ZachW commited on
Commit
480ff4d
·
verified ·
1 Parent(s): 944de51

Upload extrapolation artifact: pretrain_ckpt18809 far-OOD summary JSON

Browse files
020305/evals/far_ood/pretrain_ckpt18809/summary.json ADDED
@@ -0,0 +1,309 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "checkpoint": "pretrain_ckpt18809",
4
+ "id_acc": 0.0,
5
+ "ood_acc": 0.000875,
6
+ "total_acc": 0.000875,
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": 229.355625,
12
+ "count_id": 0,
13
+ "count_ood": 64000,
14
+ "count_total": 64000,
15
+ "template_metrics": {
16
+ "crazy_zootopia": {
17
+ "count": 23072,
18
+ "correct": 13,
19
+ "answer_accuracy": 0.0005634535367545076,
20
+ "avg_response_len": 212.57216539528432,
21
+ "resp_tokens_sum": 4904465,
22
+ "pass_at_k": {
23
+ "pass@1": 0.0005634535367545076,
24
+ "pass@2": 0.0011045367097669006,
25
+ "pass@4": 0.0021218290009395553,
26
+ "pass@8": 0.003912576618495816,
27
+ "pass@16": 0.006643998031407991,
28
+ "pass@32": 0.009708737864077669
29
+ },
30
+ "per_op_pass_at_k": {
31
+ "21": {
32
+ "pass@1": 0.0011867088607594937,
33
+ "pass@2": 0.0023478971008574923,
34
+ "pass@4": 0.004593711719069008,
35
+ "pass@8": 0.008779093507554105,
36
+ "pass@16": 0.01592486729277256,
37
+ "pass@32": 0.02531645569620253
38
+ },
39
+ "22": {
40
+ "pass@1": 0.0008680555555555555,
41
+ "pass@2": 0.001736111111111111,
42
+ "pass@4": 0.003472222222222222,
43
+ "pass@8": 0.006944444444444444,
44
+ "pass@16": 0.013888888888888888,
45
+ "pass@32": 0.027777777777777776
46
+ },
47
+ "23": {
48
+ "pass@1": 0.0,
49
+ "pass@2": 0.0,
50
+ "pass@4": 0.0,
51
+ "pass@8": 0.0,
52
+ "pass@16": 0.0,
53
+ "pass@32": 0.0
54
+ },
55
+ "24": {
56
+ "pass@1": 0.0,
57
+ "pass@2": 0.0,
58
+ "pass@4": 0.0,
59
+ "pass@8": 0.0,
60
+ "pass@16": 0.0,
61
+ "pass@32": 0.0
62
+ },
63
+ "25": {
64
+ "pass@1": 0.00248015873015873,
65
+ "pass@2": 0.004832309267793137,
66
+ "pass@4": 0.009165386584741423,
67
+ "pass@8": 0.01643625192012289,
68
+ "pass@16": 0.026113671274961597,
69
+ "pass@32": 0.031746031746031744
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": 0.0,
81
+ "pass@2": 0.0,
82
+ "pass@4": 0.0,
83
+ "pass@8": 0.0,
84
+ "pass@16": 0.0,
85
+ "pass@32": 0.0
86
+ },
87
+ "28": {
88
+ "pass@1": 0.0010775862068965517,
89
+ "pass@2": 0.0020856507230255835,
90
+ "pass@4": 0.0039024842417500925,
91
+ "pass@8": 0.006803856136447905,
92
+ "pass@16": 0.010196514645902854,
93
+ "pass@32": 0.011494252873563218
94
+ },
95
+ "29": {
96
+ "pass@1": 0.0,
97
+ "pass@2": 0.0,
98
+ "pass@4": 0.0,
99
+ "pass@8": 0.0,
100
+ "pass@16": 0.0,
101
+ "pass@32": 0.0
102
+ },
103
+ "30": {
104
+ "pass@1": 0.0,
105
+ "pass@2": 0.0,
106
+ "pass@4": 0.0,
107
+ "pass@8": 0.0,
108
+ "pass@16": 0.0,
109
+ "pass@32": 0.0
110
+ }
111
+ }
112
+ },
113
+ "teachers_in_school": {
114
+ "count": 19936,
115
+ "correct": 30,
116
+ "answer_accuracy": 0.0015048154093097914,
117
+ "avg_response_len": 223.00491573033707,
118
+ "resp_tokens_sum": 4445826,
119
+ "pass_at_k": {
120
+ "pass@1": 0.0015048154093097914,
121
+ "pass@2": 0.0028801843317972347,
122
+ "pass@4": 0.005318375866175543,
123
+ "pass@8": 0.009321288377793106,
124
+ "pass@16": 0.015424486275994192,
125
+ "pass@32": 0.024077046548956663
126
+ },
127
+ "per_op_pass_at_k": {
128
+ "21": {
129
+ "pass@1": 0.0014423076923076924,
130
+ "pass@2": 0.002853598014888337,
131
+ "pass@4": 0.005583126550868486,
132
+ "pass@8": 0.010669975186104219,
133
+ "pass@16": 0.01935483870967742,
134
+ "pass@32": 0.03076923076923077
135
+ },
136
+ "22": {
137
+ "pass@1": 0.0,
138
+ "pass@2": 0.0,
139
+ "pass@4": 0.0,
140
+ "pass@8": 0.0,
141
+ "pass@16": 0.0,
142
+ "pass@32": 0.0
143
+ },
144
+ "23": {
145
+ "pass@1": 0.0004464285714285714,
146
+ "pass@2": 0.0008928571428571428,
147
+ "pass@4": 0.0017857142857142857,
148
+ "pass@8": 0.0035714285714285713,
149
+ "pass@16": 0.007142857142857143,
150
+ "pass@32": 0.014285714285714285
151
+ },
152
+ "24": {
153
+ "pass@1": 0.0011160714285714285,
154
+ "pass@2": 0.002232142857142857,
155
+ "pass@4": 0.004464285714285714,
156
+ "pass@8": 0.008928571428571428,
157
+ "pass@16": 0.017857142857142856,
158
+ "pass@32": 0.03571428571428571
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": 0.0024509803921568627,
170
+ "pass@2": 0.004862428842504742,
171
+ "pass@4": 0.009566729917773562,
172
+ "pass@8": 0.018500948766603416,
173
+ "pass@16": 0.034471853257432,
174
+ "pass@32": 0.058823529411764705
175
+ },
176
+ "27": {
177
+ "pass@1": 0.0,
178
+ "pass@2": 0.0,
179
+ "pass@4": 0.0,
180
+ "pass@8": 0.0,
181
+ "pass@16": 0.0,
182
+ "pass@32": 0.0
183
+ },
184
+ "28": {
185
+ "pass@1": 0.007291666666666667,
186
+ "pass@2": 0.013407258064516129,
187
+ "pass@4": 0.022883759733036704,
188
+ "pass@8": 0.03463550668834318,
189
+ "pass@16": 0.04510381905821283,
190
+ "pass@32": 0.05
191
+ },
192
+ "29": {
193
+ "pass@1": 0.0,
194
+ "pass@2": 0.0,
195
+ "pass@4": 0.0,
196
+ "pass@8": 0.0,
197
+ "pass@16": 0.0,
198
+ "pass@32": 0.0
199
+ },
200
+ "30": {
201
+ "pass@1": 0.00244140625,
202
+ "pass@2": 0.0047883064516129024,
203
+ "pass@4": 0.009211189516129032,
204
+ "pass@8": 0.01706149193548387,
205
+ "pass@16": 0.029485887096774195,
206
+ "pass@32": 0.046875
207
+ }
208
+ }
209
+ },
210
+ "movie_festival_awards": {
211
+ "count": 20992,
212
+ "correct": 13,
213
+ "answer_accuracy": 0.0006192835365853659,
214
+ "avg_response_len": 253.8333174542683,
215
+ "resp_tokens_sum": 5328469,
216
+ "pass_at_k": {
217
+ "pass@1": 0.0006192835365853659,
218
+ "pass@2": 0.0011709529897718331,
219
+ "pass@4": 0.002112993095309151,
220
+ "pass@8": 0.003568171430837266,
221
+ "pass@16": 0.005723575728527029,
222
+ "pass@32": 0.009146341463414634
223
+ },
224
+ "per_op_pass_at_k": {
225
+ "21": {
226
+ "pass@1": 0.0,
227
+ "pass@2": 0.0,
228
+ "pass@4": 0.0,
229
+ "pass@8": 0.0,
230
+ "pass@16": 0.0,
231
+ "pass@32": 0.0
232
+ },
233
+ "22": {
234
+ "pass@1": 0.0004664179104477612,
235
+ "pass@2": 0.0009328358208955224,
236
+ "pass@4": 0.0018656716417910447,
237
+ "pass@8": 0.0037313432835820895,
238
+ "pass@16": 0.007462686567164179,
239
+ "pass@32": 0.014925373134328358
240
+ },
241
+ "23": {
242
+ "pass@1": 0.0009765625,
243
+ "pass@2": 0.001953125,
244
+ "pass@4": 0.00390625,
245
+ "pass@8": 0.0078125,
246
+ "pass@16": 0.015625,
247
+ "pass@32": 0.03125
248
+ },
249
+ "24": {
250
+ "pass@1": 0.0004280821917808219,
251
+ "pass@2": 0.0008561643835616438,
252
+ "pass@4": 0.0017123287671232876,
253
+ "pass@8": 0.003424657534246575,
254
+ "pass@16": 0.00684931506849315,
255
+ "pass@32": 0.0136986301369863
256
+ },
257
+ "25": {
258
+ "pass@1": 0.0,
259
+ "pass@2": 0.0,
260
+ "pass@4": 0.0,
261
+ "pass@8": 0.0,
262
+ "pass@16": 0.0,
263
+ "pass@32": 0.0
264
+ },
265
+ "26": {
266
+ "pass@1": 0.0026355421686746986,
267
+ "pass@2": 0.0047609794014768754,
268
+ "pass@4": 0.007809882466462065,
269
+ "pass@8": 0.01080930206665629,
270
+ "pass@16": 0.012007242913068664,
271
+ "pass@32": 0.012048192771084338
272
+ },
273
+ "27": {
274
+ "pass@1": 0.0,
275
+ "pass@2": 0.0,
276
+ "pass@4": 0.0,
277
+ "pass@8": 0.0,
278
+ "pass@16": 0.0,
279
+ "pass@32": 0.0
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": 0.0020497311827956982,
292
+ "pass@4": 0.0039650537634408605,
293
+ "pass@8": 0.007392473118279571,
294
+ "pass@16": 0.012634408602150538,
295
+ "pass@32": 0.016666666666666666
296
+ },
297
+ "30": {
298
+ "pass@1": 0.0,
299
+ "pass@2": 0.0,
300
+ "pass@4": 0.0,
301
+ "pass@8": 0.0,
302
+ "pass@16": 0.0,
303
+ "pass@32": 0.0
304
+ }
305
+ }
306
+ }
307
+ }
308
+ }
309
+ ]