Datasets:
Tasks:
Image Classification
Modalities:
Image
Formats:
parquet
Sub-tasks:
multi-class-image-classification
Languages:
English
Size:
10K - 100K
License:
Search is not available for this dataset
image imagewidth (px) 28 28 | label class label 10
classes | embedding_foundation sequencelengths 768 768 | embedding_ft sequencelengths 768 768 | outlier_score_ft float64 0.58 1 | outlier_score_foundation float64 0.68 0.98 | nn_image dict |
|---|---|---|---|---|---|---|
55 | [
0.055755846202373505,
0.06896422058343887,
-0.28924471139907837,
-0.2733098566532135,
-0.24934417009353638,
0.08550799638032913,
0.19516059756278992,
-0.0436483770608902,
0.05387004092335701,
-0.47074973583221436,
-0.27012011408805847,
-0.0035196023527532816,
0.16980962455272675,
0.1882610... | [
0.2898097038269043,
-0.043863970786333084,
0.18290989100933075,
0.07755671441555023,
0.00436986843124032,
-0.45789414644241333,
-0.3313480317592621,
-0.12131891399621964,
0.05815001204609871,
0.05486920475959778,
0.32228466868400574,
0.2252117544412613,
0.4642280042171478,
0.39821451902389... | 0.958487 | 0.839489 | {
"bytes": [
137,
80,
78,
71,
13,
10,
26,
10,
0,
0,
0,
13,
73,
72,
68,
82,
0,
0,
0,
28,
0,
0,
0,
28,
8,
0,
0,
0,
0,
87,
102,
128,
72,
0,
0,
0,
231,
73,
68,
65,
84,
120,
156,
99,
96,
24,
74,
128,
17,
153,
99,
39,
44,
193,
168,
88,
196,
192,
48,
237,
218,
220,
95,
168,
202,
146,
46,
92,
120,
245,
233,
207,
223,
63,
127,
254,
252,
249,
243,
135,
15,
205,
140,
251,
127,
254,
252,
249,
3,
145,
252,
56,
151,
149,
129,
129,
129,
129,
129,
5,
77,
73,
199,
127,
6,
6,
134,
197,
55,
209,
109,
143,
249,
243,
231,
79,
128,
44,
138,
16,
19,
156,
245,
236,
59,
3,
67,
177,
29,
46,
135,
135,
236,
253,
243,
231,
79,
91,
58,
14,
89,
142,
89,
207,
255,
252,
61,
128,
75,
47,
195,
252,
151,
127,
223,
197,
227,
148,
157,
250,
247,
207,
91,
156,
146,
12,
255,
255,
126,
49,
199,
41,
121,
235,
207,
159,
5,
236,
184,
36,
51,
255,
252,
249,
3,
247,
44,
19,
154,
164,
43,
3,
195,
241,
79,
216,
245,
241,
121,
255,
253,
243,
61,
14,
83,
156,
95,
142,
147,
129,
193,
254,
207,
223,
63,
205,
88,
52,
85,
92,
81,
102,
104,
120,
240,
231,
239,
159,
8,
44,
26,
191,
252,
152,
246,
238,
247,
159,
63,
55,
52,
88,
49,
37,
89,
166,
254,
249,
243,
231,
207,
151,
10,
101,
172,
110,
49,
124,
253,
231,
79,
126,
40,
118,
119,
146,
14,
0,
7,
138,
97,
32,
174,
218,
197,
82,
0,
0,
0,
0,
73,
69,
78,
68,
174,
66,
96,
130
],
"path": null
} | |
00 | [
0.15941746532917023,
-0.10303866863250732,
-0.2600635588169098,
-0.2749791741371155,
0.029650665819644928,
-0.1461169719696045,
0.025942403823137283,
-0.015339868143200874,
-0.1518344134092331,
-0.36937853693962097,
-0.21947965025901794,
0.06841541081666946,
0.15400217473506927,
-0.1605581... | [
-0.3443199396133423,
0.38727006316185,
-0.37169310450553894,
-0.030840596184134483,
-0.36843952536582947,
0.4505344331264496,
0.049530379474163055,
-0.3375467360019684,
-0.32141101360321045,
0.09777911007404327,
-0.0616067573428154,
0.30867183208465576,
-0.37873291969299316,
-0.19419284164... | 0.996789 | 0.901946 | {
"bytes": [
137,
80,
78,
71,
13,
10,
26,
10,
0,
0,
0,
13,
73,
72,
68,
82,
0,
0,
0,
28,
0,
0,
0,
28,
8,
0,
0,
0,
0,
87,
102,
128,
72,
0,
0,
1,
8,
73,
68,
65,
84,
120,
156,
197,
208,
177,
47,
3,
97,
24,
199,
241,
159,
38,
114,
76,
134,
26,
12,
53,
184,
106,
210,
50,
152,
36,
24,
58,
73,
44,
172,
18,
151,
176,
248,
31,
76,
29,
76,
6,
3,
17,
137,
169,
54,
127,
128,
38,
141,
193,
63,
32,
145,
24,
164,
185,
224,
132,
27,
170,
131,
132,
164,
99,
57,
223,
176,
156,
187,
235,
235,
181,
242,
76,
207,
243,
126,
222,
223,
155,
39,
175,
244,
255,
85,
251,
60,
169,
252,
66,
110,
240,
1,
157,
41,
171,
21,
3,
0,
58,
211,
22,
155,
12,
224,
34,
127,
6,
71,
22,
188,
132,
198,
172,
42,
15,
68,
27,
63,
114,
187,
111,
180,
75,
146,
230,
224,
206,
196,
77,
96,
73,
146,
202,
208,
45,
245,
219,
132,
207,
249,
206,
176,
36,
57,
213,
208,
140,
94,
211,
46,
124,
247,
117,
162,
117,
73,
202,
197,
243,
106,
89,
199,
79,
233,
213,
220,
88,
38,
87,
104,
209,
28,
74,
38,
143,
119,
47,
131,
91,
80,
77,
167,
58,
93,
101,
159,
213,
235,
77,
98,
51,
139,
28,
100,
215,
185,
98,
59,
181,
16,
191,
111,
215,
30,
249,
184,
115,
106,
33,
207,
158,
21,
157,
249,
83,
94,
246,
141,
47,
232,
49,
46,
201,
25,
61,
132,
219,
98,
114,
58,
16,
227,
96,
107,
207,
29,
113,
151,
197,
253,
202,
163,
140,
106,
0,
64,
228,
123,
166,
72,
90,
104,
2,
248,
107,
22,
250,
227,
250,
2,
82,
235,
113,
77,
166,
140,
102,
49,
0,
0,
0,
0,
73,
69,
78,
68,
174,
66,
96,
130
],
"path": null
} | |
44 | [
-0.12161394953727722,
0.023048682138323784,
-0.19237416982650757,
-0.1365087777376175,
-0.10920078307390213,
0.0968828797340393,
0.026648705825209618,
-0.1655183732509613,
-0.1930493712425232,
-0.47365590929985046,
-0.36887043714523315,
-0.304238498210907,
0.10931092500686646,
0.1421499699... | [
0.04633236676454544,
-0.38148313760757446,
-0.16074497997760773,
-0.10137465596199036,
0.3197562098503113,
0.035613108426332474,
0.1351068615913391,
0.08741147071123123,
-0.3504461944103241,
-0.11091212183237076,
0.09539677202701569,
-0.3301088213920593,
-0.3589787483215332,
-0.39909446239... | 0.994356 | 0.884935 | {
"bytes": [
137,
80,
78,
71,
13,
10,
26,
10,
0,
0,
0,
13,
73,
72,
68,
82,
0,
0,
0,
28,
0,
0,
0,
28,
8,
0,
0,
0,
0,
87,
102,
128,
72,
0,
0,
0,
248,
73,
68,
65,
84,
120,
156,
99,
96,
24,
36,
128,
213,
124,
191,
15,
78,
73,
253,
63,
127,
58,
25,
24,
24,
24,
152,
176,
73,
202,
51,
48,
172,
197,
169,
115,
225,
159,
13,
44,
200,
252,
189,
155,
16,
108,
175,
47,
127,
114,
24,
24,
24,
24,
24,
32,
42,
88,
131,
120,
204,
16,
146,
222,
156,
12,
135,
80,
93,
80,
10,
231,
136,
253,
253,
243,
65,
21,
33,
199,
62,
251,
207,
99,
1,
56,
111,
210,
223,
63,
219,
145,
52,
150,
253,
249,
227,
132,
224,
61,
251,
251,
39,
146,
1,
97,
167,
8,
3,
195,
107,
152,
84,
168,
135,
8,
195,
235,
51,
72,
146,
140,
140,
12,
23,
166,
127,
255,
207,
192,
144,
201,
243,
143,
129,
129,
129,
233,
238,
109,
36,
99,
139,
222,
252,
249,
243,
231,
239,
159,
63,
127,
254,
252,
60,
183,
251,
197,
159,
159,
222,
80,
113,
70,
6,
6,
6,
6,
6,
157,
28,
11,
157,
61,
255,
25,
174,
237,
248,
117,
208,
109,
27,
195,
43,
41,
6,
84,
32,
168,
11,
101,
184,
253,
249,
211,
14,
19,
132,
5,
211,
251,
247,
80,
70,
6,
3,
195,
91,
6,
28,
64,
255,
239,
223,
207,
112,
14,
11,
154,
164,
231,
127,
134,
38,
92,
26,
25,
22,
254,
249,
35,
6,
231,
96,
141,
79,
92,
146,
103,
24,
42,
222,
224,
83,
79,
115,
0,
0,
105,
151,
79,
70,
113,
114,
158,
142,
0,
0,
0,
0,
73,
69,
78,
68,
174,
66,
96,
130
],
"path": null
} | |
11 | [
0.2116268277168274,
0.07443660497665405,
-0.2436290979385376,
-0.2518989145755768,
-0.29724740982055664,
-0.166959747672081,
0.21671436727046967,
-0.06508763879537582,
0.3428446352481842,
-0.5339957475662231,
-0.23810119926929474,
-0.18990185856819153,
0.2949427366256714,
0.094982057809829... | [
0.24189753830432892,
0.4566895067691803,
0.3326575458049774,
-0.3869461715221405,
-0.28423240780830383,
0.0849863737821579,
0.12257940322160721,
-0.4565140902996063,
-0.11553803086280823,
-0.19755426049232483,
-0.34109577536582947,
-0.4502791166305542,
0.3951323330402374,
-0.23077152669429... | 0.998491 | 0.94702 | {
"bytes": [
137,
80,
78,
71,
13,
10,
26,
10,
0,
0,
0,
13,
73,
72,
68,
82,
0,
0,
0,
28,
0,
0,
0,
28,
8,
0,
0,
0,
0,
87,
102,
128,
72,
0,
0,
0,
180,
73,
68,
65,
84,
120,
156,
99,
96,
24,
252,
32,
228,
194,
191,
213,
184,
228,
130,
46,
252,
253,
155,
1,
97,
50,
97,
72,
198,
233,
226,
54,
52,
253,
199,
223,
191,
139,
56,
176,
203,
101,
124,
251,
251,
174,
75,
28,
187,
92,
230,
215,
191,
127,
175,
193,
121,
44,
40,
114,
146,
218,
28,
12,
71,
207,
96,
215,
39,
121,
248,
239,
223,
191,
126,
56,
28,
227,
255,
247,
239,
223,
117,
2,
216,
229,
156,
174,
254,
253,
123,
95,
25,
135,
198,
109,
127,
255,
190,
75,
18,
194,
46,
183,
224,
243,
223,
191,
149,
56,
244,
25,
191,
251,
251,
247,
134,
11,
118,
57,
193,
185,
159,
255,
94,
141,
17,
198,
42,
199,
191,
255,
239,
223,
191,
229,
216,
229,
24,
10,
254,
254,
253,
123,
82,
16,
77,
16,
22,
43,
199,
190,
51,
48,
92,
125,
143,
195,
57,
89,
127,
254,
254,
85,
195,
33,
199,
192,
240,
2,
159,
100,
230,
223,
110,
28,
206,
161,
58,
0,
0,
26,
239,
69,
233,
13,
77,
195,
171,
0,
0,
0,
0,
73,
69,
78,
68,
174,
66,
96,
130
],
"path": null
} | |
99 | [
0.2062450647354126,
0.12278524786233902,
-0.1381160169839859,
-0.17313197255134583,
-0.08984404802322388,
0.020129164680838585,
0.07127396762371063,
-0.013791313394904137,
0.08637185394763947,
-0.3930748403072357,
-0.12134677171707153,
-0.06922566890716553,
0.4477277994155884,
0.1099054515... | [
-0.14900259673595428,
-0.27072837948799133,
0.2449478656053543,
0.5078010559082031,
-0.48765790462493896,
0.04053984582424164,
-0.5196414589881897,
-0.21596640348434448,
0.4073738157749176,
-0.22233977913856506,
-0.18392054736614227,
0.13726510107517242,
-0.11035670340061188,
0.13771769404... | 0.992551 | 0.923266 | {
"bytes": [
137,
80,
78,
71,
13,
10,
26,
10,
0,
0,
0,
13,
73,
72,
68,
82,
0,
0,
0,
28,
0,
0,
0,
28,
8,
0,
0,
0,
0,
87,
102,
128,
72,
0,
0,
0,
240,
73,
68,
65,
84,
120,
156,
99,
96,
24,
54,
128,
17,
133,
151,
40,
202,
192,
224,
124,
228,
231,
227,
229,
152,
10,
103,
191,
254,
253,
239,
223,
191,
127,
255,
254,
253,
114,
131,
8,
176,
192,
100,
184,
123,
254,
36,
51,
48,
156,
188,
211,
166,
186,
138,
141,
133,
21,
85,
151,
231,
169,
127,
255,
158,
29,
15,
23,
100,
96,
187,
255,
239,
223,
121,
33,
20,
57,
238,
83,
255,
254,
222,
49,
98,
96,
96,
96,
72,
251,
247,
239,
223,
90,
84,
141,
190,
255,
254,
189,
132,
176,
38,
254,
251,
119,
86,
12,
42,
202,
4,
161,
220,
24,
24,
234,
224,
42,
239,
189,
98,
64,
113,
16,
59,
3,
67,
47,
199,
133,
151,
238,
12,
81,
198,
12,
12,
71,
209,
124,
225,
251,
239,
223,
191,
127,
159,
223,
66,
124,
162,
132,
38,
201,
239,
183,
243,
241,
191,
127,
255,
222,
236,
124,
246,
239,
223,
35,
73,
204,
32,
208,
116,
113,
113,
49,
98,
88,
246,
239,
223,
116,
76,
57,
40,
136,
70,
150,
100,
66,
147,
244,
102,
96,
216,
136,
83,
146,
1,
217,
177,
88,
36,
25,
112,
73,
74,
27,
224,
145,
124,
186,
142,
129,
1,
221,
155,
8,
192,
185,
241,
223,
3,
125,
156,
178,
214,
95,
254,
173,
198,
233,
160,
163,
174,
191,
240,
185,
17,
1,
0,
134,
117,
90,
27,
22,
242,
114,
34,
0,
0,
0,
0,
73,
69,
78,
68,
174,
66,
96,
130
],
"path": null
} | |
22 | [
0.08376070111989975,
0.0034862349275499582,
-0.21620474755764008,
-0.14218387007713318,
-0.2540224492549896,
-0.01811244711279869,
0.08731447160243988,
0.01712123677134514,
-0.00037671951577067375,
-0.3851156234741211,
-0.19619892537593842,
-0.12013868987560272,
0.3460136950016022,
-0.1073... | [
-0.3362293541431427,
-0.14586883783340454,
-0.09560512006282806,
0.15935714542865753,
-0.2983415722846985,
-0.10146545618772507,
0.6791165471076965,
-0.16078004240989685,
0.10026566684246063,
0.33570241928100586,
-0.07739301770925522,
-0.21037684381008148,
0.40856024622917175,
0.2520348429... | 0.995175 | 0.887807 | {
"bytes": [
137,
80,
78,
71,
13,
10,
26,
10,
0,
0,
0,
13,
73,
72,
68,
82,
0,
0,
0,
28,
0,
0,
0,
28,
8,
0,
0,
0,
0,
87,
102,
128,
72,
0,
0,
1,
9,
73,
68,
65,
84,
120,
156,
213,
207,
175,
75,
67,
81,
24,
198,
241,
239,
116,
176,
59,
29,
130,
32,
140,
91,
174,
26,
197,
48,
193,
174,
65,
17,
131,
160,
69,
49,
14,
5,
103,
17,
252,
11,
20,
52,
88,
196,
184,
98,
178,
174,
106,
156,
26,
54,
48,
92,
156,
98,
18,
54,
147,
98,
152,
97,
186,
137,
206,
43,
60,
24,
142,
92,
127,
236,
44,
139,
79,
121,
225,
124,
56,
188,
239,
3,
127,
159,
142,
120,
196,
89,
232,
137,
88,
45,
150,
83,
254,
85,
26,
177,
89,
178,
36,
61,
30,
31,
233,
208,
98,
94,
73,
65,
206,
97,
210,
130,
157,
91,
117,
5,
43,
192,
140,
5,
183,
165,
32,
3,
144,
213,
158,
121,
137,
126,
225,
83,
245,
124,
167,
8,
48,
193,
89,
235,
202,
94,
51,
250,
159,
149,
248,
236,
245,
13,
107,
102,
44,
118,
221,
4,
182,
42,
0,
221,
15,
90,
110,
107,
23,
242,
29,
90,
14,
114,
135,
166,
1,
55,
213,
152,
107,
254,
254,
49,
154,
173,
201,
228,
54,
241,
83,
146,
155,
247,
205,
242,
218,
146,
187,
42,
169,
170,
43,
143,
177,
225,
208,
82,
215,
186,
219,
24,
240,
102,
79,
222,
85,
158,
26,
44,
170,
158,
121,
171,
132,
88,
144,
242,
251,
167,
47,
82,
99,
183,
15,
162,
7,
146,
252,
16,
125,
179,
172,
176,
238,
154,
35,
199,
211,
233,
56,
64,
4,
32,
54,
15,
84,
46,
3,
181,
235,
247,
47,
242,
1,
83,
67,
108,
242,
225,
221,
148,
155,
0,
0,
0,
0,
73,
69,
78,
68,
174,
66,
96,
130
],
"path": null
} | |
11 | [0.3397919535636902,0.07246905565261841,-0.22857190668582916,-0.1418822854757309,-0.233572855591774,(...TRUNCATED) | [0.22328144311904907,0.41259539127349854,0.33726242184638977,-0.3560121953487396,-0.3558163642883301(...TRUNCATED) | 0.997133 | 0.937616 | {"bytes":"iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAAAeUlEQVR4nMXRwQ2AIAwFUNo4Bs4hi7CIi7iGOgeaOF(...TRUNCATED) | |
33 | [0.173781618475914,0.03911104053258896,-0.1764281690120697,-0.17309260368347168,-0.2629533112049103,(...TRUNCATED) | [-0.33131131529808044,-0.4200515151023865,0.32494524121284485,-0.06822431832551956,0.250309139490127(...TRUNCATED) | 0.997522 | 0.941296 | {"bytes":"iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAABEElEQVR4nGNgGGjAGzT9379//6YJYci4TJz47y8EhK(...TRUNCATED) | |
11 | [0.29078754782676697,0.0492263026535511,-0.22514574229717255,-0.15281744301319122,-0.240966692566871(...TRUNCATED) | [0.2694103419780731,0.44531911611557007,0.3512427508831024,-0.41870051622390747,-0.33192893862724304(...TRUNCATED) | 0.998058 | 0.931529 | {"bytes":"iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAAAcElEQVR4nGNgGMwg+pslCp8JmePKZodbkoHBELexZX(...TRUNCATED) | |
44 | [0.11276884377002716,-0.04172897711396217,-0.1634320467710495,-0.08415620028972626,-0.16161626577377(...TRUNCATED) | [0.10163689404726028,-0.40715184807777405,-0.18369340896606445,-0.12446718662977219,0.27636379003524(...TRUNCATED) | 0.99506 | 0.907182 | {"bytes":"iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAAA70lEQVR4nM3PP0tCcRjF8QNZYGkh1xxEaBGcEpx6CQ(...TRUNCATED) |
End of preview. Expand in Data Studio
Dataset Card for "mnist-outlier"
📚 This dataset is an enriched version of the MNIST Dataset.
The workflow is described in the medium article: Changes of Embeddings during Fine-Tuning of Transformers.
Explore the Dataset
The open source data curation tool Renumics Spotlight allows you to explorer this dataset. You can find a Hugging Face Space running Spotlight with this dataset here: https://huggingface.co/spaces/renumics/mnist-outlier.
Or you can explorer it locally:
!pip install renumics-spotlight datasets
from renumics import spotlight
import datasets
ds = datasets.load_dataset("renumics/mnist-outlier", split="train")
df = ds.rename_columns({"label":"labels"}).to_pandas()
df["label_str"] = df["labels"].apply(lambda x: ds.features["label"].int2str(x))
dtypes = {
"nn_image": spotlight.Image,
"image": spotlight.Image,
"embedding_ft": spotlight.Embedding,
"embedding_foundation": spotlight.Embedding,
}
spotlight.show(
df,
dtype=dtypes,
layout="https://spotlight.renumics.com/resources/layout_pre_post_ft.json",
)
- Downloads last month
- 18