ZachW commited on
Commit
acc618d
·
verified ·
1 Parent(s): 480ff4d

Upload extrapolation artifact: pretrain_ckpt18809 far-OOD metrics

Browse files
020305/evals/far_ood/pretrain_ckpt18809/pretrain_ckpt18809_metrics.json ADDED
@@ -0,0 +1,904 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "pretrain_ckpt18809",
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": 0.000875,
48
+ "avg_response_len": 229.355625,
49
+ "avg_loss": 0.0,
50
+ "count": 64000,
51
+ "validation_metrics": {},
52
+ "pass_at_k": {
53
+ "pass@1": 0.000875,
54
+ "pass@2": 0.001679435483870968,
55
+ "pass@4": 0.0031146551724137933,
56
+ "pass@8": 0.005484425429964918,
57
+ "pass@16": 0.009077221604251636,
58
+ "pass@32": 0.014
59
+ },
60
+ "per_op_pass_at_k": {
61
+ "21": {
62
+ "pass@1": 0.0009375,
63
+ "pass@2": 0.0018548387096774188,
64
+ "pass@4": 0.0036290322580645163,
65
+ "pass@8": 0.0069354838709677425,
66
+ "pass@16": 0.012580645161290323,
67
+ "pass@32": 0.02
68
+ },
69
+ "22": {
70
+ "pass@1": 0.00046875,
71
+ "pass@2": 0.0009375,
72
+ "pass@4": 0.001875,
73
+ "pass@8": 0.00375,
74
+ "pass@16": 0.0075,
75
+ "pass@32": 0.015
76
+ },
77
+ "23": {
78
+ "pass@1": 0.00046875,
79
+ "pass@2": 0.0009375,
80
+ "pass@4": 0.001875,
81
+ "pass@8": 0.00375,
82
+ "pass@16": 0.0075,
83
+ "pass@32": 0.015
84
+ },
85
+ "24": {
86
+ "pass@1": 0.00046875,
87
+ "pass@2": 0.0009375,
88
+ "pass@4": 0.001875,
89
+ "pass@8": 0.00375,
90
+ "pass@16": 0.0075,
91
+ "pass@32": 0.015
92
+ },
93
+ "25": {
94
+ "pass@1": 0.0009375,
95
+ "pass@2": 0.0018346774193548382,
96
+ "pass@4": 0.0035120967741935484,
97
+ "pass@8": 0.006427419354838711,
98
+ "pass@16": 0.010725806451612903,
99
+ "pass@32": 0.015
100
+ },
101
+ "26": {
102
+ "pass@1": 0.00171875,
103
+ "pass@2": 0.0032157258064516125,
104
+ "pass@4": 0.005680617352614015,
105
+ "pass@8": 0.009203602293146231,
106
+ "pass@16": 0.013773328389568658,
107
+ "pass@32": 0.02
108
+ },
109
+ "27": {
110
+ "pass@1": 0.0,
111
+ "pass@2": 0.0,
112
+ "pass@4": 0.0,
113
+ "pass@8": 0.0,
114
+ "pass@16": 0.0,
115
+ "pass@32": 0.0
116
+ },
117
+ "28": {
118
+ "pass@1": 0.00265625,
119
+ "pass@2": 0.004929435483870968,
120
+ "pass@4": 0.008562708565072302,
121
+ "pass@8": 0.013350329425857793,
122
+ "pass@16": 0.01796662958843159,
123
+ "pass@32": 0.02
124
+ },
125
+ "29": {
126
+ "pass@1": 0.0003125,
127
+ "pass@2": 0.0006149193548387094,
128
+ "pass@4": 0.001189516129032258,
129
+ "pass@8": 0.0022177419354838714,
130
+ "pass@16": 0.0037903225806451613,
131
+ "pass@32": 0.005
132
+ },
133
+ "30": {
134
+ "pass@1": 0.00078125,
135
+ "pass@2": 0.0015322580645161289,
136
+ "pass@4": 0.0029475806451612903,
137
+ "pass@8": 0.005459677419354838,
138
+ "pass@16": 0.009435483870967742,
139
+ "pass@32": 0.015
140
+ }
141
+ },
142
+ "per_op_accuracy": {
143
+ "21": 0.0009375,
144
+ "22": 0.00046875,
145
+ "23": 0.00046875,
146
+ "24": 0.00046875,
147
+ "25": 0.0009375,
148
+ "26": 0.00171875,
149
+ "27": 0.0,
150
+ "28": 0.00265625,
151
+ "29": 0.0003125,
152
+ "30": 0.00078125
153
+ },
154
+ "per_op_avg_response_len": {
155
+ "21": 226.6790625,
156
+ "22": 229.86078125,
157
+ "23": 227.5865625,
158
+ "24": 230.08515625,
159
+ "25": 226.731875,
160
+ "26": 233.328125,
161
+ "27": 230.444375,
162
+ "28": 229.16796875,
163
+ "29": 230.10484375,
164
+ "30": 229.5675
165
+ },
166
+ "per_op_avg_loss": {},
167
+ "per_op_length": {
168
+ "21": 226.6790625,
169
+ "22": 229.86078125,
170
+ "23": 227.5865625,
171
+ "24": 230.08515625,
172
+ "25": 226.731875,
173
+ "26": 233.328125,
174
+ "27": 230.444375,
175
+ "28": 229.16796875,
176
+ "29": 230.10484375,
177
+ "30": 229.5675
178
+ },
179
+ "per_op_loss": {},
180
+ "per_template": {
181
+ "crazy_zootopia": {
182
+ "count": 23072,
183
+ "correct": 13,
184
+ "answer_accuracy": 0.0005634535367545076,
185
+ "avg_response_len": 212.57216539528432,
186
+ "resp_tokens_sum": 4904465,
187
+ "pass_at_k": {
188
+ "pass@1": 0.0005634535367545076,
189
+ "pass@2": 0.0011045367097669006,
190
+ "pass@4": 0.0021218290009395553,
191
+ "pass@8": 0.003912576618495816,
192
+ "pass@16": 0.006643998031407991,
193
+ "pass@32": 0.009708737864077669
194
+ },
195
+ "per_op_pass_at_k": {
196
+ "21": {
197
+ "pass@1": 0.0011867088607594937,
198
+ "pass@2": 0.0023478971008574923,
199
+ "pass@4": 0.004593711719069008,
200
+ "pass@8": 0.008779093507554105,
201
+ "pass@16": 0.01592486729277256,
202
+ "pass@32": 0.02531645569620253
203
+ },
204
+ "22": {
205
+ "pass@1": 0.0008680555555555555,
206
+ "pass@2": 0.001736111111111111,
207
+ "pass@4": 0.003472222222222222,
208
+ "pass@8": 0.006944444444444444,
209
+ "pass@16": 0.013888888888888888,
210
+ "pass@32": 0.027777777777777776
211
+ },
212
+ "23": {
213
+ "pass@1": 0.0,
214
+ "pass@2": 0.0,
215
+ "pass@4": 0.0,
216
+ "pass@8": 0.0,
217
+ "pass@16": 0.0,
218
+ "pass@32": 0.0
219
+ },
220
+ "24": {
221
+ "pass@1": 0.0,
222
+ "pass@2": 0.0,
223
+ "pass@4": 0.0,
224
+ "pass@8": 0.0,
225
+ "pass@16": 0.0,
226
+ "pass@32": 0.0
227
+ },
228
+ "25": {
229
+ "pass@1": 0.00248015873015873,
230
+ "pass@2": 0.004832309267793137,
231
+ "pass@4": 0.009165386584741423,
232
+ "pass@8": 0.01643625192012289,
233
+ "pass@16": 0.026113671274961597,
234
+ "pass@32": 0.031746031746031744
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": 0.0,
246
+ "pass@2": 0.0,
247
+ "pass@4": 0.0,
248
+ "pass@8": 0.0,
249
+ "pass@16": 0.0,
250
+ "pass@32": 0.0
251
+ },
252
+ "28": {
253
+ "pass@1": 0.0010775862068965517,
254
+ "pass@2": 0.0020856507230255835,
255
+ "pass@4": 0.0039024842417500925,
256
+ "pass@8": 0.006803856136447905,
257
+ "pass@16": 0.010196514645902854,
258
+ "pass@32": 0.011494252873563218
259
+ },
260
+ "29": {
261
+ "pass@1": 0.0,
262
+ "pass@2": 0.0,
263
+ "pass@4": 0.0,
264
+ "pass@8": 0.0,
265
+ "pass@16": 0.0,
266
+ "pass@32": 0.0
267
+ },
268
+ "30": {
269
+ "pass@1": 0.0,
270
+ "pass@2": 0.0,
271
+ "pass@4": 0.0,
272
+ "pass@8": 0.0,
273
+ "pass@16": 0.0,
274
+ "pass@32": 0.0
275
+ }
276
+ }
277
+ },
278
+ "teachers_in_school": {
279
+ "count": 19936,
280
+ "correct": 30,
281
+ "answer_accuracy": 0.0015048154093097914,
282
+ "avg_response_len": 223.00491573033707,
283
+ "resp_tokens_sum": 4445826,
284
+ "pass_at_k": {
285
+ "pass@1": 0.0015048154093097914,
286
+ "pass@2": 0.0028801843317972347,
287
+ "pass@4": 0.005318375866175543,
288
+ "pass@8": 0.009321288377793106,
289
+ "pass@16": 0.015424486275994192,
290
+ "pass@32": 0.024077046548956663
291
+ },
292
+ "per_op_pass_at_k": {
293
+ "21": {
294
+ "pass@1": 0.0014423076923076924,
295
+ "pass@2": 0.002853598014888337,
296
+ "pass@4": 0.005583126550868486,
297
+ "pass@8": 0.010669975186104219,
298
+ "pass@16": 0.01935483870967742,
299
+ "pass@32": 0.03076923076923077
300
+ },
301
+ "22": {
302
+ "pass@1": 0.0,
303
+ "pass@2": 0.0,
304
+ "pass@4": 0.0,
305
+ "pass@8": 0.0,
306
+ "pass@16": 0.0,
307
+ "pass@32": 0.0
308
+ },
309
+ "23": {
310
+ "pass@1": 0.0004464285714285714,
311
+ "pass@2": 0.0008928571428571428,
312
+ "pass@4": 0.0017857142857142857,
313
+ "pass@8": 0.0035714285714285713,
314
+ "pass@16": 0.007142857142857143,
315
+ "pass@32": 0.014285714285714285
316
+ },
317
+ "24": {
318
+ "pass@1": 0.0011160714285714285,
319
+ "pass@2": 0.002232142857142857,
320
+ "pass@4": 0.004464285714285714,
321
+ "pass@8": 0.008928571428571428,
322
+ "pass@16": 0.017857142857142856,
323
+ "pass@32": 0.03571428571428571
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": 0.0024509803921568627,
335
+ "pass@2": 0.004862428842504742,
336
+ "pass@4": 0.009566729917773562,
337
+ "pass@8": 0.018500948766603416,
338
+ "pass@16": 0.034471853257432,
339
+ "pass@32": 0.058823529411764705
340
+ },
341
+ "27": {
342
+ "pass@1": 0.0,
343
+ "pass@2": 0.0,
344
+ "pass@4": 0.0,
345
+ "pass@8": 0.0,
346
+ "pass@16": 0.0,
347
+ "pass@32": 0.0
348
+ },
349
+ "28": {
350
+ "pass@1": 0.007291666666666667,
351
+ "pass@2": 0.013407258064516129,
352
+ "pass@4": 0.022883759733036704,
353
+ "pass@8": 0.03463550668834318,
354
+ "pass@16": 0.04510381905821283,
355
+ "pass@32": 0.05
356
+ },
357
+ "29": {
358
+ "pass@1": 0.0,
359
+ "pass@2": 0.0,
360
+ "pass@4": 0.0,
361
+ "pass@8": 0.0,
362
+ "pass@16": 0.0,
363
+ "pass@32": 0.0
364
+ },
365
+ "30": {
366
+ "pass@1": 0.00244140625,
367
+ "pass@2": 0.0047883064516129024,
368
+ "pass@4": 0.009211189516129032,
369
+ "pass@8": 0.01706149193548387,
370
+ "pass@16": 0.029485887096774195,
371
+ "pass@32": 0.046875
372
+ }
373
+ }
374
+ },
375
+ "movie_festival_awards": {
376
+ "count": 20992,
377
+ "correct": 13,
378
+ "answer_accuracy": 0.0006192835365853659,
379
+ "avg_response_len": 253.8333174542683,
380
+ "resp_tokens_sum": 5328469,
381
+ "pass_at_k": {
382
+ "pass@1": 0.0006192835365853659,
383
+ "pass@2": 0.0011709529897718331,
384
+ "pass@4": 0.002112993095309151,
385
+ "pass@8": 0.003568171430837266,
386
+ "pass@16": 0.005723575728527029,
387
+ "pass@32": 0.009146341463414634
388
+ },
389
+ "per_op_pass_at_k": {
390
+ "21": {
391
+ "pass@1": 0.0,
392
+ "pass@2": 0.0,
393
+ "pass@4": 0.0,
394
+ "pass@8": 0.0,
395
+ "pass@16": 0.0,
396
+ "pass@32": 0.0
397
+ },
398
+ "22": {
399
+ "pass@1": 0.0004664179104477612,
400
+ "pass@2": 0.0009328358208955224,
401
+ "pass@4": 0.0018656716417910447,
402
+ "pass@8": 0.0037313432835820895,
403
+ "pass@16": 0.007462686567164179,
404
+ "pass@32": 0.014925373134328358
405
+ },
406
+ "23": {
407
+ "pass@1": 0.0009765625,
408
+ "pass@2": 0.001953125,
409
+ "pass@4": 0.00390625,
410
+ "pass@8": 0.0078125,
411
+ "pass@16": 0.015625,
412
+ "pass@32": 0.03125
413
+ },
414
+ "24": {
415
+ "pass@1": 0.0004280821917808219,
416
+ "pass@2": 0.0008561643835616438,
417
+ "pass@4": 0.0017123287671232876,
418
+ "pass@8": 0.003424657534246575,
419
+ "pass@16": 0.00684931506849315,
420
+ "pass@32": 0.0136986301369863
421
+ },
422
+ "25": {
423
+ "pass@1": 0.0,
424
+ "pass@2": 0.0,
425
+ "pass@4": 0.0,
426
+ "pass@8": 0.0,
427
+ "pass@16": 0.0,
428
+ "pass@32": 0.0
429
+ },
430
+ "26": {
431
+ "pass@1": 0.0026355421686746986,
432
+ "pass@2": 0.0047609794014768754,
433
+ "pass@4": 0.007809882466462065,
434
+ "pass@8": 0.01080930206665629,
435
+ "pass@16": 0.012007242913068664,
436
+ "pass@32": 0.012048192771084338
437
+ },
438
+ "27": {
439
+ "pass@1": 0.0,
440
+ "pass@2": 0.0,
441
+ "pass@4": 0.0,
442
+ "pass@8": 0.0,
443
+ "pass@16": 0.0,
444
+ "pass@32": 0.0
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": 0.0020497311827956982,
457
+ "pass@4": 0.0039650537634408605,
458
+ "pass@8": 0.007392473118279571,
459
+ "pass@16": 0.012634408602150538,
460
+ "pass@32": 0.016666666666666666
461
+ },
462
+ "30": {
463
+ "pass@1": 0.0,
464
+ "pass@2": 0.0,
465
+ "pass@4": 0.0,
466
+ "pass@8": 0.0,
467
+ "pass@16": 0.0,
468
+ "pass@32": 0.0
469
+ }
470
+ }
471
+ }
472
+ }
473
+ },
474
+ "total": {
475
+ "answer_accuracy": 0.000875,
476
+ "avg_loss": 0.0,
477
+ "count": 64000,
478
+ "validation_metrics": {
479
+ "id": {},
480
+ "ood": {}
481
+ },
482
+ "pass_at_k": {
483
+ "pass@1": 0.000875,
484
+ "pass@2": 0.001679435483870968,
485
+ "pass@4": 0.0031146551724137933,
486
+ "pass@8": 0.005484425429964918,
487
+ "pass@16": 0.009077221604251636,
488
+ "pass@32": 0.014
489
+ },
490
+ "per_op_pass_at_k": {
491
+ "21": {
492
+ "pass@1": 0.0009375,
493
+ "pass@2": 0.0018548387096774188,
494
+ "pass@4": 0.0036290322580645163,
495
+ "pass@8": 0.0069354838709677425,
496
+ "pass@16": 0.012580645161290323,
497
+ "pass@32": 0.02
498
+ },
499
+ "22": {
500
+ "pass@1": 0.00046875,
501
+ "pass@2": 0.0009375,
502
+ "pass@4": 0.001875,
503
+ "pass@8": 0.00375,
504
+ "pass@16": 0.0075,
505
+ "pass@32": 0.015
506
+ },
507
+ "23": {
508
+ "pass@1": 0.00046875,
509
+ "pass@2": 0.0009375,
510
+ "pass@4": 0.001875,
511
+ "pass@8": 0.00375,
512
+ "pass@16": 0.0075,
513
+ "pass@32": 0.015
514
+ },
515
+ "24": {
516
+ "pass@1": 0.00046875,
517
+ "pass@2": 0.0009375,
518
+ "pass@4": 0.001875,
519
+ "pass@8": 0.00375,
520
+ "pass@16": 0.0075,
521
+ "pass@32": 0.015
522
+ },
523
+ "25": {
524
+ "pass@1": 0.0009375,
525
+ "pass@2": 0.0018346774193548382,
526
+ "pass@4": 0.0035120967741935484,
527
+ "pass@8": 0.006427419354838711,
528
+ "pass@16": 0.010725806451612903,
529
+ "pass@32": 0.015
530
+ },
531
+ "26": {
532
+ "pass@1": 0.00171875,
533
+ "pass@2": 0.0032157258064516125,
534
+ "pass@4": 0.005680617352614015,
535
+ "pass@8": 0.009203602293146231,
536
+ "pass@16": 0.013773328389568658,
537
+ "pass@32": 0.02
538
+ },
539
+ "27": {
540
+ "pass@1": 0.0,
541
+ "pass@2": 0.0,
542
+ "pass@4": 0.0,
543
+ "pass@8": 0.0,
544
+ "pass@16": 0.0,
545
+ "pass@32": 0.0
546
+ },
547
+ "28": {
548
+ "pass@1": 0.00265625,
549
+ "pass@2": 0.004929435483870968,
550
+ "pass@4": 0.008562708565072302,
551
+ "pass@8": 0.013350329425857793,
552
+ "pass@16": 0.01796662958843159,
553
+ "pass@32": 0.02
554
+ },
555
+ "29": {
556
+ "pass@1": 0.0003125,
557
+ "pass@2": 0.0006149193548387094,
558
+ "pass@4": 0.001189516129032258,
559
+ "pass@8": 0.0022177419354838714,
560
+ "pass@16": 0.0037903225806451613,
561
+ "pass@32": 0.005
562
+ },
563
+ "30": {
564
+ "pass@1": 0.00078125,
565
+ "pass@2": 0.0015322580645161289,
566
+ "pass@4": 0.0029475806451612903,
567
+ "pass@8": 0.005459677419354838,
568
+ "pass@16": 0.009435483870967742,
569
+ "pass@32": 0.015
570
+ }
571
+ },
572
+ "per_op_accuracy": {
573
+ "21": 0.0009375,
574
+ "22": 0.00046875,
575
+ "23": 0.00046875,
576
+ "24": 0.00046875,
577
+ "25": 0.0009375,
578
+ "26": 0.00171875,
579
+ "27": 0.0,
580
+ "28": 0.00265625,
581
+ "29": 0.0003125,
582
+ "30": 0.00078125
583
+ },
584
+ "per_op_avg_response_len": {
585
+ "21": 226.6790625,
586
+ "22": 229.86078125,
587
+ "23": 227.5865625,
588
+ "24": 230.08515625,
589
+ "25": 226.731875,
590
+ "26": 233.328125,
591
+ "27": 230.444375,
592
+ "28": 229.16796875,
593
+ "29": 230.10484375,
594
+ "30": 229.5675
595
+ },
596
+ "per_op_avg_loss": {},
597
+ "per_op_length": {
598
+ "21": 226.6790625,
599
+ "22": 229.86078125,
600
+ "23": 227.5865625,
601
+ "24": 230.08515625,
602
+ "25": 226.731875,
603
+ "26": 233.328125,
604
+ "27": 230.444375,
605
+ "28": 229.16796875,
606
+ "29": 230.10484375,
607
+ "30": 229.5675
608
+ },
609
+ "per_op_loss": {},
610
+ "per_template": {
611
+ "crazy_zootopia": {
612
+ "count": 23072,
613
+ "correct": 13,
614
+ "answer_accuracy": 0.0005634535367545076,
615
+ "avg_response_len": 212.57216539528432,
616
+ "resp_tokens_sum": 4904465,
617
+ "pass_at_k": {
618
+ "pass@1": 0.0005634535367545076,
619
+ "pass@2": 0.0011045367097669006,
620
+ "pass@4": 0.0021218290009395553,
621
+ "pass@8": 0.003912576618495816,
622
+ "pass@16": 0.006643998031407991,
623
+ "pass@32": 0.009708737864077669
624
+ },
625
+ "per_op_pass_at_k": {
626
+ "21": {
627
+ "pass@1": 0.0011867088607594937,
628
+ "pass@2": 0.0023478971008574923,
629
+ "pass@4": 0.004593711719069008,
630
+ "pass@8": 0.008779093507554105,
631
+ "pass@16": 0.01592486729277256,
632
+ "pass@32": 0.02531645569620253
633
+ },
634
+ "22": {
635
+ "pass@1": 0.0008680555555555555,
636
+ "pass@2": 0.001736111111111111,
637
+ "pass@4": 0.003472222222222222,
638
+ "pass@8": 0.006944444444444444,
639
+ "pass@16": 0.013888888888888888,
640
+ "pass@32": 0.027777777777777776
641
+ },
642
+ "23": {
643
+ "pass@1": 0.0,
644
+ "pass@2": 0.0,
645
+ "pass@4": 0.0,
646
+ "pass@8": 0.0,
647
+ "pass@16": 0.0,
648
+ "pass@32": 0.0
649
+ },
650
+ "24": {
651
+ "pass@1": 0.0,
652
+ "pass@2": 0.0,
653
+ "pass@4": 0.0,
654
+ "pass@8": 0.0,
655
+ "pass@16": 0.0,
656
+ "pass@32": 0.0
657
+ },
658
+ "25": {
659
+ "pass@1": 0.00248015873015873,
660
+ "pass@2": 0.004832309267793137,
661
+ "pass@4": 0.009165386584741423,
662
+ "pass@8": 0.01643625192012289,
663
+ "pass@16": 0.026113671274961597,
664
+ "pass@32": 0.031746031746031744
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": 0.0,
676
+ "pass@2": 0.0,
677
+ "pass@4": 0.0,
678
+ "pass@8": 0.0,
679
+ "pass@16": 0.0,
680
+ "pass@32": 0.0
681
+ },
682
+ "28": {
683
+ "pass@1": 0.0010775862068965517,
684
+ "pass@2": 0.0020856507230255835,
685
+ "pass@4": 0.0039024842417500925,
686
+ "pass@8": 0.006803856136447905,
687
+ "pass@16": 0.010196514645902854,
688
+ "pass@32": 0.011494252873563218
689
+ },
690
+ "29": {
691
+ "pass@1": 0.0,
692
+ "pass@2": 0.0,
693
+ "pass@4": 0.0,
694
+ "pass@8": 0.0,
695
+ "pass@16": 0.0,
696
+ "pass@32": 0.0
697
+ },
698
+ "30": {
699
+ "pass@1": 0.0,
700
+ "pass@2": 0.0,
701
+ "pass@4": 0.0,
702
+ "pass@8": 0.0,
703
+ "pass@16": 0.0,
704
+ "pass@32": 0.0
705
+ }
706
+ }
707
+ },
708
+ "teachers_in_school": {
709
+ "count": 19936,
710
+ "correct": 30,
711
+ "answer_accuracy": 0.0015048154093097914,
712
+ "avg_response_len": 223.00491573033707,
713
+ "resp_tokens_sum": 4445826,
714
+ "pass_at_k": {
715
+ "pass@1": 0.0015048154093097914,
716
+ "pass@2": 0.0028801843317972347,
717
+ "pass@4": 0.005318375866175543,
718
+ "pass@8": 0.009321288377793106,
719
+ "pass@16": 0.015424486275994192,
720
+ "pass@32": 0.024077046548956663
721
+ },
722
+ "per_op_pass_at_k": {
723
+ "21": {
724
+ "pass@1": 0.0014423076923076924,
725
+ "pass@2": 0.002853598014888337,
726
+ "pass@4": 0.005583126550868486,
727
+ "pass@8": 0.010669975186104219,
728
+ "pass@16": 0.01935483870967742,
729
+ "pass@32": 0.03076923076923077
730
+ },
731
+ "22": {
732
+ "pass@1": 0.0,
733
+ "pass@2": 0.0,
734
+ "pass@4": 0.0,
735
+ "pass@8": 0.0,
736
+ "pass@16": 0.0,
737
+ "pass@32": 0.0
738
+ },
739
+ "23": {
740
+ "pass@1": 0.0004464285714285714,
741
+ "pass@2": 0.0008928571428571428,
742
+ "pass@4": 0.0017857142857142857,
743
+ "pass@8": 0.0035714285714285713,
744
+ "pass@16": 0.007142857142857143,
745
+ "pass@32": 0.014285714285714285
746
+ },
747
+ "24": {
748
+ "pass@1": 0.0011160714285714285,
749
+ "pass@2": 0.002232142857142857,
750
+ "pass@4": 0.004464285714285714,
751
+ "pass@8": 0.008928571428571428,
752
+ "pass@16": 0.017857142857142856,
753
+ "pass@32": 0.03571428571428571
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": 0.0024509803921568627,
765
+ "pass@2": 0.004862428842504742,
766
+ "pass@4": 0.009566729917773562,
767
+ "pass@8": 0.018500948766603416,
768
+ "pass@16": 0.034471853257432,
769
+ "pass@32": 0.058823529411764705
770
+ },
771
+ "27": {
772
+ "pass@1": 0.0,
773
+ "pass@2": 0.0,
774
+ "pass@4": 0.0,
775
+ "pass@8": 0.0,
776
+ "pass@16": 0.0,
777
+ "pass@32": 0.0
778
+ },
779
+ "28": {
780
+ "pass@1": 0.007291666666666667,
781
+ "pass@2": 0.013407258064516129,
782
+ "pass@4": 0.022883759733036704,
783
+ "pass@8": 0.03463550668834318,
784
+ "pass@16": 0.04510381905821283,
785
+ "pass@32": 0.05
786
+ },
787
+ "29": {
788
+ "pass@1": 0.0,
789
+ "pass@2": 0.0,
790
+ "pass@4": 0.0,
791
+ "pass@8": 0.0,
792
+ "pass@16": 0.0,
793
+ "pass@32": 0.0
794
+ },
795
+ "30": {
796
+ "pass@1": 0.00244140625,
797
+ "pass@2": 0.0047883064516129024,
798
+ "pass@4": 0.009211189516129032,
799
+ "pass@8": 0.01706149193548387,
800
+ "pass@16": 0.029485887096774195,
801
+ "pass@32": 0.046875
802
+ }
803
+ }
804
+ },
805
+ "movie_festival_awards": {
806
+ "count": 20992,
807
+ "correct": 13,
808
+ "answer_accuracy": 0.0006192835365853659,
809
+ "avg_response_len": 253.8333174542683,
810
+ "resp_tokens_sum": 5328469,
811
+ "pass_at_k": {
812
+ "pass@1": 0.0006192835365853659,
813
+ "pass@2": 0.0011709529897718331,
814
+ "pass@4": 0.002112993095309151,
815
+ "pass@8": 0.003568171430837266,
816
+ "pass@16": 0.005723575728527029,
817
+ "pass@32": 0.009146341463414634
818
+ },
819
+ "per_op_pass_at_k": {
820
+ "21": {
821
+ "pass@1": 0.0,
822
+ "pass@2": 0.0,
823
+ "pass@4": 0.0,
824
+ "pass@8": 0.0,
825
+ "pass@16": 0.0,
826
+ "pass@32": 0.0
827
+ },
828
+ "22": {
829
+ "pass@1": 0.0004664179104477612,
830
+ "pass@2": 0.0009328358208955224,
831
+ "pass@4": 0.0018656716417910447,
832
+ "pass@8": 0.0037313432835820895,
833
+ "pass@16": 0.007462686567164179,
834
+ "pass@32": 0.014925373134328358
835
+ },
836
+ "23": {
837
+ "pass@1": 0.0009765625,
838
+ "pass@2": 0.001953125,
839
+ "pass@4": 0.00390625,
840
+ "pass@8": 0.0078125,
841
+ "pass@16": 0.015625,
842
+ "pass@32": 0.03125
843
+ },
844
+ "24": {
845
+ "pass@1": 0.0004280821917808219,
846
+ "pass@2": 0.0008561643835616438,
847
+ "pass@4": 0.0017123287671232876,
848
+ "pass@8": 0.003424657534246575,
849
+ "pass@16": 0.00684931506849315,
850
+ "pass@32": 0.0136986301369863
851
+ },
852
+ "25": {
853
+ "pass@1": 0.0,
854
+ "pass@2": 0.0,
855
+ "pass@4": 0.0,
856
+ "pass@8": 0.0,
857
+ "pass@16": 0.0,
858
+ "pass@32": 0.0
859
+ },
860
+ "26": {
861
+ "pass@1": 0.0026355421686746986,
862
+ "pass@2": 0.0047609794014768754,
863
+ "pass@4": 0.007809882466462065,
864
+ "pass@8": 0.01080930206665629,
865
+ "pass@16": 0.012007242913068664,
866
+ "pass@32": 0.012048192771084338
867
+ },
868
+ "27": {
869
+ "pass@1": 0.0,
870
+ "pass@2": 0.0,
871
+ "pass@4": 0.0,
872
+ "pass@8": 0.0,
873
+ "pass@16": 0.0,
874
+ "pass@32": 0.0
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": 0.0020497311827956982,
887
+ "pass@4": 0.0039650537634408605,
888
+ "pass@8": 0.007392473118279571,
889
+ "pass@16": 0.012634408602150538,
890
+ "pass@32": 0.016666666666666666
891
+ },
892
+ "30": {
893
+ "pass@1": 0.0,
894
+ "pass@2": 0.0,
895
+ "pass@4": 0.0,
896
+ "pass@8": 0.0,
897
+ "pass@16": 0.0,
898
+ "pass@32": 0.0
899
+ }
900
+ }
901
+ }
902
+ }
903
+ }
904
+ }