ZachW commited on
Commit
6402b41
·
verified ·
1 Parent(s): 0d7243f

Upload extrapolation artifact: opd_ckpt18809_refkl_op11_14_s200_merged_hf far-OOD summary JSON

Browse files
020305/evals/far_ood/opd_ckpt18809_refkl_op11_14_s200_merged_hf/summary.json ADDED
@@ -0,0 +1,309 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "checkpoint": "opd_ckpt18809_refkl_op11_14_s200_merged_hf",
4
+ "id_acc": 0.0,
5
+ "ood_acc": 0.002484375,
6
+ "total_acc": 0.002484375,
7
+ "id_avg_loss": 0.0,
8
+ "ood_avg_loss": 0.0,
9
+ "total_avg_loss": 0.0,
10
+ "id_avg_resp_len": 0.0,
11
+ "ood_avg_resp_len": 297.913984375,
12
+ "count_id": 0,
13
+ "count_ood": 64000,
14
+ "count_total": 64000,
15
+ "template_metrics": {
16
+ "crazy_zootopia": {
17
+ "count": 23072,
18
+ "correct": 61,
19
+ "answer_accuracy": 0.0026438973647711513,
20
+ "avg_response_len": 276.247572815534,
21
+ "resp_tokens_sum": 6373584,
22
+ "pass_at_k": {
23
+ "pass@1": 0.0026438973647711513,
24
+ "pass@2": 0.004840387454700014,
25
+ "pass@4": 0.008337704553834667,
26
+ "pass@8": 0.013432497404332345,
27
+ "pass@16": 0.020918036351159127,
28
+ "pass@32": 0.033287101248266296
29
+ },
30
+ "per_op_pass_at_k": {
31
+ "21": {
32
+ "pass@1": 0.0011867088607594937,
33
+ "pass@2": 0.0023734177215189874,
34
+ "pass@4": 0.004746835443037975,
35
+ "pass@8": 0.00949367088607595,
36
+ "pass@16": 0.0189873417721519,
37
+ "pass@32": 0.0379746835443038
38
+ },
39
+ "22": {
40
+ "pass@1": 0.00390625,
41
+ "pass@2": 0.007560483870967741,
42
+ "pass@4": 0.014168520578420465,
43
+ "pass@8": 0.024950562353231984,
44
+ "pass@16": 0.03939562476826103,
45
+ "pass@32": 0.05555555555555555
46
+ },
47
+ "23": {
48
+ "pass@1": 0.0004734848484848485,
49
+ "pass@2": 0.000946969696969697,
50
+ "pass@4": 0.001893939393939394,
51
+ "pass@8": 0.003787878787878788,
52
+ "pass@16": 0.007575757575757576,
53
+ "pass@32": 0.015151515151515152
54
+ },
55
+ "24": {
56
+ "pass@1": 0.006161971830985915,
57
+ "pass@2": 0.011869604725124942,
58
+ "pass@4": 0.02214314809882655,
59
+ "pass@8": 0.03936646700124117,
60
+ "pass@16": 0.06689052868689084,
61
+ "pass@32": 0.11267605633802817
62
+ },
63
+ "25": {
64
+ "pass@1": 0.011408730158730158,
65
+ "pass@2": 0.019169226830517153,
66
+ "pass@4": 0.028110599078341014,
67
+ "pass@8": 0.0348603001661956,
68
+ "pass@16": 0.03967604011614265,
69
+ "pass@32": 0.047619047619047616
70
+ },
71
+ "26": {
72
+ "pass@1": 0.0,
73
+ "pass@2": 0.0,
74
+ "pass@4": 0.0,
75
+ "pass@8": 0.0,
76
+ "pass@16": 0.0,
77
+ "pass@32": 0.0
78
+ },
79
+ "27": {
80
+ "pass@1": 0.00043402777777777775,
81
+ "pass@2": 0.0008680555555555555,
82
+ "pass@4": 0.001736111111111111,
83
+ "pass@8": 0.003472222222222222,
84
+ "pass@16": 0.006944444444444444,
85
+ "pass@32": 0.013888888888888888
86
+ },
87
+ "28": {
88
+ "pass@1": 0.0028735632183908046,
89
+ "pass@2": 0.005260474601408972,
90
+ "pass@4": 0.008887589019728177,
91
+ "pass@8": 0.013185885879683586,
92
+ "pass@16": 0.017202312204421828,
93
+ "pass@32": 0.022988505747126436
94
+ },
95
+ "29": {
96
+ "pass@1": 0.0004166666666666667,
97
+ "pass@2": 0.0008333333333333334,
98
+ "pass@4": 0.0016666666666666668,
99
+ "pass@8": 0.0033333333333333335,
100
+ "pass@16": 0.006666666666666667,
101
+ "pass@32": 0.013333333333333334
102
+ },
103
+ "30": {
104
+ "pass@1": 0.0004464285714285714,
105
+ "pass@2": 0.0008928571428571428,
106
+ "pass@4": 0.0017857142857142857,
107
+ "pass@8": 0.0035714285714285713,
108
+ "pass@16": 0.007142857142857143,
109
+ "pass@32": 0.014285714285714285
110
+ }
111
+ }
112
+ },
113
+ "teachers_in_school": {
114
+ "count": 19936,
115
+ "correct": 29,
116
+ "answer_accuracy": 0.0014546548956661316,
117
+ "avg_response_len": 290.34595706260035,
118
+ "resp_tokens_sum": 5788337,
119
+ "pass_at_k": {
120
+ "pass@1": 0.0014546548956661316,
121
+ "pass@2": 0.0028607673587738823,
122
+ "pass@4": 0.005535087139803991,
123
+ "pass@8": 0.0103836615322536,
124
+ "pass@16": 0.018455498083299257,
125
+ "pass@32": 0.030497592295345103
126
+ },
127
+ "per_op_pass_at_k": {
128
+ "21": {
129
+ "pass@1": 0.004326923076923077,
130
+ "pass@2": 0.008374689826302729,
131
+ "pass@4": 0.015694361256096516,
132
+ "pass@8": 0.027637545991272352,
133
+ "pass@16": 0.0436382305125353,
134
+ "pass@32": 0.06153846153846154
135
+ },
136
+ "22": {
137
+ "pass@1": 0.0030737704918032786,
138
+ "pass@2": 0.006114489687995769,
139
+ "pass@4": 0.012096774193548387,
140
+ "pass@8": 0.023664727657324167,
141
+ "pass@16": 0.0452141723955579,
142
+ "pass@32": 0.08196721311475409
143
+ },
144
+ "23": {
145
+ "pass@1": 0.0013392857142857143,
146
+ "pass@2": 0.0026497695852534552,
147
+ "pass@4": 0.005184331797235023,
148
+ "pass@8": 0.00990783410138249,
149
+ "pass@16": 0.017972350230414748,
150
+ "pass@32": 0.02857142857142857
151
+ },
152
+ "24": {
153
+ "pass@1": 0.0027901785714285715,
154
+ "pass@2": 0.005472350230414745,
155
+ "pass@4": 0.010527073732718894,
156
+ "pass@8": 0.019498847926267278,
157
+ "pass@16": 0.03369815668202765,
158
+ "pass@32": 0.05357142857142857
159
+ },
160
+ "25": {
161
+ "pass@1": 0.0004280821917808219,
162
+ "pass@2": 0.0008561643835616438,
163
+ "pass@4": 0.0017123287671232876,
164
+ "pass@8": 0.003424657534246575,
165
+ "pass@16": 0.00684931506849315,
166
+ "pass@32": 0.0136986301369863
167
+ },
168
+ "26": {
169
+ "pass@1": 0.0006127450980392157,
170
+ "pass@2": 0.0012254901960784314,
171
+ "pass@4": 0.0024509803921568627,
172
+ "pass@8": 0.004901960784313725,
173
+ "pass@16": 0.00980392156862745,
174
+ "pass@32": 0.0196078431372549
175
+ },
176
+ "27": {
177
+ "pass@1": 0.0005387931034482759,
178
+ "pass@2": 0.0010775862068965517,
179
+ "pass@4": 0.0021551724137931034,
180
+ "pass@8": 0.004310344827586207,
181
+ "pass@16": 0.008620689655172414,
182
+ "pass@32": 0.017241379310344827
183
+ },
184
+ "28": {
185
+ "pass@1": 0.0,
186
+ "pass@2": 0.0,
187
+ "pass@4": 0.0,
188
+ "pass@8": 0.0,
189
+ "pass@16": 0.0,
190
+ "pass@32": 0.0
191
+ },
192
+ "29": {
193
+ "pass@1": 0.0014423076923076924,
194
+ "pass@2": 0.002853598014888337,
195
+ "pass@4": 0.005583126550868486,
196
+ "pass@8": 0.010669975186104219,
197
+ "pass@16": 0.01935483870967742,
198
+ "pass@32": 0.03076923076923077
199
+ },
200
+ "30": {
201
+ "pass@1": 0.0,
202
+ "pass@2": 0.0,
203
+ "pass@4": 0.0,
204
+ "pass@8": 0.0,
205
+ "pass@16": 0.0,
206
+ "pass@32": 0.0
207
+ }
208
+ }
209
+ },
210
+ "movie_festival_awards": {
211
+ "count": 20992,
212
+ "correct": 69,
213
+ "answer_accuracy": 0.003286966463414634,
214
+ "avg_response_len": 328.9145388719512,
215
+ "resp_tokens_sum": 6904574,
216
+ "pass_at_k": {
217
+ "pass@1": 0.003286966463414634,
218
+ "pass@2": 0.0060207267899291895,
219
+ "pass@4": 0.010398961244200873,
220
+ "pass@8": 0.016888463323967543,
221
+ "pass@16": 0.026588695555185955,
222
+ "pass@32": 0.042682926829268296
223
+ },
224
+ "per_op_pass_at_k": {
225
+ "21": {
226
+ "pass@1": 0.005580357142857143,
227
+ "pass@2": 0.010440668202764977,
228
+ "pass@4": 0.01828420467185762,
229
+ "pass@8": 0.028176148101064676,
230
+ "pass@16": 0.03493961544573335,
231
+ "pass@32": 0.03571428571428571
232
+ },
233
+ "22": {
234
+ "pass@1": 0.004197761194029851,
235
+ "pass@2": 0.008094607607125662,
236
+ "pass@4": 0.015103846728537512,
237
+ "pass@8": 0.026700479803430013,
238
+ "pass@16": 0.04445237660418707,
239
+ "pass@32": 0.07462686567164178
240
+ },
241
+ "23": {
242
+ "pass@1": 0.00830078125,
243
+ "pass@2": 0.014112903225806451,
244
+ "pass@4": 0.021564324944382644,
245
+ "pass@8": 0.030255666017797552,
246
+ "pass@16": 0.04309473287549774,
247
+ "pass@32": 0.0625
248
+ },
249
+ "24": {
250
+ "pass@1": 0.0021404109589041095,
251
+ "pass@2": 0.0042532037118868755,
252
+ "pass@4": 0.008395934600088379,
253
+ "pass@8": 0.016349977905435263,
254
+ "pass@16": 0.030932390631904547,
255
+ "pass@32": 0.0547945205479452
256
+ },
257
+ "25": {
258
+ "pass@1": 0.001953125,
259
+ "pass@2": 0.00390625,
260
+ "pass@4": 0.0078125,
261
+ "pass@8": 0.015625,
262
+ "pass@16": 0.03125,
263
+ "pass@32": 0.0625
264
+ },
265
+ "26": {
266
+ "pass@1": 0.004518072289156626,
267
+ "pass@2": 0.007700155460551886,
268
+ "pass@4": 0.01154897677472962,
269
+ "pass@8": 0.014827153226579631,
270
+ "pass@16": 0.018071881276617813,
271
+ "pass@32": 0.024096385542168676
272
+ },
273
+ "27": {
274
+ "pass@1": 0.0035714285714285713,
275
+ "pass@2": 0.0067108294930875575,
276
+ "pass@4": 0.011918004131574764,
277
+ "pass@8": 0.01930672528559069,
278
+ "pass@16": 0.028445186009145966,
279
+ "pass@32": 0.04285714285714286
280
+ },
281
+ "28": {
282
+ "pass@1": 0.0,
283
+ "pass@2": 0.0,
284
+ "pass@4": 0.0,
285
+ "pass@8": 0.0,
286
+ "pass@16": 0.0,
287
+ "pass@32": 0.0
288
+ },
289
+ "29": {
290
+ "pass@1": 0.0010416666666666667,
291
+ "pass@2": 0.0020833333333333333,
292
+ "pass@4": 0.004166666666666667,
293
+ "pass@8": 0.008333333333333333,
294
+ "pass@16": 0.016666666666666666,
295
+ "pass@32": 0.03333333333333333
296
+ },
297
+ "30": {
298
+ "pass@1": 0.000946969696969697,
299
+ "pass@2": 0.001893939393939394,
300
+ "pass@4": 0.003787878787878788,
301
+ "pass@8": 0.007575757575757576,
302
+ "pass@16": 0.015151515151515152,
303
+ "pass@32": 0.030303030303030304
304
+ }
305
+ }
306
+ }
307
+ }
308
+ }
309
+ ]