Datasets:

Languages:
English
ArXiv:
License:
Daniel O'Connell commited on
Commit
4364273
·
1 Parent(s): 0fbf0bc

update datafiles

Browse files
README.md CHANGED
@@ -22,16 +22,24 @@ dataset_info:
22
  dtype: string
23
  - name: date_published
24
  dtype: string
 
 
25
  - name: source_filetype
26
  dtype: string
27
  - name: converted_with
28
  dtype: string
29
  - name: book_title
30
  dtype: string
31
- - name: authors
32
- dtype: string
33
  - name: paged_url
34
  dtype: string
 
 
 
 
 
 
 
 
35
  - name: karma
36
  dtype: int32
37
  - name: votes
@@ -92,18 +100,12 @@ dataset_info:
92
  dtype: string
93
  - name: filename
94
  dtype: string
95
- - name: question
96
- dtype: string
97
- - name: answer
98
- dtype: string
99
- - name: entry
100
- dtype: string
101
  splits:
102
  - name: train
103
- num_bytes: 313193281
104
- num_examples: 9681
105
- download_size: 359283302
106
- dataset_size: 313193281
107
  - config_name: agentmodels
108
  features:
109
  - name: id
@@ -118,21 +120,21 @@ dataset_info:
118
  dtype: string
119
  - name: date_published
120
  dtype: string
 
 
121
  - name: source_filetype
122
  dtype: string
123
  - name: converted_with
124
  dtype: string
125
  - name: book_title
126
  dtype: string
127
- - name: authors
128
- sequence: string
129
  splits:
130
  - name: train
131
- num_bytes: 385017
132
  num_examples: 21
133
- download_size: 399788
134
- dataset_size: 385017
135
- - config_name: aiimpacts.org
136
  features:
137
  - name: id
138
  dtype: string
@@ -146,14 +148,18 @@ dataset_info:
146
  dtype: string
147
  - name: date_published
148
  dtype: string
 
 
149
  - name: paged_url
150
  dtype: string
 
 
151
  splits:
152
  - name: train
153
- num_bytes: 2199965
154
  num_examples: 253
155
- download_size: 2231312
156
- dataset_size: 2199965
157
  - config_name: aipulse.org
158
  features:
159
  - name: id
@@ -168,14 +174,16 @@ dataset_info:
168
  dtype: string
169
  - name: date_published
170
  dtype: string
 
 
171
  - name: paged_url
172
  dtype: string
173
  splits:
174
  - name: train
175
- num_bytes: 652237
176
  num_examples: 23
177
  download_size: 655946
178
- dataset_size: 652237
179
  - config_name: aisafety.camp
180
  features:
181
  - name: id
@@ -190,14 +198,48 @@ dataset_info:
190
  dtype: string
191
  - name: date_published
192
  dtype: string
 
 
193
  - name: paged_url
194
  dtype: string
 
 
195
  splits:
196
  - name: train
197
- num_bytes: 83463
198
  num_examples: 9
199
- download_size: 84634
200
- dataset_size: 83463
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
201
  - config_name: alignmentforum
202
  features:
203
  - name: id
@@ -213,11 +255,7 @@ dataset_info:
213
  - name: date_published
214
  dtype: string
215
  - name: authors
216
- sequence:
217
- - name: username
218
- dtype: string
219
- - name: displayName
220
- dtype: string
221
  - name: karma
222
  dtype: int32
223
  - name: votes
@@ -234,10 +272,10 @@ dataset_info:
234
  dtype: string
235
  splits:
236
  - name: train
237
- num_bytes: 73475866
238
- num_examples: 2642
239
- download_size: 74837221
240
- dataset_size: 73475866
241
  - config_name: alignment_newsletter
242
  features:
243
  - name: id
@@ -252,6 +290,8 @@ dataset_info:
252
  dtype: string
253
  - name: date_published
254
  dtype: string
 
 
255
  - name: converted_with
256
  dtype: string
257
  - name: source_type
@@ -272,14 +312,12 @@ dataset_info:
272
  dtype: string
273
  - name: read_more
274
  dtype: string
275
- - name: authors
276
- dtype: string
277
  splits:
278
  - name: train
279
- num_bytes: 3171348
280
- num_examples: 1507
281
- download_size: 3771103
282
- dataset_size: 3171348
283
  - config_name: arbital
284
  features:
285
  - name: id
@@ -294,19 +332,21 @@ dataset_info:
294
  dtype: string
295
  - name: date_published
296
  dtype: string
297
- - name: source_filetype
298
- dtype: string
299
  - name: authors
 
 
300
  dtype: string
301
  - name: alias
302
  dtype: string
 
 
303
  splits:
304
  - name: train
305
- num_bytes: 3296285
306
- num_examples: 731
307
- download_size: 4183968
308
- dataset_size: 3296285
309
- - config_name: arxiv_papers
310
  features:
311
  - name: id
312
  dtype: string
@@ -321,7 +361,7 @@ dataset_info:
321
  - name: date_published
322
  dtype: string
323
  - name: authors
324
- dtype: string
325
  - name: source_type
326
  dtype: string
327
  - name: converted_with
@@ -340,12 +380,14 @@ dataset_info:
340
  dtype: string
341
  - name: categories
342
  sequence: string
 
 
343
  splits:
344
  - name: train
345
- num_bytes: 139825760
346
  num_examples: 1144
347
- download_size: 155468988
348
- dataset_size: 139825760
349
  - config_name: audio_transcripts
350
  features:
351
  - name: id
@@ -360,18 +402,18 @@ dataset_info:
360
  dtype: string
361
  - name: date_published
362
  dtype: string
 
 
363
  - name: source_filetype
364
  dtype: string
365
  - name: converted_with
366
  dtype: string
367
- - name: authors
368
- dtype: string
369
  splits:
370
  - name: train
371
- num_bytes: 1312506
372
  num_examples: 37
373
- download_size: 1334641
374
- dataset_size: 1312506
375
  - config_name: carado.moe
376
  features:
377
  - name: id
@@ -386,17 +428,17 @@ dataset_info:
386
  dtype: string
387
  - name: date_published
388
  dtype: string
389
- - name: source_type
390
- dtype: string
391
  - name: authors
 
 
392
  dtype: string
393
  splits:
394
  - name: train
395
- num_bytes: 822934
396
- num_examples: 247
397
- download_size: 845692
398
- dataset_size: 822934
399
- - config_name: cold.takes
400
  features:
401
  - name: id
402
  dtype: string
@@ -410,15 +452,17 @@ dataset_info:
410
  dtype: string
411
  - name: date_published
412
  dtype: string
 
 
413
  - name: source_type
414
  dtype: string
415
  splits:
416
  - name: train
417
- num_bytes: 1630528
418
  num_examples: 102
419
- download_size: 2553428
420
- dataset_size: 1630528
421
- - config_name: deepmind.blog
422
  features:
423
  - name: id
424
  dtype: string
@@ -432,14 +476,16 @@ dataset_info:
432
  dtype: string
433
  - name: date_published
434
  dtype: string
 
 
435
  - name: source_type
436
  dtype: string
437
  splits:
438
  - name: train
439
- num_bytes: 111079
440
  num_examples: 10
441
- download_size: 111779
442
- dataset_size: 111079
443
  - config_name: distill
444
  features:
445
  - name: id
@@ -454,12 +500,12 @@ dataset_info:
454
  dtype: string
455
  - name: date_published
456
  dtype: string
 
 
457
  - name: source_type
458
  dtype: string
459
  - name: converted_with
460
  dtype: string
461
- - name: authors
462
- sequence: string
463
  - name: abstract
464
  dtype: string
465
  - name: journal_ref
@@ -472,10 +518,10 @@ dataset_info:
472
  dtype: string
473
  splits:
474
  - name: train
475
- num_bytes: 1503276
476
- num_examples: 54
477
- download_size: 1597604
478
- dataset_size: 1503276
479
  - config_name: eaforum
480
  features:
481
  - name: id
@@ -491,11 +537,7 @@ dataset_info:
491
  - name: date_published
492
  dtype: string
493
  - name: authors
494
- sequence:
495
- - name: username
496
- dtype: string
497
- - name: displayName
498
- dtype: string
499
  - name: karma
500
  dtype: int32
501
  - name: votes
@@ -510,12 +552,14 @@ dataset_info:
510
  dtype: string
511
  - name: source_type
512
  dtype: string
 
 
513
  splits:
514
  - name: train
515
- num_bytes: 8459932
516
- num_examples: 403
517
- download_size: 8598593
518
- dataset_size: 8459932
519
  - config_name: gdocs
520
  features:
521
  - name: id
@@ -530,12 +574,12 @@ dataset_info:
530
  dtype: string
531
  - name: date_published
532
  dtype: string
 
 
533
  - name: source_filetype
534
  dtype: string
535
  - name: converted_with
536
  dtype: string
537
- - name: authors
538
- dtype: string
539
  - name: docx_name
540
  dtype: string
541
  splits:
@@ -558,6 +602,8 @@ dataset_info:
558
  dtype: string
559
  - name: date_published
560
  dtype: string
 
 
561
  - name: source_filetype
562
  dtype: string
563
  - name: converted_with
@@ -568,10 +614,10 @@ dataset_info:
568
  dtype: string
569
  splits:
570
  - name: train
571
- num_bytes: 19582819
572
  num_examples: 20
573
- download_size: 19745156
574
- dataset_size: 19582819
575
  - config_name: generative.ink
576
  features:
577
  - name: id
@@ -586,14 +632,16 @@ dataset_info:
586
  dtype: string
587
  - name: date_published
588
  dtype: string
 
 
589
  - name: source_type
590
  dtype: string
591
  splits:
592
  - name: train
593
- num_bytes: 446616
594
  num_examples: 19
595
- download_size: 449238
596
- dataset_size: 446616
597
  - config_name: gwern_blog
598
  features:
599
  - name: id
@@ -609,16 +657,16 @@ dataset_info:
609
  - name: date_published
610
  dtype: string
611
  - name: authors
612
- dtype: string
613
  - name: description
614
  dtype: string
615
  splits:
616
  - name: train
617
- num_bytes: 685779
618
  num_examples: 7
619
- download_size: 692140
620
- dataset_size: 685779
621
- - config_name: intelligence.org
622
  features:
623
  - name: id
624
  dtype: string
@@ -632,15 +680,17 @@ dataset_info:
632
  dtype: string
633
  - name: date_published
634
  dtype: string
635
- - name: paged_url
 
 
636
  dtype: string
637
  splits:
638
  - name: train
639
- num_bytes: 5657100
640
- num_examples: 492
641
- download_size: 5733645
642
- dataset_size: 5657100
643
- - config_name: jsteinhardt.wordpress.com
644
  features:
645
  - name: id
646
  dtype: string
@@ -654,14 +704,18 @@ dataset_info:
654
  dtype: string
655
  - name: date_published
656
  dtype: string
 
 
657
  - name: paged_url
658
  dtype: string
 
 
659
  splits:
660
  - name: train
661
- num_bytes: 373205
662
  num_examples: 39
663
- download_size: 377419
664
- dataset_size: 373205
665
  - config_name: lesswrong
666
  features:
667
  - name: id
@@ -677,11 +731,7 @@ dataset_info:
677
  - name: date_published
678
  dtype: string
679
  - name: authors
680
- sequence:
681
- - name: username
682
- dtype: string
683
- - name: displayName
684
- dtype: string
685
  - name: karma
686
  dtype: int32
687
  - name: votes
@@ -696,12 +746,14 @@ dataset_info:
696
  dtype: string
697
  - name: source_type
698
  dtype: string
 
 
699
  splits:
700
  - name: train
701
- num_bytes: 50397518
702
  num_examples: 1741
703
- download_size: 51442341
704
- dataset_size: 50397518
705
  - config_name: markdown.ebooks
706
  features:
707
  - name: id
@@ -716,19 +768,19 @@ dataset_info:
716
  dtype: string
717
  - name: date_published
718
  dtype: string
719
- - name: source_type
720
- dtype: string
721
  - name: authors
 
 
722
  dtype: string
723
  - name: filename
724
  dtype: string
725
  splits:
726
  - name: train
727
- num_bytes: 1878412
728
  num_examples: 4
729
- download_size: 1897759
730
- dataset_size: 1878412
731
- - config_name: nonarxiv_papers
732
  features:
733
  - name: id
734
  dtype: string
@@ -742,21 +794,19 @@ dataset_info:
742
  dtype: string
743
  - name: date_published
744
  dtype: string
745
- - name: source_filetype
746
- dtype: string
747
- - name: abstract
748
- dtype: string
749
  - name: authors
750
  sequence: string
751
- - name: filename
752
  dtype: string
 
 
753
  splits:
754
  - name: train
755
- num_bytes: 10107066
756
- num_examples: 200
757
- download_size: 10168097
758
- dataset_size: 10107066
759
- - config_name: qualiacomputing.com
760
  features:
761
  - name: id
762
  dtype: string
@@ -770,15 +820,23 @@ dataset_info:
770
  dtype: string
771
  - name: date_published
772
  dtype: string
773
- - name: paged_url
 
 
 
 
 
 
774
  dtype: string
 
 
775
  splits:
776
  - name: train
777
- num_bytes: 4330805
778
- num_examples: 299
779
- download_size: 4367941
780
- dataset_size: 4330805
781
- - config_name: reports
782
  features:
783
  - name: id
784
  dtype: string
@@ -792,21 +850,19 @@ dataset_info:
792
  dtype: string
793
  - name: date_published
794
  dtype: string
795
- - name: source_filetype
796
- dtype: string
797
- - name: abstract
798
- dtype: string
799
  - name: authors
800
  sequence: string
801
- - name: filename
802
  dtype: string
 
 
803
  splits:
804
  - name: train
805
- num_bytes: 4604618
806
- num_examples: 55
807
- download_size: 4626267
808
- dataset_size: 4604618
809
- - config_name: stampy
810
  features:
811
  - name: id
812
  dtype: string
@@ -820,23 +876,21 @@ dataset_info:
820
  dtype: string
821
  - name: date_published
822
  dtype: string
823
- - name: source_filetype
824
- dtype: string
825
  - name: authors
 
 
826
  dtype: string
827
- - name: question
828
- dtype: string
829
- - name: answer
830
  dtype: string
831
- - name: entry
832
  dtype: string
833
  splits:
834
  - name: train
835
- num_bytes: 1574712
836
- num_examples: 291
837
- download_size: 1617016
838
- dataset_size: 1574712
839
- - config_name: vkrakovna.wordpress.com
840
  features:
841
  - name: id
842
  dtype: string
@@ -850,14 +904,18 @@ dataset_info:
850
  dtype: string
851
  - name: date_published
852
  dtype: string
 
 
853
  - name: paged_url
854
  dtype: string
 
 
855
  splits:
856
  - name: train
857
- num_bytes: 298604
858
  num_examples: 46
859
- download_size: 303703
860
- dataset_size: 298604
861
  - config_name: waitbutwhy
862
  features:
863
  - name: id
@@ -872,17 +930,17 @@ dataset_info:
872
  dtype: string
873
  - name: date_published
874
  dtype: string
875
- - name: source_type
876
- dtype: string
877
  - name: authors
878
  dtype: string
 
 
879
  splits:
880
  - name: train
881
- num_bytes: 157197
882
  num_examples: 2
883
- download_size: 158517
884
- dataset_size: 157197
885
- - config_name: yudkowsky.net
886
  features:
887
  - name: id
888
  dtype: string
@@ -896,14 +954,18 @@ dataset_info:
896
  dtype: string
897
  - name: date_published
898
  dtype: string
 
 
899
  - name: paged_url
900
  dtype: string
 
 
901
  splits:
902
  - name: train
903
- num_bytes: 571526
904
  num_examples: 23
905
- download_size: 575271
906
- dataset_size: 571526
907
  ---
908
  # AI Alignment Research Dataset
909
 
@@ -981,4 +1043,4 @@ The scraper to generate this dataset is open-sourced on [GitHub](https://github.
981
 
982
  For more information, here is the [paper](https://arxiv.org/abs/2206.02841) and [LessWrong](https://www.lesswrong.com/posts/FgjcHiWvADgsocE34/a-descriptive-not-prescriptive-overview-of-current-ai) post. Please use the following citation when using the dataset:
983
 
984
- Kirchner, J. H., Smith, L., Thibodeau, J., McDonnell, K., and Reynolds, L. "Understanding AI alignment research: A Systematic Analysis." arXiv preprint arXiv:2022.4338861 (2022).
 
22
  dtype: string
23
  - name: date_published
24
  dtype: string
25
+ - name: authors
26
+ dtype: string
27
  - name: source_filetype
28
  dtype: string
29
  - name: converted_with
30
  dtype: string
31
  - name: book_title
32
  dtype: string
 
 
33
  - name: paged_url
34
  dtype: string
35
+ - name: summary
36
+ sequence: string
37
+ - name: question
38
+ dtype: string
39
+ - name: answer
40
+ dtype: string
41
+ - name: entry
42
+ dtype: string
43
  - name: karma
44
  dtype: int32
45
  - name: votes
 
100
  dtype: string
101
  - name: filename
102
  dtype: string
 
 
 
 
 
 
103
  splits:
104
  - name: train
105
+ num_bytes: 319462301
106
+ num_examples: 9688
107
+ download_size: 366146519
108
+ dataset_size: 319462301
109
  - config_name: agentmodels
110
  features:
111
  - name: id
 
120
  dtype: string
121
  - name: date_published
122
  dtype: string
123
+ - name: authors
124
+ sequence: string
125
  - name: source_filetype
126
  dtype: string
127
  - name: converted_with
128
  dtype: string
129
  - name: book_title
130
  dtype: string
 
 
131
  splits:
132
  - name: train
133
+ num_bytes: 386274
134
  num_examples: 21
135
+ download_size: 1203135
136
+ dataset_size: 386274
137
+ - config_name: aiimpacts
138
  features:
139
  - name: id
140
  dtype: string
 
148
  dtype: string
149
  - name: date_published
150
  dtype: string
151
+ - name: authors
152
+ sequence: string
153
  - name: paged_url
154
  dtype: string
155
+ - name: summary
156
+ sequence: string
157
  splits:
158
  - name: train
159
+ num_bytes: 3219417
160
  num_examples: 253
161
+ download_size: 3290483
162
+ dataset_size: 3219417
163
  - config_name: aipulse.org
164
  features:
165
  - name: id
 
174
  dtype: string
175
  - name: date_published
176
  dtype: string
177
+ - name: authors
178
+ sequence: string
179
  - name: paged_url
180
  dtype: string
181
  splits:
182
  - name: train
183
+ num_bytes: 652332
184
  num_examples: 23
185
  download_size: 655946
186
+ dataset_size: 652332
187
  - config_name: aisafety.camp
188
  features:
189
  - name: id
 
198
  dtype: string
199
  - name: date_published
200
  dtype: string
201
+ - name: authors
202
+ sequence: string
203
  - name: paged_url
204
  dtype: string
205
+ - name: summary
206
+ sequence: string
207
  splits:
208
  - name: train
209
+ num_bytes: 96728
210
  num_examples: 9
211
+ download_size: 99056
212
+ dataset_size: 96728
213
+ - config_name: aisafety.info
214
+ features:
215
+ - name: id
216
+ dtype: string
217
+ - name: source
218
+ dtype: string
219
+ - name: title
220
+ dtype: string
221
+ - name: text
222
+ dtype: large_string
223
+ - name: url
224
+ dtype: string
225
+ - name: date_published
226
+ dtype: string
227
+ - name: authors
228
+ sequence: string
229
+ - name: source_filetype
230
+ dtype: string
231
+ - name: question
232
+ dtype: string
233
+ - name: answer
234
+ dtype: string
235
+ - name: entry
236
+ dtype: string
237
+ splits:
238
+ - name: train
239
+ num_bytes: 1573839
240
+ num_examples: 291
241
+ download_size: 1616143
242
+ dataset_size: 1573839
243
  - config_name: alignmentforum
244
  features:
245
  - name: id
 
255
  - name: date_published
256
  dtype: string
257
  - name: authors
258
+ sequence: string
 
 
 
 
259
  - name: karma
260
  dtype: int32
261
  - name: votes
 
272
  dtype: string
273
  splits:
274
  - name: train
275
+ num_bytes: 73451397
276
+ num_examples: 2643
277
+ download_size: 75161883
278
+ dataset_size: 73451397
279
  - config_name: alignment_newsletter
280
  features:
281
  - name: id
 
290
  dtype: string
291
  - name: date_published
292
  dtype: string
293
+ - name: authors
294
+ sequence: string
295
  - name: converted_with
296
  dtype: string
297
  - name: source_type
 
312
  dtype: string
313
  - name: read_more
314
  dtype: string
 
 
315
  splits:
316
  - name: train
317
+ num_bytes: 3202660
318
+ num_examples: 1504
319
+ download_size: 3553667
320
+ dataset_size: 3202660
321
  - config_name: arbital
322
  features:
323
  - name: id
 
332
  dtype: string
333
  - name: date_published
334
  dtype: string
 
 
335
  - name: authors
336
+ sequence: string
337
+ - name: source_filetype
338
  dtype: string
339
  - name: alias
340
  dtype: string
341
+ - name: summary
342
+ sequence: string
343
  splits:
344
  - name: train
345
+ num_bytes: 3295285
346
+ num_examples: 728
347
+ download_size: 3624231
348
+ dataset_size: 3295285
349
+ - config_name: arxiv
350
  features:
351
  - name: id
352
  dtype: string
 
361
  - name: date_published
362
  dtype: string
363
  - name: authors
364
+ sequence: string
365
  - name: source_type
366
  dtype: string
367
  - name: converted_with
 
380
  dtype: string
381
  - name: categories
382
  sequence: string
383
+ - name: summary
384
+ sequence: string
385
  splits:
386
  - name: train
387
+ num_bytes: 140132334
388
  num_examples: 1144
389
+ download_size: 155841593
390
+ dataset_size: 140132334
391
  - config_name: audio_transcripts
392
  features:
393
  - name: id
 
402
  dtype: string
403
  - name: date_published
404
  dtype: string
405
+ - name: authors
406
+ sequence: string
407
  - name: source_filetype
408
  dtype: string
409
  - name: converted_with
410
  dtype: string
 
 
411
  splits:
412
  - name: train
413
+ num_bytes: 1312932
414
  num_examples: 37
415
+ download_size: 1334992
416
+ dataset_size: 1312932
417
  - config_name: carado.moe
418
  features:
419
  - name: id
 
428
  dtype: string
429
  - name: date_published
430
  dtype: string
 
 
431
  - name: authors
432
+ sequence: string
433
+ - name: source_type
434
  dtype: string
435
  splits:
436
  - name: train
437
+ num_bytes: 1046524
438
+ num_examples: 248
439
+ download_size: 1077923
440
+ dataset_size: 1046524
441
+ - config_name: cold_takes
442
  features:
443
  - name: id
444
  dtype: string
 
452
  dtype: string
453
  - name: date_published
454
  dtype: string
455
+ - name: authors
456
+ sequence: string
457
  - name: source_type
458
  dtype: string
459
  splits:
460
  - name: train
461
+ num_bytes: 2122519
462
  num_examples: 102
463
+ download_size: 3304055
464
+ dataset_size: 2122519
465
+ - config_name: deepmind_blog
466
  features:
467
  - name: id
468
  dtype: string
 
476
  dtype: string
477
  - name: date_published
478
  dtype: string
479
+ - name: authors
480
+ sequence: string
481
  - name: source_type
482
  dtype: string
483
  splits:
484
  - name: train
485
+ num_bytes: 131197
486
  num_examples: 10
487
+ download_size: 132884
488
+ dataset_size: 131197
489
  - config_name: distill
490
  features:
491
  - name: id
 
500
  dtype: string
501
  - name: date_published
502
  dtype: string
503
+ - name: authors
504
+ sequence: string
505
  - name: source_type
506
  dtype: string
507
  - name: converted_with
508
  dtype: string
 
 
509
  - name: abstract
510
  dtype: string
511
  - name: journal_ref
 
518
  dtype: string
519
  splits:
520
  - name: train
521
+ num_bytes: 2246011
522
+ num_examples: 51
523
+ download_size: 2435649
524
+ dataset_size: 2246011
525
  - config_name: eaforum
526
  features:
527
  - name: id
 
537
  - name: date_published
538
  dtype: string
539
  - name: authors
540
+ sequence: string
 
 
 
 
541
  - name: karma
542
  dtype: int32
543
  - name: votes
 
552
  dtype: string
553
  - name: source_type
554
  dtype: string
555
+ - name: summary
556
+ sequence: string
557
  splits:
558
  - name: train
559
+ num_bytes: 8427856
560
+ num_examples: 401
561
+ download_size: 8558545
562
+ dataset_size: 8427856
563
  - config_name: gdocs
564
  features:
565
  - name: id
 
574
  dtype: string
575
  - name: date_published
576
  dtype: string
577
+ - name: authors
578
+ sequence: string
579
  - name: source_filetype
580
  dtype: string
581
  - name: converted_with
582
  dtype: string
 
 
583
  - name: docx_name
584
  dtype: string
585
  splits:
 
602
  dtype: string
603
  - name: date_published
604
  dtype: string
605
+ - name: authors
606
+ sequence: string
607
  - name: source_filetype
608
  dtype: string
609
  - name: converted_with
 
614
  dtype: string
615
  splits:
616
  - name: train
617
+ num_bytes: 19583452
618
  num_examples: 20
619
+ download_size: 19746001
620
+ dataset_size: 19583452
621
  - config_name: generative.ink
622
  features:
623
  - name: id
 
632
  dtype: string
633
  - name: date_published
634
  dtype: string
635
+ - name: authors
636
+ sequence: string
637
  - name: source_type
638
  dtype: string
639
  splits:
640
  - name: train
641
+ num_bytes: 516173
642
  num_examples: 19
643
+ download_size: 525315
644
+ dataset_size: 516173
645
  - config_name: gwern_blog
646
  features:
647
  - name: id
 
657
  - name: date_published
658
  dtype: string
659
  - name: authors
660
+ sequence: string
661
  - name: description
662
  dtype: string
663
  splits:
664
  - name: train
665
+ num_bytes: 735083
666
  num_examples: 7
667
+ download_size: 742259
668
+ dataset_size: 735083
669
+ - config_name: importai
670
  features:
671
  - name: id
672
  dtype: string
 
680
  dtype: string
681
  - name: date_published
682
  dtype: string
683
+ - name: authors
684
+ sequence: string
685
+ - name: source_type
686
  dtype: string
687
  splits:
688
  - name: train
689
+ num_bytes: 243239
690
+ num_examples: 15
691
+ download_size: 246559
692
+ dataset_size: 243239
693
+ - config_name: jsteinhardt_blog
694
  features:
695
  - name: id
696
  dtype: string
 
704
  dtype: string
705
  - name: date_published
706
  dtype: string
707
+ - name: authors
708
+ sequence: string
709
  - name: paged_url
710
  dtype: string
711
+ - name: summary
712
+ sequence: string
713
  splits:
714
  - name: train
715
+ num_bytes: 657649
716
  num_examples: 39
717
+ download_size: 669942
718
+ dataset_size: 657649
719
  - config_name: lesswrong
720
  features:
721
  - name: id
 
731
  - name: date_published
732
  dtype: string
733
  - name: authors
734
+ sequence: string
 
 
 
 
735
  - name: karma
736
  dtype: int32
737
  - name: votes
 
746
  dtype: string
747
  - name: source_type
748
  dtype: string
749
+ - name: summary
750
+ sequence: string
751
  splits:
752
  - name: train
753
+ num_bytes: 50402273
754
  num_examples: 1741
755
+ download_size: 51420245
756
+ dataset_size: 50402273
757
  - config_name: markdown.ebooks
758
  features:
759
  - name: id
 
768
  dtype: string
769
  - name: date_published
770
  dtype: string
 
 
771
  - name: authors
772
+ sequence: string
773
+ - name: source_type
774
  dtype: string
775
  - name: filename
776
  dtype: string
777
  splits:
778
  - name: train
779
+ num_bytes: 1878436
780
  num_examples: 4
781
+ download_size: 1897775
782
+ dataset_size: 1878436
783
+ - config_name: miri
784
  features:
785
  - name: id
786
  dtype: string
 
794
  dtype: string
795
  - name: date_published
796
  dtype: string
 
 
 
 
797
  - name: authors
798
  sequence: string
799
+ - name: paged_url
800
  dtype: string
801
+ - name: summary
802
+ sequence: string
803
  splits:
804
  - name: train
805
+ num_bytes: 6712562
806
+ num_examples: 492
807
+ download_size: 6856015
808
+ dataset_size: 6712562
809
+ - config_name: nonarxiv_papers
810
  features:
811
  - name: id
812
  dtype: string
 
820
  dtype: string
821
  - name: date_published
822
  dtype: string
823
+ - name: authors
824
+ sequence: string
825
+ - name: source_filetype
826
+ dtype: string
827
+ - name: abstract
828
+ dtype: string
829
+ - name: filename
830
  dtype: string
831
+ - name: summary
832
+ sequence: string
833
  splits:
834
  - name: train
835
+ num_bytes: 10369405
836
+ num_examples: 200
837
+ download_size: 10408494
838
+ dataset_size: 10369405
839
+ - config_name: qualiacomputing
840
  features:
841
  - name: id
842
  dtype: string
 
850
  dtype: string
851
  - name: date_published
852
  dtype: string
 
 
 
 
853
  - name: authors
854
  sequence: string
855
+ - name: paged_url
856
  dtype: string
857
+ - name: summary
858
+ sequence: string
859
  splits:
860
  - name: train
861
+ num_bytes: 5456683
862
+ num_examples: 301
863
+ download_size: 5536074
864
+ dataset_size: 5456683
865
+ - config_name: reports
866
  features:
867
  - name: id
868
  dtype: string
 
876
  dtype: string
877
  - name: date_published
878
  dtype: string
 
 
879
  - name: authors
880
+ sequence: string
881
+ - name: source_filetype
882
  dtype: string
883
+ - name: abstract
 
 
884
  dtype: string
885
+ - name: filename
886
  dtype: string
887
  splits:
888
  - name: train
889
+ num_bytes: 4604618
890
+ num_examples: 55
891
+ download_size: 4621199
892
+ dataset_size: 4604618
893
+ - config_name: vkrakovna_blog
894
  features:
895
  - name: id
896
  dtype: string
 
904
  dtype: string
905
  - name: date_published
906
  dtype: string
907
+ - name: authors
908
+ sequence: string
909
  - name: paged_url
910
  dtype: string
911
+ - name: summary
912
+ sequence: string
913
  splits:
914
  - name: train
915
+ num_bytes: 360866
916
  num_examples: 46
917
+ download_size: 369604
918
+ dataset_size: 360866
919
  - config_name: waitbutwhy
920
  features:
921
  - name: id
 
930
  dtype: string
931
  - name: date_published
932
  dtype: string
 
 
933
  - name: authors
934
  dtype: string
935
+ - name: source_type
936
+ dtype: string
937
  splits:
938
  - name: train
939
+ num_bytes: 157217
940
  num_examples: 2
941
+ download_size: 158533
942
+ dataset_size: 157217
943
+ - config_name: yudkowsky_blog
944
  features:
945
  - name: id
946
  dtype: string
 
954
  dtype: string
955
  - name: date_published
956
  dtype: string
957
+ - name: authors
958
+ sequence: string
959
  - name: paged_url
960
  dtype: string
961
+ - name: summary
962
+ sequence: string
963
  splits:
964
  - name: train
965
+ num_bytes: 594207
966
  num_examples: 23
967
+ download_size: 604224
968
+ dataset_size: 594207
969
  ---
970
  # AI Alignment Research Dataset
971
 
 
1043
 
1044
  For more information, here is the [paper](https://arxiv.org/abs/2206.02841) and [LessWrong](https://www.lesswrong.com/posts/FgjcHiWvADgsocE34/a-descriptive-not-prescriptive-overview-of-current-ai) post. Please use the following citation when using the dataset:
1045
 
1046
+ Kirchner, J. H., Smith, L., Thibodeau, J., McDonnell, K., and Reynolds, L. "Understanding AI alignment research: A Systematic Analysis." arXiv preprint arXiv:2022.4338861 (2022).
agentmodels.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
aiimpacts.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
aiimpacts.org.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
aisafety.camp.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
aisafety.info.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
alignment-research-dataset.py CHANGED
@@ -24,6 +24,7 @@ _VERSION_ = '0.0.0'
24
 
25
 
26
  def iterate_file(filename):
 
27
  with open(filename) as f:
28
  for l in f:
29
  try:
@@ -88,179 +89,172 @@ DEFAULT_FEATURES = {
88
  'text': Value('large_string'),
89
  'url': Value('string'),
90
  'date_published': Value(dtype='string'),
 
91
  }
92
 
 
93
  # Per datasource additional features
94
  DATASOURCES = {
95
  'agentmodels': {
96
- 'source_filetype': Value(dtype='string', id=None),
97
- 'converted_with': Value(dtype='string', id=None),
98
- 'book_title': Value(dtype='string', id=None),
99
- 'authors': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None)
100
  },
101
- 'aiimpacts.org': {
102
- 'paged_url': Value(dtype='string', id=None)
 
103
  },
104
  'aipulse.org': {
105
- 'paged_url': Value(dtype='string', id=None)
106
  },
107
  'aisafety.camp': {
108
- 'paged_url': Value(dtype='string', id=None)
 
 
 
 
 
 
 
109
  },
110
  'alignmentforum': {
111
- 'authors': Sequence({
112
- 'username': Value(dtype='string', id=None),
113
- 'displayName': Value(dtype='string', id=None),
114
- }),
115
- 'karma': Value(dtype='int32', id=None),
116
- 'votes': Value(dtype='int32', id=None),
117
- 'words': Value(dtype='int32', id=None),
118
- 'comment_count': Value(dtype='int32', id=None),
119
- 'tags': Sequence(feature=Value(dtype='string', id=None)),
120
- 'modified_at': Value(dtype='string', id=None),
121
- 'source_type': Value(dtype='string', id=None),
122
  },
123
  'alignment_newsletter': {
124
- 'converted_with': Value(dtype='string', id=None),
125
- 'source_type': Value(dtype='string', id=None),
126
- 'venue': Value(dtype='string', id=None),
127
- 'newsletter_category': Value(dtype='string', id=None),
128
- 'highlight': Value(dtype='int32', id=None),
129
- 'newsletter_number': Value(dtype='string', id=None),
130
- 'summarizer': Value(dtype='string', id=None),
131
- 'opinion': Value(dtype='string', id=None),
132
- 'prerequisites': Value(dtype='string', id=None),
133
- 'read_more': Value(dtype='string', id=None),
134
- 'authors': Value(dtype='string', id=None)
135
  },
136
- 'arbital': {
137
- 'source_filetype': Value(dtype='string', id=None),
138
- 'authors': Value(dtype='string', id=None),
139
- 'alias': Value(dtype='string', id=None)
140
  },
141
- 'arxiv_papers': {
142
- 'authors': Value(dtype='string', id=None),
143
- 'source_type': Value(dtype='string', id=None),
144
- 'converted_with': Value(dtype='string', id=None),
145
- 'data_last_modified': Value(dtype='string', id=None),
146
- 'abstract': Value(dtype='string', id=None),
147
- 'author_comment': Value(dtype='string', id=None),
148
- 'journal_ref': Value(dtype='string', id=None),
149
- 'doi': Value(dtype='string', id=None),
150
- 'primary_category': Value(dtype='string', id=None),
151
- 'categories': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None)
152
  },
153
  'audio_transcripts': {
154
- 'source_filetype': Value(dtype='string', id=None),
155
- 'converted_with': Value(dtype='string', id=None),
156
- 'authors': Value(dtype='string', id=None)
157
  },
158
  'carado.moe': {
159
- 'source_type': Value(dtype='string', id=None),
160
- 'authors': Value(dtype='string', id=None)
161
  },
162
- 'cold.takes': {
163
- 'source_type': Value(dtype='string', id=None)
164
  },
165
- 'deepmind.blog': {
166
- 'source_type': Value(dtype='string', id=None)
167
  },
168
  'distill': {
169
- 'source_type': Value(dtype='string', id=None),
170
- 'converted_with': Value(dtype='string', id=None),
171
- 'authors': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None),
172
- 'abstract': Value(dtype='string', id=None),
173
- 'journal_ref': Value(dtype='string', id=None),
174
- 'doi': Value(dtype='string', id=None),
175
- 'bibliography_bib': Sequence(feature={'title': Value(dtype='string', id=None)}, length=-1, id=None)
176
  },
177
  'eaforum': {
178
- 'authors': Sequence({
179
- 'username': Value(dtype='string', id=None),
180
- 'displayName': Value(dtype='string', id=None),
181
- }),
182
- 'karma': Value(dtype='int32', id=None),
183
- 'votes': Value(dtype='int32', id=None),
184
- 'words': Value(dtype='int32', id=None),
185
- 'comment_count': Value(dtype='int32', id=None),
186
- 'tags': Sequence(feature=Value(dtype='string', id=None)),
187
- 'modified_at': Value(dtype='string', id=None),
188
- 'source_type': Value(dtype='string', id=None),
189
  },
190
  'gdocs': {
191
- 'source_filetype': Value(dtype='string', id=None),
192
- 'converted_with': Value(dtype='string', id=None),
193
- 'authors': Value(dtype='string', id=None),
194
- 'docx_name': Value(dtype='string', id=None)
195
  },
196
  'gdrive_ebooks': {
197
- 'source_filetype': Value(dtype='string', id=None),
198
- 'converted_with': Value(dtype='string', id=None),
199
- 'chapter_names': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None),
200
- 'file_name': Value(dtype='string', id=None)
201
  },
202
  'generative.ink': {
203
- 'source_type': Value(dtype='string', id=None)
204
  },
205
  'gwern_blog': {
206
- 'authors': Value(dtype='string', id=None),
207
- 'description': Value(dtype='string', id=None)
208
  },
209
- 'intelligence.org': {
210
- 'paged_url': Value(dtype='string', id=None)
211
  },
212
- 'jsteinhardt.wordpress.com': {
213
- 'paged_url': Value(dtype='string', id=None)
 
214
  },
215
  'lesswrong': {
216
- 'authors': Sequence({
217
- 'username': Value(dtype='string', id=None),
218
- 'displayName': Value(dtype='string', id=None),
219
- }),
220
- 'karma': Value(dtype='int32', id=None),
221
- 'votes': Value(dtype='int32', id=None),
222
- 'words': Value(dtype='int32', id=None),
223
- 'comment_count': Value(dtype='int32', id=None),
224
- 'tags': Sequence(feature=Value(dtype='string', id=None)),
225
- 'modified_at': Value(dtype='string', id=None),
226
- 'source_type': Value(dtype='string', id=None),
227
  },
228
  'markdown.ebooks': {
229
- 'source_type': Value(dtype='string', id=None),
230
- 'authors': Value(dtype='string', id=None),
231
- 'filename': Value(dtype='string', id=None)
 
 
 
232
  },
233
  'nonarxiv_papers': {
234
- 'source_filetype': Value(dtype='string', id=None),
235
- 'abstract': Value(dtype='string', id=None),
236
- 'authors': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None),
237
- 'filename': Value(dtype='string', id=None)
238
  },
239
- 'qualiacomputing.com': {
240
- 'paged_url': Value(dtype='string', id=None)
 
241
  },
242
  'reports': {
243
- 'source_filetype': Value(dtype='string', id=None),
244
- 'abstract': Value(dtype='string', id=None),
245
- 'authors': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None),
246
- 'filename': Value(dtype='string', id=None)
247
- },
248
- 'stampy': {
249
- 'source_filetype': Value(dtype='string', id=None),
250
- 'authors': Value(dtype='string', id=None),
251
- 'question': Value(dtype='string', id=None),
252
- 'answer': Value(dtype='string', id=None),
253
- 'entry': Value(dtype='string', id=None)
254
  },
255
- 'vkrakovna.wordpress.com': {
256
- 'paged_url': Value(dtype='string', id=None)
 
257
  },
258
  'waitbutwhy': {
259
- 'source_type': Value(dtype='string', id=None),
260
- 'authors': Value(dtype='string', id=None)
261
  },
262
- 'yudkowsky.net': {
263
- 'paged_url': Value(dtype='string', id=None)
 
264
  },
265
  }
266
 
 
24
 
25
 
26
  def iterate_file(filename):
27
+ print(filename)
28
  with open(filename) as f:
29
  for l in f:
30
  try:
 
89
  'text': Value('large_string'),
90
  'url': Value('string'),
91
  'date_published': Value(dtype='string'),
92
+ 'authors': Sequence(feature=Value(dtype='string'), length=-1)
93
  }
94
 
95
+
96
  # Per datasource additional features
97
  DATASOURCES = {
98
  'agentmodels': {
99
+ 'source_filetype': Value(dtype='string'),
100
+ 'converted_with': Value(dtype='string'),
101
+ 'book_title': Value(dtype='string'),
 
102
  },
103
+ 'aiimpacts': {
104
+ 'paged_url': Value(dtype='string'),
105
+ 'summary': Sequence(feature=Value(dtype='string')),
106
  },
107
  'aipulse.org': {
108
+ 'paged_url': Value(dtype='string')
109
  },
110
  'aisafety.camp': {
111
+ 'paged_url': Value(dtype='string'),
112
+ 'summary': Sequence(feature=Value(dtype='string')),
113
+ },
114
+ 'aisafety.info': {
115
+ 'source_filetype': Value(dtype='string'),
116
+ 'question': Value(dtype='string'),
117
+ 'answer': Value(dtype='string'),
118
+ 'entry': Value(dtype='string')
119
  },
120
  'alignmentforum': {
121
+ 'karma': Value(dtype='int32'),
122
+ 'votes': Value(dtype='int32'),
123
+ 'words': Value(dtype='int32'),
124
+ 'comment_count': Value(dtype='int32'),
125
+ 'tags': Sequence(feature=Value(dtype='string')),
126
+ 'modified_at': Value(dtype='string'),
127
+ 'source_type': Value(dtype='string'),
 
 
 
 
128
  },
129
  'alignment_newsletter': {
130
+ 'converted_with': Value(dtype='string'),
131
+ 'source_type': Value(dtype='string'),
132
+ 'venue': Value(dtype='string'),
133
+ 'newsletter_category': Value(dtype='string'),
134
+ 'highlight': Value(dtype='int32'),
135
+ 'newsletter_number': Value(dtype='string'),
136
+ 'summarizer': Value(dtype='string'),
137
+ 'opinion': Value(dtype='string'),
138
+ 'prerequisites': Value(dtype='string'),
139
+ 'read_more': Value(dtype='string'),
 
140
  },
141
+ 'arbital': {
142
+ 'source_filetype': Value(dtype='string'),
143
+ 'alias': Value(dtype='string'),
144
+ 'summary': Sequence(feature=Value(dtype='string')),
145
  },
146
+ 'arxiv': {
147
+ 'source_type': Value(dtype='string'),
148
+ 'converted_with': Value(dtype='string'),
149
+ 'data_last_modified': Value(dtype='string'),
150
+ 'abstract': Value(dtype='string'),
151
+ 'author_comment': Value(dtype='string'),
152
+ 'journal_ref': Value(dtype='string'),
153
+ 'doi': Value(dtype='string'),
154
+ 'primary_category': Value(dtype='string'),
155
+ 'categories': Sequence(feature=Value(dtype='string'), length=-1),
156
+ 'summary': Sequence(feature=Value(dtype='string')),
157
  },
158
  'audio_transcripts': {
159
+ 'source_filetype': Value(dtype='string'),
160
+ 'converted_with': Value(dtype='string'),
 
161
  },
162
  'carado.moe': {
163
+ 'source_type': Value(dtype='string'),
 
164
  },
165
+ 'cold_takes': {
166
+ 'source_type': Value(dtype='string')
167
  },
168
+ 'deepmind_blog': {
169
+ 'source_type': Value(dtype='string')
170
  },
171
  'distill': {
172
+ 'source_type': Value(dtype='string'),
173
+ 'converted_with': Value(dtype='string'),
174
+ 'abstract': Value(dtype='string'),
175
+ 'journal_ref': Value(dtype='string'),
176
+ 'doi': Value(dtype='string'),
177
+ 'bibliography_bib': Sequence(feature={'title': Value(dtype='string')}, length=-1)
 
178
  },
179
  'eaforum': {
180
+ 'karma': Value(dtype='int32'),
181
+ 'votes': Value(dtype='int32'),
182
+ 'words': Value(dtype='int32'),
183
+ 'comment_count': Value(dtype='int32'),
184
+ 'tags': Sequence(feature=Value(dtype='string')),
185
+ 'modified_at': Value(dtype='string'),
186
+ 'source_type': Value(dtype='string'),
187
+ 'summary': Sequence(feature=Value(dtype='string')),
 
 
 
188
  },
189
  'gdocs': {
190
+ 'source_filetype': Value(dtype='string'),
191
+ 'converted_with': Value(dtype='string'),
192
+ 'docx_name': Value(dtype='string')
 
193
  },
194
  'gdrive_ebooks': {
195
+ 'source_filetype': Value(dtype='string'),
196
+ 'converted_with': Value(dtype='string'),
197
+ 'chapter_names': Sequence(feature=Value(dtype='string'), length=-1),
198
+ 'file_name': Value(dtype='string')
199
  },
200
  'generative.ink': {
201
+ 'source_type': Value(dtype='string')
202
  },
203
  'gwern_blog': {
204
+ 'description': Value(dtype='string')
 
205
  },
206
+ 'importai': {
207
+ 'source_type': Value(dtype='string')
208
  },
209
+ 'jsteinhardt_blog': {
210
+ 'paged_url': Value(dtype='string'),
211
+ 'summary': Sequence(feature=Value(dtype='string')),
212
  },
213
  'lesswrong': {
214
+ 'karma': Value(dtype='int32'),
215
+ 'votes': Value(dtype='int32'),
216
+ 'words': Value(dtype='int32'),
217
+ 'comment_count': Value(dtype='int32'),
218
+ 'tags': Sequence(feature=Value(dtype='string')),
219
+ 'modified_at': Value(dtype='string'),
220
+ 'source_type': Value(dtype='string'),
221
+ 'summary': Sequence(feature=Value(dtype='string')),
 
 
 
222
  },
223
  'markdown.ebooks': {
224
+ 'source_type': Value(dtype='string'),
225
+ 'filename': Value(dtype='string')
226
+ },
227
+ 'miri': {
228
+ 'paged_url': Value(dtype='string'),
229
+ 'summary': Sequence(feature=Value(dtype='string')),
230
  },
231
  'nonarxiv_papers': {
232
+ 'source_filetype': Value(dtype='string'),
233
+ 'abstract': Value(dtype='string'),
234
+ 'filename': Value(dtype='string'),
235
+ 'summary': Sequence(feature=Value(dtype='string')),
236
  },
237
+ 'qualiacomputing': {
238
+ 'paged_url': Value(dtype='string'),
239
+ 'summary': Sequence(feature=Value(dtype='string')),
240
  },
241
  'reports': {
242
+ 'source_filetype': Value(dtype='string'),
243
+ 'abstract': Value(dtype='string'),
244
+ 'authors': Sequence(feature=Value(dtype='string'), length=-1),
245
+ 'filename': Value(dtype='string')
 
 
 
 
 
 
 
246
  },
247
+ 'vkrakovna_blog': {
248
+ 'paged_url': Value(dtype='string'),
249
+ 'summary': Sequence(feature=Value(dtype='string')),
250
  },
251
  'waitbutwhy': {
252
+ 'source_type': Value(dtype='string'),
253
+ 'authors': Value(dtype='string')
254
  },
255
+ 'yudkowsky_blog': {
256
+ 'paged_url': Value(dtype='string'),
257
+ 'summary': Sequence(feature=Value(dtype='string')),
258
  },
259
  }
260
 
alignment_newsletter.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
arbital.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
audio_transcripts.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
carado.moe.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
cold.takes.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
cold_takes.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
deepmind.blog.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
deepmind_blog.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
distill.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
eaforum.jsonl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d85a64727ec4603dbbdeb82aa3f6b4a1c6338e1591bf6c7e834cfdac6ae6f138
3
- size 8598593
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5be79fd78ce3a9d7558dd864bf140b6632fc13f1588955e442e4299a610284e0
3
+ size 8558545
gdocs.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
generative.ink.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
gwern_blog.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
import.ai.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
importai.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
jsteinhardt.wordpress.com.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
jsteinhardt_blog.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
markdown.ebooks.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
intelligence.org.jsonl → miri.jsonl RENAMED
The diff for this file is too large to render. See raw diff
 
ml.safety.newsletter.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
ml_safety_newsletter.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
qualiacomputing.com.jsonl → qualiacomputing.jsonl RENAMED
The diff for this file is too large to render. See raw diff
 
reports.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
stampy.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
vkrakovna.wordpress.com.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
vkrakovna_blog.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
waitbutwhy.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
yudkowsky.net.jsonl DELETED
The diff for this file is too large to render. See raw diff
 
yudkowsky_blog.jsonl ADDED
The diff for this file is too large to render. See raw diff