Datasets:
Search is not available for this dataset
input_ids listlengths 2.05k 2.05k |
|---|
[
35,
99,
7,
71,
58,
58,
22,
29,
19,
93,
26,
18,
83,
62,
28,
126,
86,
90,
122,
3,
67,
32,
96,
31,
122,
92,
48,
11,
82,
18,
44,
35,
108,
39,
22,
4,
112,
12,
76,
75,
17,
82,
81,
68,
86,
61,
99,
29,
125,
93,
95,
103,
... |
[
56,
120,
13,
25,
49,
17,
77,
113,
89,
56,
12,
12,
39,
81,
59,
103,
51,
41,
12,
49,
115,
51,
120,
105,
123,
113,
25,
89,
61,
37,
52,
116,
18,
82,
76,
0,
20,
115,
49,
11,
75,
22,
113,
19,
125,
57,
64,
121,
86,
20,
101,... |
[
32,
96,
6,
34,
20,
38,
98,
70,
84,
102,
7,
28,
52,
48,
92,
112,
116,
47,
0,
111,
64,
71,
31,
42,
95,
46,
106,
110,
23,
87,
21,
85,
47,
43,
111,
107,
31,
95,
54,
27,
91,
118,
62,
126,
28,
46,
110,
17,
92,
81,
12,
45... |
[
57,
38,
57,
2,
11,
27,
102,
75,
12,
36,
36,
100,
32,
66,
91,
121,
121,
22,
86,
76,
54,
96,
27,
47,
62,
59,
126,
91,
2,
51,
115,
22,
59,
30,
123,
21,
85,
100,
19,
3,
118,
83,
67,
66,
43,
107,
94,
111,
123,
4,
68,
7,... |
[
26,
90,
59,
8,
123,
72,
2,
66,
39,
29,
26,
29,
62,
93,
126,
103,
34,
24,
30,
88,
98,
90,
1,
93,
59,
65,
94,
123,
0,
64,
17,
63,
36,
62,
100,
127,
81,
53,
126,
31,
95,
59,
123,
47,
117,
111,
43,
41,
32,
96,
41,
5,
... |
[
16,
80,
16,
17,
11,
41,
13,
8,
105,
17,
77,
81,
18,
82,
72,
75,
8,
25,
35,
45,
72,
81,
109,
89,
6,
99,
80,
70,
0,
64,
28,
92,
36,
100,
53,
46,
117,
43,
0,
107,
31,
51,
42,
115,
24,
64,
53,
49,
88,
23,
110,
106,
6... |
[
5,
1,
65,
22,
69,
36,
41,
86,
105,
100,
21,
45,
85,
109,
28,
8,
14,
72,
8,
92,
78,
62,
29,
72,
126,
93,
22,
86,
4,
3,
15,
79,
68,
67,
11,
9,
27,
75,
12,
91,
51,
73,
9,
73,
34,
115,
54,
57,
41,
121,
12,
13,
53,
... |
[
28,
92,
43,
8,
62,
29,
40,
18,
17,
82,
31,
126,
26,
95,
23,
59,
93,
72,
87,
104,
107,
13,
123,
54,
90,
81,
52,
116,
118,
38,
77,
12,
76,
102,
13,
28,
51,
92,
10,
1,
47,
77,
111,
115,
56,
74,
14,
63,
120,
78,
127,
6... |
[
58,
122,
46,
110,
55,
44,
40,
57,
31,
2,
35,
119,
95,
12,
26,
4,
104,
20,
76,
28,
68,
36,
108,
121,
92,
84,
66,
5,
13,
77,
12,
53,
69,
100,
99,
32,
10,
18,
117,
96,
82,
76,
74,
11,
75,
90,
54,
50,
60,
39,
124,
37,
... |
[
32,
3,
42,
39,
103,
106,
67,
11,
52,
116,
43,
59,
123,
96,
107,
41,
75,
31,
95,
105,
41,
4,
105,
68,
7,
71,
9,
73,
45,
109,
0,
64,
21,
85,
36,
44,
49,
100,
113,
11,
60,
124,
25,
108,
89,
62,
42,
75,
126,
106,
50,
1... |
[
34,
23,
4,
98,
25,
50,
44,
60,
114,
18,
108,
33,
60,
24,
82,
30,
68,
88,
124,
94,
61,
3,
87,
29,
125,
93,
124,
89,
97,
67,
41,
105,
28,
44,
34,
52,
37,
53,
92,
116,
22,
27,
117,
98,
86,
101,
63,
91,
127,
30,
53,
59... |
[
24,
88,
24,
88,
54,
118,
33,
97,
28,
92,
25,
3,
21,
67,
61,
125,
85,
89,
51,
23,
34,
30,
87,
115,
94,
98,
33,
42,
97,
106,
43,
35,
31,
107,
95,
99,
37,
35,
101,
99,
40,
32,
43,
104,
3,
96,
67,
107,
47,
111,
0,
29,
... |
[
63,
127,
29,
93,
7,
71,
11,
75,
31,
44,
23,
95,
22,
108,
23,
86,
87,
87,
53,
7,
12,
27,
50,
17,
71,
43,
81,
19,
76,
0,
91,
16,
3,
80,
15,
16,
107,
67,
22,
79,
83,
59,
5,
117,
26,
90,
86,
69,
30,
17,
45,
109,
27,
... |
[
40,
104,
31,
95,
54,
21,
37,
46,
85,
34,
118,
101,
110,
2,
66,
98,
35,
99,
15,
79,
20,
5,
55,
38,
44,
119,
108,
32,
84,
102,
47,
7,
55,
31,
96,
119,
71,
14,
95,
111,
60,
69,
78,
124,
54,
118,
54,
118,
4,
68,
52,
12... |
[
11,
3,
31,
20,
46,
67,
75,
1,
8,
72,
58,
84,
110,
23,
60,
58,
122,
95,
124,
9,
65,
26,
55,
12,
122,
31,
50,
95,
42,
59,
90,
123,
56,
39,
76,
19,
103,
45,
73,
120,
87,
83,
114,
106,
22,
119,
26,
109,
90,
35,
99,
86,... |
[
2,
66,
2,
30,
41,
37,
34,
40,
55,
66,
101,
11,
6,
119,
32,
15,
79,
94,
98,
57,
61,
70,
96,
58,
75,
104,
125,
16,
121,
58,
80,
54,
60,
1,
124,
19,
83,
65,
37,
105,
118,
101,
122,
122,
31,
50,
23,
87,
95,
57,
0,
64,
... |
[
2,
28,
66,
23,
14,
18,
92,
34,
87,
82,
98,
78,
11,
75,
0,
1,
65,
3,
64,
67,
1,
35,
14,
65,
28,
4,
78,
92,
99,
35,
99,
68,
34,
34,
98,
0,
98,
16,
32,
59,
96,
46,
80,
26,
110,
64,
28,
92,
123,
56,
90,
120,
28,
92... |
[
0,
54,
52,
118,
42,
55,
106,
119,
64,
116,
8,
72,
10,
52,
116,
74,
54,
118,
15,
49,
4,
2,
22,
33,
68,
50,
66,
113,
14,
97,
86,
114,
32,
96,
29,
78,
93,
22,
46,
86,
79,
110,
42,
106,
53,
36,
100,
117,
46,
110,
36,
4... |
[
8,
7,
72,
25,
71,
89,
59,
123,
6,
21,
70,
57,
121,
85,
50,
114,
55,
119,
63,
127,
49,
42,
106,
2,
4,
41,
68,
30,
51,
44,
53,
66,
108,
105,
7,
117,
41,
48,
115,
113,
105,
112,
71,
94,
59,
34,
123,
51,
115,
98,
28,
9... |
[
62,
21,
33,
37,
101,
85,
61,
37,
126,
101,
47,
62,
58,
34,
126,
8,
23,
125,
98,
122,
12,
111,
39,
87,
30,
94,
72,
97,
36,
103,
4,
68,
76,
100,
61,
6,
33,
125,
97,
43,
107,
70,
60,
124,
63,
127,
44,
108,
26,
90,
58,
... |
[
53,
117,
36,
11,
34,
17,
81,
38,
98,
35,
8,
99,
100,
18,
62,
38,
6,
72,
13,
27,
75,
47,
82,
21,
77,
42,
70,
91,
51,
102,
18,
28,
85,
5,
111,
12,
92,
34,
102,
46,
115,
7,
71,
59,
123,
82,
69,
98,
2,
25,
63,
1,
89,... |
[
21,
42,
106,
59,
123,
14,
8,
56,
27,
78,
91,
42,
85,
6,
70,
120,
63,
12,
34,
1,
76,
127,
72,
50,
5,
114,
52,
65,
69,
13,
60,
29,
77,
55,
106,
93,
25,
124,
98,
61,
56,
120,
125,
89,
62,
116,
126,
119,
34,
98,
37,
7,... |
[
17,
24,
81,
88,
12,
3,
47,
44,
76,
67,
38,
15,
111,
79,
46,
108,
57,
110,
102,
26,
90,
49,
113,
121,
50,
114,
37,
101,
60,
124,
33,
60,
97,
28,
92,
14,
78,
46,
124,
30,
94,
110,
5,
45,
109,
51,
32,
115,
69,
96,
26,
... |
[
2,
66,
19,
14,
83,
78,
0,
54,
64,
118,
12,
76,
9,
73,
45,
109,
20,
0,
64,
84,
28,
92,
29,
93,
4,
68,
28,
92,
42,
106,
7,
71,
53,
59,
30,
94,
123,
20,
84,
117,
32,
96,
18,
17,
81,
56,
15,
82,
14,
29,
79,
32,
11,
... |
[
19,
83,
7,
71,
58,
122,
36,
100,
33,
9,
97,
73,
44,
108,
14,
37,
52,
11,
62,
15,
101,
79,
47,
36,
3,
100,
111,
78,
58,
60,
124,
116,
28,
19,
42,
49,
106,
1,
113,
65,
62,
41,
83,
122,
92,
13,
105,
77,
126,
75,
17,
1... |
[
3,
3,
28,
92,
67,
62,
49,
5,
3,
113,
14,
78,
126,
58,
63,
127,
69,
67,
122,
30,
94,
67,
45,
109,
50,
11,
75,
3,
1,
38,
63,
0,
65,
114,
64,
42,
29,
27,
23,
106,
102,
127,
42,
14,
67,
91,
0,
87,
64,
62,
1,
78,
106,... |
[
29,
16,
21,
93,
12,
28,
46,
50,
21,
45,
85,
38,
109,
16,
92,
11,
75,
80,
80,
102,
85,
76,
110,
114,
60,
44,
46,
32,
18,
124,
25,
34,
89,
35,
98,
110,
26,
90,
33,
99,
9,
108,
13,
97,
31,
95,
82,
0,
59,
56,
47,
73,
... |
[
3,
67,
6,
59,
40,
48,
8,
31,
112,
72,
25,
104,
42,
0,
2,
66,
35,
89,
64,
70,
61,
106,
95,
59,
19,
8,
22,
41,
72,
60,
86,
7,
99,
58,
83,
62,
124,
21,
122,
71,
85,
105,
40,
126,
123,
43,
4,
4,
25,
68,
125,
68,
52,
... |
[
22,
86,
14,
78,
41,
54,
105,
118,
35,
99,
41,
58,
57,
121,
122,
38,
105,
102,
28,
4,
42,
33,
97,
106,
43,
48,
107,
92,
25,
37,
19,
112,
50,
101,
68,
114,
40,
44,
24,
104,
88,
49,
37,
83,
89,
101,
108,
113,
11,
13,
11... |
[
36,
100,
20,
84,
34,
4,
98,
49,
68,
113,
4,
68,
39,
6,
18,
53,
60,
47,
103,
55,
25,
124,
2,
36,
117,
11,
119,
37,
70,
43,
66,
49,
100,
1,
82,
75,
113,
89,
107,
38,
102,
111,
43,
47,
107,
65,
40,
101,
6,
1,
29,
26,
... |
End of preview. Expand in Data Studio
PPT Corpus: shuffle_dyck
Raw stage-1 pre-pretraining corpus from the ppt research framework, exported for information-theoretic analysis (e.g. m-local entropy) independent of this repo's training pipeline.
IMPORTANT: token IDs are NOT standard BPE tokens
Raw Shuffle-Dyck bracket-symbol IDs (integer i = opening bracket type i, i+k = its closing counterpart), generated by submodules/pre-pretraining/src/grammar.py::generate_shuff_dyck.
These input_ids are not decodable with the Pythia (or any other)
tokenizer. They are integers in [0, 128) with regimen-specific
meaning (see above) — treat this dataset as a corpus of symbol/event
sequences, not text.
Corpus stats
| Field | Value |
|---|---|
| Vocabulary size | 128 |
Chunk length (seq_len) |
2048 |
| Number of chunks | 50,000 |
| Total tokens | 102,400,000 |
| Generation seed | None |
Each row's input_ids is a fixed-length list of 2048 integers — the
exact chunk granularity fed to the model during stage-1 training (formed
by concatenating generated sequences and splitting into non-overlapping
2048-token blocks).
Generation config
{
"num_symbols": 64,
"n": 200000,
"target_length": 512,
"min_depth": 1,
"max_depth": 8,
"raw_ids": true,
"file_dir": "data/raw/shuffle_dyck",
"cache_dir": "data/tokenized/shuffle_dyck_raw",
"seq_len": 2048
}
Usage
from datasets import load_dataset
ds = load_dataset("sashaboguraev/ppt-shuffle_dyck-corpus")
example = ds["train"][0]["input_ids"] # list[int], length 2048
- Downloads last month
- 12