0.004197761194029851,
400
+ "pass@2": 0.008094607607125662,
401
+ "pass@4": 0.015103846728537512,
402
+ "pass@8": 0.026700479803430013,
403
+ "pass@16": 0.04445237660418707,
404
+ "pass@32": 0.07462686567164178
405
+ },
406
+ "23": {
407
+ "pass@1": 0.00830078125,
408
+ "pass@2": 0.014112903225806451,
409
+ "pass@4": 0.021564324944382644,
410
+ "pass@8": 0.030255666017797552,
411
+ "pass@16": 0.04309473287549774,
412
+ "pass@32": 0.0625
413
+ },
414
+ "24": {
415
+ "pass@1": 0.0021404109589041095,
416
+ "pass@2": 0.0042532037118868755,
417
+ "pass@4": 0.008395934600088379,
418
+ "pass@8": 0.016349977905435263,
419
+ "pass@16": 0.030932390631904547,
420
+ "pass@32": 0.0547945205479452
421
+ },
422
+ "25": {
423
+ "pass@1": 0.001953125,
424
+ "pass@2": 0.00390625,
425
+ "pass@4": 0.0078125,
426
+ "pass@8": 0.015625,
427
+ "pass@16": 0.03125,
428
+ "pass@32": 0.0625
429
+ },
430
+ "26": {
431
+ "pass@1": 0.004518072289156626,
432
+ "pass@2": 0.007700155460551886,
433
+ "pass@4": 0.01154897677472962,
434
+ "pass@8": 0.014827153226579631,
435
+ "pass@16": 0.018071881276617813,
436
+ "pass@32": 0.024096385542168676
437
+ },
438
+ "27": {
439
+ "pass@1": 0.0035714285714285713,
440
+ "pass@2": 0.0067108294930875575,
441
+ "pass@4": 0.011918004131574764,
442
+ "pass@8": 0.01930672528559069,
443
+ "pass@16": 0.028445186009145966,
444
+ "pass@32": 0.04285714285714286
445
+ },
446
+ "28": {
447
+ "pass@1": 0.0,
448
+ "pass@2": 0.0,
449
+ "pass@4": 0.0,
450
+ "pass@8": 0.0,
451
+ "pass@16": 0.0,
452
+ "pass@32": 0.0
453
+ },
454
+ "29": {
455
+ "pass@1": 0.0010416666666666667,
456
+ "pass@2": 0.0020833333333333333,
457
+ "pass@4": 0.004166666666666667,
458
+ "pass@8": 0.008333333333333333,
459
+ "pass@16": 0.016666666666666666,
460
+ "pass@32": 0.03333333333333333
461
+ },
462
+ "30": {
463
+ "pass@1": 0.000946969696969697,
464
+ "pass@2": 0.001893939393939394,
465
+ "pass@4": 0.003787878787878788,
466
+ "pass@8": 0.007575757575757576,
467
+ "pass@16": 0.015151515151515152,
468
+ "pass@32": 0.030303030303030304
469
+ }
470
+ }
471
+ }
472
+ }
473
+ },
474
+ "total": {
475
+ "answer_accuracy": 0.002484375,
476
+ "avg_loss": 0.0,
477
+ "count": 64000,
478
+ "validation_metrics": {
479
+ "id": {},
480
+ "ood": {}
481
+ },
482
+ "pass_at_k": {
483
+ "pass@1": 0.002484375,
484
+ "pass@2": 0.004610887096774193,
485
+ "pass@4": 0.008140781423804228,
486
+ "pass@8": 0.013616341851820162,
487
+ "pass@16": 0.022010931899641574,
488
+ "pass@32": 0.0355
489
+ },
490
+ "per_op_pass_at_k": {
491
+ "21": {
492
+ "pass@1": 0.0034375,
493
+ "pass@2": 0.00658266129032258,
494
+ "pass@4": 0.0120952447163515,
495
+ "pass@8": 0.020621523915461622,
496
+ "pass@16": 0.03146551724137931,
497
+ "pass@32": 0.045
498
+ },
499
+ "22": {
500
+ "pass@1": 0.00375,
501
+ "pass@2": 0.007298387096774193,
502
+ "pass@4": 0.013849972191323692,
503
+ "pass@8": 0.02514460511679644,
504
+ "pass@16": 0.0428642936596218,
505
+ "pass@32": 0.07
506
+ },
507
+ "23": {
508
+ "pass@1": 0.00328125,
509
+ "pass@2": 0.005756048387096774,
510
+ "pass@4": 0.009340100111234706,
511
+ "pass@8": 0.014399555061179087,
512
+ "pass@16": 0.02258063710080444,
513
+ "pass@32": 0.035
514
+ },
515
+ "24": {
516
+ "pass@1": 0.00375,
517
+ "pass@2": 0.007298387096774193,
518
+ "pass@4": 0.013872914349276975,
519
+ "pass@8": 0.025402515140279327,
520
+ "pass@16": 0.04447194413545916,
521
+ "pass@32": 0.075
522
+ },
523
+ "25": {
524
+ "pass@1": 0.004375,
525
+ "pass@2": 0.007600806451612902,
526
+ "pass@4": 0.011979838709677419,
527
+ "pass@8": 0.017230994552351615,
528
+ "pass@16": 0.024997952636584934,
529
+ "pass@32": 0.04
530
+ },
531
+ "26": {
532
+ "pass@1": 0.00203125,
533
+ "pass@2": 0.0035080645161290323,
534
+ "pass@4": 0.005417825361512791,
535
+ "pass@8": 0.007403268589030548,
536
+ "pass@16": 0.009999830729796393,
537
+ "pass@32": 0.015
538
+ },
539
+ "27": {
540
+ "pass@1": 0.0015625,
541
+ "pass@2": 0.002973790322580645,
542
+ "pass@4": 0.005421301446051168,
543
+ "pass@8": 0.009257353849956742,
544
+ "pass@16": 0.014955815103201089,
545
+ "pass@32": 0.025
546
+ },
547
+ "28": {
548
+ "pass@1": 0.00125,
549
+ "pass@2": 0.0022883064516129033,
550
+ "pass@4": 0.003866101223581757,
551
+ "pass@8": 0.00573586035766236,
552
+ "pass@16": 0.007483005808923495,
553
+ "pass@32": 0.01
554
+ },
555
+ "29": {
556
+ "pass@1": 0.0009375,
557
+ "pass@2": 0.0018649193548387095,
558
+ "pass@4": 0.0036895161290322583,
559
+ "pass@8": 0.0072177419354838715,
560
+ "pass@16": 0.01379032258064516,
561
+ "pass@32": 0.025
562
+ },
563
+ "30": {
564
+ "pass@1": 0.00046875,
565
+ "pass@2": 0.0009375,
566
+ "pass@4": 0.001875,
567
+ "pass@8": 0.00375,
568
+ "pass@16": 0.0075,
569
+ "pass@32": 0.015
570
+ }
571
+ },
572
+ "per_op_accuracy": {
573
+ "21": 0.0034375,
574
+ "22": 0.00375,
575
+ "23": 0.00328125,
576
+ "24": 0.00375,
577
+ "25": 0.004375,
578
+ "26": 0.00203125,
579
+ "27": 0.0015625,
580
+ "28": 0.00125,
581
+ "29": 0.0009375,
582
+ "30": 0.00046875
583
+ },
584
+ "per_op_avg_response_len": {
585
+ "21": 291.64796875,
586
+ "22": 295.7371875,
587
+ "23": 293.74296875,
588
+ "24": 300.16828125,
589
+ "25": 297.91453125,
590
+ "26": 304.6490625,
591
+ "27": 304.83328125,
592
+ "28": 294.66640625,
593
+ "29": 299.793125,
594
+ "30": 295.98703125
595
+ },
596
+ "per_op_avg_loss": {},
597
+ "per_op_length": {
598
+ "21": 291.64796875,
599
+ "22": 295.7371875,
600
+ "23": 293.74296875,
601
+ "24": 300.16828125,
602
+ "25": 297.91453125,
603
+ "26": 304.6490625,
604
+ "27": 304.83328125,
605
+ "28": 294.66640625,
606
+ "29": 299.793125,
607
+ "30": 295.98703125
608
+ },
609
+ "per_op_loss": {},
610
+ "per_template": {
611
+ "crazy_zootopia": {
612
+ "count": 23072,
613
+ "correct": 61,
614
+ "answer_accuracy": 0.0026438973647711513,
615
+ "avg_response_len": 276.247572815534,
616
+ "resp_tokens_sum": 6373584,
617
+ "pass_at_k": {
618
+ "pass@1": 0.0026438973647711513,
619
+ "pass@2": 0.004840387454700014,
620
+ "pass@4": 0.008337704553834667,
621
+ "pass@8": 0.013432497404332345,
622
+ "pass@16": 0.020918036351159127,
623
+ "pass@32": 0.033287101248266296
624
+ },
625
+ "per_op_pass_at_k": {
626
+ "21": {
627
+ "pass@1": 0.0011867088607594937,
628
+ "pass@2": 0.0023734177215189874,
629
+ "pass@4": 0.004746835443037975,
630
+ "pass@8": 0.00949367088607595,
631
+ "pass@16": 0.0189873417721519,
632
+ "pass@32": 0.0379746835443038
633
+ },
634
+ "22": {
635
+ "pass@1": 0.00390625,
636
+ "pass@2": 0.007560483870967741,
637
+ "pass@4": 0.014168520578420465,
638
+ "pass@8": 0.024950562353231984,
639
+ "pass@16": 0.03939562476826103,
640
+ "pass@32": 0.05555555555555555
641
+ },
642
+ "23": {
643
+ "pass@1": 0.0004734848484848485,
644
+ "pass@2": 0.000946969696969697,
645
+ "pass@4": 0.001893939393939394,
646
+ "pass@8": 0.003787878787878788,
647
+ "pass@16": 0.007575757575757576,
648
+ "pass@32": 0.015151515151515152
649
+ },
650
+ "24": {
651
+ "pass@1": 0.006161971830985915,
652
+ "pass@2": 0.011869604725124942,
653
+ "pass@4": 0.02214314809882655,
654
+ "pass@8": 0.03936646700124117,
655
+ "pass@16": 0.06689052868689084,
656
+ "pass@32": 0.11267605633802817
657
+ },
658
+ "25": {
659
+ "pass@1": 0.011408730158730158,
660
+ "pass@2": 0.019169226830517153,
661
+ "pass@4": 0.028110599078341014,
662
+ "pass@8": 0.0348603001661956,
663
+ "pass@16": 0.03967604011614265,
664
+ "pass@32": 0.047619047619047616
665
+ },
666
+ "26": {
667
+ "pass@1": 0.0,
668
+ "pass@2": 0.0,
669
+ "pass@4": 0.0,
670
+ "pass@8": 0.0,
671
+ "pass@16": 0.0,
672
+ "pass@32": 0.0
673
+ },
674
+ "27": {
675
+ "pass@1": 0.00043402777777777775,
676
+ "pass@2": 0.0008680555555555555,
677
+ "pass@4": 0.001736111111111111,
678
+ "pass@8": 0.003472222222222222,
679
+ "pass@16": 0.006944444444444444,
680
+ "pass@32": 0.013888888888888888
681
+ },
682
+ "28": {
683
+ "pass@1": 0.0028735632183908046,
684
+ "pass@2": 0.005260474601408972,
685
+ "pass@4": 0.008887589019728177,
686
+ "pass@8": 0.013185885879683586,
687
+ "pass@16": 0.017202312204421828,
688
+ "pass@32": 0.022988505747126436
689
+ },
690
+ "29": {
691
+ "pass@1": 0.0004166666666666667,
692
+ "pass@2": 0.0008333333333333334,
693
+ "pass@4": 0.0016666666666666668,
694
+ "pass@8": 0.0033333333333333335,
695
+ "pass@16": 0.006666666666666667,
696
+ "pass@32": 0.013333333333333334
697
+ },
698
+ "30": {
699
+ "pass@1": 0.0004464285714285714,
700
+ "pass@2": 0.0008928571428571428,
701
+ "pass@4": 0.0017857142857142857,
702
+ "pass@8": 0.0035714285714285713,
703
+ "pass@16": 0.007142857142857143,
704
+ "pass@32": 0.014285714285714285
705
+ }
706
+ }
707
+ },
708
+ "teachers_in_school": {
709
+ "count": 19936,
710
+ "correct": 29,
711
+ "answer_accuracy": 0.0014546548956661316,
712
+ "avg_response_len": 290.34595706260035,
713
+ "resp_tokens_sum": 5788337,
714
+ "pass_at_k": {
715
+ "pass@1": 0.0014546548956661316,
716
+ "pass@2": 0.0028607673587738823,
717
+ "pass@4": 0.005535087139803991,
718
+ "pass@8": 0.0103836615322536,
719
+ "pass@16": 0.018455498083299257,
720
+ "pass@32": 0.030497592295345103
721
+ },
722
+ "per_op_pass_at_k": {
723
+ "21": {
724
+ "pass@1": 0.004326923076923077,
725
+ "pass@2": 0.008374689826302729,
726
+ "pass@4": 0.015694361256096516,
727
+ "pass@8": 0.027637545991272352,
728
+ "pass@16": 0.0436382305125353,
729
+ "pass@32": 0.06153846153846154
730
+ },
731
+ "22": {
732
+ "pass@1": 0.0030737704918032786,
733
+ "pass@2": 0.006114489687995769,
734
+ "pass@4": 0.012096774193548387,
735
+ "pass@8": 0.023664727657324167,
736
+ "pass@16": 0.0452141723955579,
737
+ "pass@32": 0.08196721311475409
738
+ },
739
+ "23": {
740
+ "pass@1": 0.0013392857142857143,
741
+ "pass@2": 0.0026497695852534552,
742
+ "pass@4": 0.005184331797235023,
743
+ "pass@8": 0.00990783410138249,
744
+ "pass@16": 0.017972350230414748,
745
+ "pass@32": 0.02857142857142857
746
+ },
747
+ "24": {
748
+ "pass@1": 0.0027901785714285715,
749
+ "pass@2": 0.005472350230414745,
750
+ "pass@4": 0.010527073732718894,
751
+ "pass@8": 0.019498847926267278,
752
+ "pass@16": 0.03369815668202765,
753
+ "pass@32": 0.05357142857142857
754
+ },
755
+ "25": {
756
+ "pass@1": 0.0004280821917808219,
757
+ "pass@2": 0.0008561643835616438,
758
+ "pass@4": 0.0017123287671232876,
759
+ "pass@8": 0.003424657534246575,
760
+ "pass@16": 0.00684931506849315,
761
+ "pass@32": 0.0136986301369863
762
+ },
763
+ "26": {
764
+ "pass@1": 0.0006127450980392157,
765
+ "pass@2": 0.0012254901960784314,
766
+ "pass@4": 0.0024509803921568627,
767
+ "pass@8": 0.004901960784313725,
768
+ "pass@16": 0.00980392156862745,
769
+ "pass@32": 0.0196078431372549
770
+ },
771
+ "27": {
772
+ "pass@1": 0.0005387931034482759,
773
+ "pass@2": 0.0010775862068965517,
774
+ "pass@4": 0.0021551724137931034,
775
+ "pass@8": 0.004310344827586207,
776
+ "pass@16": 0.008620689655172414,
777
+ "pass@32": 0.017241379310344827
778
+ },
779
+ "28": {
780
+ "pass@1": 0.0,
781
+ "pass@2": 0.0,
782
+ "pass@4": 0.0,
783
+ "pass@8": 0.0,
784
+ "pass@16": 0.0,
785
+ "pass@32": 0.0
786
+ },
787
+ "29": {
788
+ "pass@1": 0.0014423076923076924,
789
+ "pass@2": 0.002853598014888337,
790
+ "pass@4": 0.005583126550868486,
791
+ "pass@8": 0.010669975186104219,
792
+ "pass@16": 0.01935483870967742,
793
+ "pass@32": 0.03076923076923077
794
+ },
795
+ "30": {
796
+ "pass@1": 0.0,
797
+ "pass@2": 0.0,
798
+ "pass@4": 0.0,
799
+ "pass@8": 0.0,
800
+ "pass@16": 0.0,
801
+ "pass@32": 0.0
802
+ }
803
+ }
804
+ },
805
+ "movie_festival_awards": {
806
+ "count": 20992,
807
+ "correct": 69,
808
+ "answer_accuracy": 0.003286966463414634,
809
+ "avg_response_len": 328.9145388719512,
810
+ "resp_tokens_sum": 6904574,
811
+ "pass_at_k": {
812
+ "pass@1": 0.003286966463414634,
813
+ "pass@2": 0.0060207267899291895,
814
+ "pass@4": 0.010398961244200873,
815
+ "pass@8": 0.016888463323967543,
816
+ "pass@16": 0.026588695555185955,
817
+ "pass@32": 0.042682926829268296
818
+ },
819
+ "per_op_pass_at_k": {
820
+ "21": {
821
+ "pass@1": 0.005580357142857143,
822
+ "pass@2": 0.010440668202764977,
823
+ "pass@4": 0.01828420467185762,
824
+ "pass@8": 0.028176148101064676,
825
+ "pass@16": 0.03493961544573335,
826
+ "pass@32": 0.03571428571428571
827
+ },
828
+ "22": {
829
+ "pass@1": 0.004197761194029851,
830
+ "pass@2": 0.008094607607125662,
831
+ "pass@4": 0.015103846728537512,
832
+ "pass@8": 0.026700479803430013,
833
+ "pass@16": 0.04445237660418707,
834
+ "pass@32": 0.07462686567164178
835
+ },
836
+ "23": {
837
+ "pass@1": 0.00830078125,
838
+ "pass@2": 0.014112903225806451,
839
+ "pass@4": 0.021564324944382644,
840
+ "pass@8": 0.030255666017797552,
841
+ "pass@16": 0.04309473287549774,
842
+ "pass@32": 0.0625
843
+ },
844
+ "24": {
845
+ "pass@1": 0.0021404109589041095,
846
+ "pass@2": 0.0042532037118868755,
847
+ "pass@4": 0.008395934600088379,
848
+ "pass@8": 0.016349977905435263,
849
+ "pass@16": 0.030932390631904547,
850
+ "pass@32": 0.0547945205479452
851
+ },
852
+ "25": {
853
+ "pass@1": 0.001953125,
854
+ "pass@2": 0.00390625,
855
+ "pass@4": 0.0078125,
856
+ "pass@8": 0.015625,
857
+ "pass@16": 0.03125,
858
+ "pass@32": 0.0625
859
+ },
860
+ "26": {
861
+ "pass@1": 0.004518072289156626,
862
+ "pass@2": 0.007700155460551886,
863
+ "pass@4": 0.01154897677472962,
864
+ "pass@8": 0.014827153226579631,
865
+ "pass@16": 0.018071881276617813,
866
+ "pass@32": 0.024096385542168676
867
+ },
868
+ "27": {
869
+ "pass@1": 0.0035714285714285713,
870
+ "pass@2": 0.0067108294930875575,
871
+ "pass@4": 0.011918004131574764,
872
+ "pass@8": 0.01930672528559069,
873
+ "pass@16": 0.028445186009145966,
874
+ "pass@32": 0.04285714285714286
875
+ },
876
+ "28": {
877
+ "pass@1": 0.0,
878
+ "pass@2": 0.0,
879
+ "pass@4": 0.0,
880
+ "pass@8": 0.0,
881
+ "pass@16": 0.0,
882
+ "pass@32": 0.0
883
+ },
884
+ "29": {
885
+ "pass@1": 0.0010416666666666667,
886
+ "pass@2": 0.0020833333333333333,
887
+ "pass@4": 0.004166666666666667,
888
+ "pass@8": 0.008333333333333333,
889
+ "pass@16": 0.016666666666666666,
890
+ "pass@32": 0.03333333333333333
891
+ },
892
+ "30": {
893
+ "pass@1": 0.000946969696969697,
894
+ "pass@2": 0.001893939393939394,
895
+ "pass@4": 0.003787878787878788,
896
+ "pass@8": 0.007575757575757576,
897
+ "pass@16": 0.015151515151515152,
898
+ "pass@32": 0.030303030303030304
899
+ }
900
+ }
901
+ }
902
+ }
903
+ }
904
+ }