ZachW commited on
Commit
a265173
·
verified ·
1 Parent(s): 798d552

Upload extrapolation artifact: rl_ckpt18809_op11_14_step200_merged_hf far-OOD summary JSON

Browse files
020305/evals/far_ood/rl_ckpt18809_op11_14_step200_merged_hf/summary.json ADDED
@@ -0,0 +1,309 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "checkpoint": "rl_ckpt18809_op11_14_step200_merged_hf",
4
+ "id_acc": 0.0,
5
+ "ood_acc": 0.0025,
6
+ "total_acc": 0.0025,
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": 298.277109375,
12
+ "count_id": 0,
13
+ "count_ood": 64000,
14
+ "count_total": 64000,
15
+ "template_metrics": {
16
+ "crazy_zootopia": {
17
+ "count": 23072,
18
+ "correct": 60,
19
+ "answer_accuracy": 0.0026005547850208046,
20
+ "avg_response_len": 276.871012482663,
21
+ "resp_tokens_sum": 6387968,
22
+ "pass_at_k": {
23
+ "pass@1": 0.0026005547850208046,
24
+ "pass@2": 0.004667017135698627,
25
+ "pass@4": 0.007791173425859214,
26
+ "pass@8": 0.012087750689001269,
27
+ "pass@16": 0.018247633171405726,
28
+ "pass@32": 0.027739251040221916
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": 0.0026041666666666665,
41
+ "pass@2": 0.005124327956989246,
42
+ "pass@4": 0.009923835125448028,
43
+ "pass@8": 0.018637992831541217,
44
+ "pass@16": 0.033154121863799284,
45
+ "pass@32": 0.05555555555555555
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": 0.008802816901408451,
57
+ "pass@2": 0.014964788732394365,
58
+ "pass@4": 0.023108618339626186,
59
+ "pass@8": 0.033562487270676344,
60
+ "pass@16": 0.049522940531216975,
61
+ "pass@32": 0.07042253521126761
62
+ },
63
+ "25": {
64
+ "pass@1": 0.009424603174603174,
65
+ "pass@2": 0.016737071172555042,
66
+ "pass@4": 0.026901583770326816,
67
+ "pass@8": 0.037567790014953535,
68
+ "pass@16": 0.04756312746511196,
69
+ "pass@32": 0.06349206349206349
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": 0.0035919540229885057,
89
+ "pass@2": 0.006511865035224323,
90
+ "pass@4": 0.010832278521473412,
91
+ "pass@8": 0.015788797652170748,
92
+ "pass@16": 0.020193573958293378,
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": 0.0,
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": 32,
116
+ "answer_accuracy": 0.0016051364365971107,
117
+ "avg_response_len": 290.62841091492777,
118
+ "resp_tokens_sum": 5793968,
119
+ "pass_at_k": {
120
+ "pass@1": 0.0016051364365971107,
121
+ "pass@2": 0.0031520219541241647,
122
+ "pass@4": 0.006083761697052369,
123
+ "pass@8": 0.01137611435570466,
124
+ "pass@16": 0.020132946005638506,
125
+ "pass@32": 0.03210272873194221
126
+ },
127
+ "per_op_pass_at_k": {
128
+ "21": {
129
+ "pass@1": 0.0033653846153846156,
130
+ "pass@2": 0.006637717121588087,
131
+ "pass@4": 0.012903225806451613,
132
+ "pass@8": 0.024317617866004965,
133
+ "pass@16": 0.042679900744416875,
134
+ "pass@32": 0.06153846153846154
135
+ },
136
+ "22": {
137
+ "pass@1": 0.005122950819672131,
138
+ "pass@2": 0.010146747752511896,
139
+ "pass@4": 0.019896879957694343,
140
+ "pass@8": 0.03820729772607087,
141
+ "pass@16": 0.07006874669487044,
142
+ "pass@32": 0.11475409836065574
143
+ },
144
+ "23": {
145
+ "pass@1": 0.0008928571428571428,
146
+ "pass@2": 0.0017857142857142857,
147
+ "pass@4": 0.0035714285714285713,
148
+ "pass@8": 0.007142857142857143,
149
+ "pass@16": 0.014285714285714285,
150
+ "pass@32": 0.02857142857142857
151
+ },
152
+ "24": {
153
+ "pass@1": 0.0033482142857142855,
154
+ "pass@2": 0.006624423963133639,
155
+ "pass@4": 0.012960829493087559,
156
+ "pass@8": 0.02476958525345622,
157
+ "pass@16": 0.044930875576036866,
158
+ "pass@32": 0.07142857142857142
159
+ },
160
+ "25": {
161
+ "pass@1": 0.0,
162
+ "pass@2": 0.0,
163
+ "pass@4": 0.0,
164
+ "pass@8": 0.0,
165
+ "pass@16": 0.0,
166
+ "pass@32": 0.0
167
+ },
168
+ "26": {
169
+ "pass@1": 0.0,
170
+ "pass@2": 0.0,
171
+ "pass@4": 0.0,
172
+ "pass@8": 0.0,
173
+ "pass@16": 0.0,
174
+ "pass@32": 0.0
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": 0.0,
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": 0.0033653846153846156,
194
+ "pass@2": 0.006420595533498759,
195
+ "pass@4": 0.011722426627877128,
196
+ "pass@8": 0.019829725335843245,
197
+ "pass@16": 0.030435526653546673,
198
+ "pass@32": 0.046153846153846156
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": 68,
213
+ "answer_accuracy": 0.0032393292682926828,
214
+ "avg_response_len": 329.0681688262195,
215
+ "resp_tokens_sum": 6907799,
216
+ "pass_at_k": {
217
+ "pass@1": 0.0032393292682926828,
218
+ "pass@2": 0.006011506687647521,
219
+ "pass@4": 0.010611002401041809,
220
+ "pass@8": 0.017712778837487186,
221
+ "pass@16": 0.028303276404021957,
222
+ "pass@32": 0.04420731707317073
223
+ },
224
+ "per_op_pass_at_k": {
225
+ "21": {
226
+ "pass@1": 0.006696428571428571,
227
+ "pass@2": 0.012708813364055297,
228
+ "pass@4": 0.0229764222151597,
229
+ "pass@8": 0.0381599043028409,
230
+ "pass@16": 0.056005791267192816,
231
+ "pass@32": 0.07142857142857142
232
+ },
233
+ "22": {
234
+ "pass@1": 0.0032649253731343282,
235
+ "pass@2": 0.006349301877708233,
236
+ "pass@4": 0.012024139591253964,
237
+ "pass@8": 0.021709029933757243,
238
+ "pass@16": 0.036558032971958886,
239
+ "pass@32": 0.05970149253731343
240
+ },
241
+ "23": {
242
+ "pass@1": 0.0048828125,
243
+ "pass@2": 0.008631552419354838,
244
+ "pass@4": 0.013730533926585095,
245
+ "pass@8": 0.018802885506688342,
246
+ "pass@16": 0.023431127178346313,
247
+ "pass@32": 0.03125
248
+ },
249
+ "24": {
250
+ "pass@1": 0.0025684931506849314,
251
+ "pass@2": 0.004971277065841803,
252
+ "pass@4": 0.009323525378274185,
253
+ "pass@8": 0.01650006856933884,
254
+ "pass@16": 0.02670394807015405,
255
+ "pass@32": 0.0410958904109589
256
+ },
257
+ "25": {
258
+ "pass@1": 0.0009765625,
259
+ "pass@2": 0.001953125,
260
+ "pass@4": 0.00390625,
261
+ "pass@8": 0.0078125,
262
+ "pass@16": 0.015625,
263
+ "pass@32": 0.03125
264
+ },
265
+ "26": {
266
+ "pass@1": 0.006024096385542169,
267
+ "pass@2": 0.010153517294986397,
268
+ "pass@4": 0.015267633381132984,
269
+ "pass@8": 0.020997761904123724,
270
+ "pass@16": 0.03012046250485329,
271
+ "pass@32": 0.04819277108433735
272
+ },
273
+ "27": {
274
+ "pass@1": 0.0040178571428571425,
275
+ "pass@2": 0.007862903225806451,
276
+ "pass@4": 0.015057603686635944,
277
+ "pass@8": 0.027626728110599075,
278
+ "pass@16": 0.0467741935483871,
279
+ "pass@32": 0.07142857142857142
280
+ },
281
+ "28": {
282
+ "pass@1": 0.0005896226415094339,
283
+ "pass@2": 0.0011792452830188679,
284
+ "pass@4": 0.0023584905660377358,
285
+ "pass@8": 0.0047169811320754715,
286
+ "pass@16": 0.009433962264150943,
287
+ "pass@32": 0.018867924528301886
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": 0.0014204545454545455,
299
+ "pass@2": 0.0028103616813294225,
300
+ "pass@4": 0.005498533724340176,
301
+ "pass@8": 0.01050830889540567,
302
+ "pass@16": 0.01906158357771261,
303
+ "pass@32": 0.030303030303030304
304
+ }
305
+ }
306
+ }
307
+ }
308
+ }
309
+ ]