Hugging Face's logo

jobs-git
/
switch-c-2048

Text2Text Generation
PyTorch
English
switch_transformers
Model card Files Community
jobs-git ybelkada commited on
Commit
46b3959
·
verified ·
0 Parent(s):

Duplicate from google/switch-c-2048

Browse files

Co-authored-by: Younes Belkada <[email protected]>

This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +399 -0
  2. README.md +240 -0
  3. config.json +39 -0
  4. pytorch_model-00001-of-00364.bin +3 -0
  5. pytorch_model-00002-of-00364.bin +3 -0
  6. pytorch_model-00003-of-00364.bin +3 -0
  7. pytorch_model-00004-of-00364.bin +3 -0
  8. pytorch_model-00005-of-00364.bin +3 -0
  9. pytorch_model-00006-of-00364.bin +3 -0
  10. pytorch_model-00007-of-00364.bin +3 -0
  11. pytorch_model-00008-of-00364.bin +3 -0
  12. pytorch_model-00009-of-00364.bin +3 -0
  13. pytorch_model-00010-of-00364.bin +3 -0
  14. pytorch_model-00011-of-00364.bin +3 -0
  15. pytorch_model-00012-of-00364.bin +3 -0
  16. pytorch_model-00013-of-00364.bin +3 -0
  17. pytorch_model-00014-of-00364.bin +3 -0
  18. pytorch_model-00015-of-00364.bin +3 -0
  19. pytorch_model-00016-of-00364.bin +3 -0
  20. pytorch_model-00017-of-00364.bin +3 -0
  21. pytorch_model-00018-of-00364.bin +3 -0
  22. pytorch_model-00019-of-00364.bin +3 -0
  23. pytorch_model-00020-of-00364.bin +3 -0
  24. pytorch_model-00021-of-00364.bin +3 -0
  25. pytorch_model-00022-of-00364.bin +3 -0
  26. pytorch_model-00023-of-00364.bin +3 -0
  27. pytorch_model-00024-of-00364.bin +3 -0
  28. pytorch_model-00025-of-00364.bin +3 -0
  29. pytorch_model-00026-of-00364.bin +3 -0
  30. pytorch_model-00027-of-00364.bin +3 -0
  31. pytorch_model-00028-of-00364.bin +3 -0
  32. pytorch_model-00029-of-00364.bin +3 -0
  33. pytorch_model-00030-of-00364.bin +3 -0
  34. pytorch_model-00031-of-00364.bin +3 -0
  35. pytorch_model-00032-of-00364.bin +3 -0
  36. pytorch_model-00033-of-00364.bin +3 -0
  37. pytorch_model-00034-of-00364.bin +3 -0
  38. pytorch_model-00035-of-00364.bin +3 -0
  39. pytorch_model-00036-of-00364.bin +3 -0
  40. pytorch_model-00037-of-00364.bin +3 -0
  41. pytorch_model-00038-of-00364.bin +3 -0
  42. pytorch_model-00039-of-00364.bin +3 -0
  43. pytorch_model-00040-of-00364.bin +3 -0
  44. pytorch_model-00041-of-00364.bin +3 -0
  45. pytorch_model-00042-of-00364.bin +3 -0
  46. pytorch_model-00043-of-00364.bin +3 -0
  47. pytorch_model-00044-of-00364.bin +3 -0
  48. pytorch_model-00045-of-00364.bin +3 -0
  49. pytorch_model-00046-of-00364.bin +3 -0
  50. pytorch_model-00047-of-00364.bin +3 -0
.gitattributes ADDED
@@ -0,0 +1,399 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tflite filter=lfs diff=lfs merge=lfs -text
29
+ *.tgz filter=lfs diff=lfs merge=lfs -text
30
+ *.wasm filter=lfs diff=lfs merge=lfs -text
31
+ *.xz filter=lfs diff=lfs merge=lfs -text
32
+ *.zip filter=lfs diff=lfs merge=lfs -text
33
+ *.zst filter=lfs diff=lfs merge=lfs -text
34
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ pytorch_model-00299-of-00364.bin filter=lfs diff=lfs merge=lfs -text
36
+ pytorch_model-00335-of-00364.bin filter=lfs diff=lfs merge=lfs -text
37
+ pytorch_model-00339-of-00364.bin filter=lfs diff=lfs merge=lfs -text
38
+ pytorch_model-00003-of-00364.bin filter=lfs diff=lfs merge=lfs -text
39
+ pytorch_model-00067-of-00364.bin filter=lfs diff=lfs merge=lfs -text
40
+ pytorch_model-00084-of-00364.bin filter=lfs diff=lfs merge=lfs -text
41
+ pytorch_model-00163-of-00364.bin filter=lfs diff=lfs merge=lfs -text
42
+ pytorch_model-00230-of-00364.bin filter=lfs diff=lfs merge=lfs -text
43
+ pytorch_model-00293-of-00364.bin filter=lfs diff=lfs merge=lfs -text
44
+ pytorch_model-00006-of-00364.bin filter=lfs diff=lfs merge=lfs -text
45
+ pytorch_model-00033-of-00364.bin filter=lfs diff=lfs merge=lfs -text
46
+ pytorch_model-00120-of-00364.bin filter=lfs diff=lfs merge=lfs -text
47
+ pytorch_model-00173-of-00364.bin filter=lfs diff=lfs merge=lfs -text
48
+ pytorch_model-00122-of-00364.bin filter=lfs diff=lfs merge=lfs -text
49
+ pytorch_model-00223-of-00364.bin filter=lfs diff=lfs merge=lfs -text
50
+ pytorch_model-00323-of-00364.bin filter=lfs diff=lfs merge=lfs -text
51
+ pytorch_model-00331-of-00364.bin filter=lfs diff=lfs merge=lfs -text
52
+ pytorch_model-00321-of-00364.bin filter=lfs diff=lfs merge=lfs -text
53
+ pytorch_model-00329-of-00364.bin filter=lfs diff=lfs merge=lfs -text
54
+ pytorch_model-00070-of-00364.bin filter=lfs diff=lfs merge=lfs -text
55
+ pytorch_model-00156-of-00364.bin filter=lfs diff=lfs merge=lfs -text
56
+ pytorch_model-00257-of-00364.bin filter=lfs diff=lfs merge=lfs -text
57
+ pytorch_model-00305-of-00364.bin filter=lfs diff=lfs merge=lfs -text
58
+ pytorch_model-00081-of-00364.bin filter=lfs diff=lfs merge=lfs -text
59
+ pytorch_model-00222-of-00364.bin filter=lfs diff=lfs merge=lfs -text
60
+ pytorch_model-00359-of-00364.bin filter=lfs diff=lfs merge=lfs -text
61
+ pytorch_model-00057-of-00364.bin filter=lfs diff=lfs merge=lfs -text
62
+ pytorch_model-00105-of-00364.bin filter=lfs diff=lfs merge=lfs -text
63
+ pytorch_model-00113-of-00364.bin filter=lfs diff=lfs merge=lfs -text
64
+ pytorch_model-00151-of-00364.bin filter=lfs diff=lfs merge=lfs -text
65
+ pytorch_model-00008-of-00364.bin filter=lfs diff=lfs merge=lfs -text
66
+ pytorch_model-00011-of-00364.bin filter=lfs diff=lfs merge=lfs -text
67
+ pytorch_model-00051-of-00364.bin filter=lfs diff=lfs merge=lfs -text
68
+ pytorch_model-00056-of-00364.bin filter=lfs diff=lfs merge=lfs -text
69
+ pytorch_model-00158-of-00364.bin filter=lfs diff=lfs merge=lfs -text
70
+ pytorch_model-00185-of-00364.bin filter=lfs diff=lfs merge=lfs -text
71
+ pytorch_model-00315-of-00364.bin filter=lfs diff=lfs merge=lfs -text
72
+ pytorch_model-00336-of-00364.bin filter=lfs diff=lfs merge=lfs -text
73
+ pytorch_model-00183-of-00364.bin filter=lfs diff=lfs merge=lfs -text
74
+ pytorch_model-00244-of-00364.bin filter=lfs diff=lfs merge=lfs -text
75
+ pytorch_model-00255-of-00364.bin filter=lfs diff=lfs merge=lfs -text
76
+ pytorch_model-00260-of-00364.bin filter=lfs diff=lfs merge=lfs -text
77
+ pytorch_model-00270-of-00364.bin filter=lfs diff=lfs merge=lfs -text
78
+ pytorch_model-00313-of-00364.bin filter=lfs diff=lfs merge=lfs -text
79
+ pytorch_model-00010-of-00364.bin filter=lfs diff=lfs merge=lfs -text
80
+ pytorch_model-00017-of-00364.bin filter=lfs diff=lfs merge=lfs -text
81
+ pytorch_model-00128-of-00364.bin filter=lfs diff=lfs merge=lfs -text
82
+ pytorch_model-00133-of-00364.bin filter=lfs diff=lfs merge=lfs -text
83
+ pytorch_model-00210-of-00364.bin filter=lfs diff=lfs merge=lfs -text
84
+ pytorch_model-00247-of-00364.bin filter=lfs diff=lfs merge=lfs -text
85
+ pytorch_model-00324-of-00364.bin filter=lfs diff=lfs merge=lfs -text
86
+ pytorch_model-00037-of-00364.bin filter=lfs diff=lfs merge=lfs -text
87
+ pytorch_model-00058-of-00364.bin filter=lfs diff=lfs merge=lfs -text
88
+ pytorch_model-00125-of-00364.bin filter=lfs diff=lfs merge=lfs -text
89
+ pytorch_model-00203-of-00364.bin filter=lfs diff=lfs merge=lfs -text
90
+ pytorch_model-00019-of-00364.bin filter=lfs diff=lfs merge=lfs -text
91
+ pytorch_model-00199-of-00364.bin filter=lfs diff=lfs merge=lfs -text
92
+ pytorch_model-00224-of-00364.bin filter=lfs diff=lfs merge=lfs -text
93
+ pytorch_model-00283-of-00364.bin filter=lfs diff=lfs merge=lfs -text
94
+ pytorch_model-00317-of-00364.bin filter=lfs diff=lfs merge=lfs -text
95
+ pytorch_model-00358-of-00364.bin filter=lfs diff=lfs merge=lfs -text
96
+ pytorch_model-00080-of-00364.bin filter=lfs diff=lfs merge=lfs -text
97
+ pytorch_model-00136-of-00364.bin filter=lfs diff=lfs merge=lfs -text
98
+ pytorch_model-00169-of-00364.bin filter=lfs diff=lfs merge=lfs -text
99
+ pytorch_model-00267-of-00364.bin filter=lfs diff=lfs merge=lfs -text
100
+ pytorch_model-00314-of-00364.bin filter=lfs diff=lfs merge=lfs -text
101
+ pytorch_model-00050-of-00364.bin filter=lfs diff=lfs merge=lfs -text
102
+ pytorch_model-00160-of-00364.bin filter=lfs diff=lfs merge=lfs -text
103
+ pytorch_model-00197-of-00364.bin filter=lfs diff=lfs merge=lfs -text
104
+ pytorch_model-00307-of-00364.bin filter=lfs diff=lfs merge=lfs -text
105
+ pytorch_model-00234-of-00364.bin filter=lfs diff=lfs merge=lfs -text
106
+ pytorch_model-00242-of-00364.bin filter=lfs diff=lfs merge=lfs -text
107
+ pytorch_model-00243-of-00364.bin filter=lfs diff=lfs merge=lfs -text
108
+ pytorch_model-00275-of-00364.bin filter=lfs diff=lfs merge=lfs -text
109
+ pytorch_model-00063-of-00364.bin filter=lfs diff=lfs merge=lfs -text
110
+ pytorch_model-00170-of-00364.bin filter=lfs diff=lfs merge=lfs -text
111
+ pytorch_model-00200-of-00364.bin filter=lfs diff=lfs merge=lfs -text
112
+ pytorch_model-00219-of-00364.bin filter=lfs diff=lfs merge=lfs -text
113
+ pytorch_model-00297-of-00364.bin filter=lfs diff=lfs merge=lfs -text
114
+ pytorch_model-00330-of-00364.bin filter=lfs diff=lfs merge=lfs -text
115
+ pytorch_model-00360-of-00364.bin filter=lfs diff=lfs merge=lfs -text
116
+ pytorch_model-00139-of-00364.bin filter=lfs diff=lfs merge=lfs -text
117
+ pytorch_model-00174-of-00364.bin filter=lfs diff=lfs merge=lfs -text
118
+ pytorch_model-00194-of-00364.bin filter=lfs diff=lfs merge=lfs -text
119
+ pytorch_model-00268-of-00364.bin filter=lfs diff=lfs merge=lfs -text
120
+ pytorch_model-00146-of-00364.bin filter=lfs diff=lfs merge=lfs -text
121
+ pytorch_model-00214-of-00364.bin filter=lfs diff=lfs merge=lfs -text
122
+ pytorch_model-00089-of-00364.bin filter=lfs diff=lfs merge=lfs -text
123
+ pytorch_model-00132-of-00364.bin filter=lfs diff=lfs merge=lfs -text
124
+ pytorch_model-00300-of-00364.bin filter=lfs diff=lfs merge=lfs -text
125
+ pytorch_model-00310-of-00364.bin filter=lfs diff=lfs merge=lfs -text
126
+ pytorch_model-00016-of-00364.bin filter=lfs diff=lfs merge=lfs -text
127
+ pytorch_model-00086-of-00364.bin filter=lfs diff=lfs merge=lfs -text
128
+ pytorch_model-00111-of-00364.bin filter=lfs diff=lfs merge=lfs -text
129
+ pytorch_model-00295-of-00364.bin filter=lfs diff=lfs merge=lfs -text
130
+ pytorch_model-00276-of-00364.bin filter=lfs diff=lfs merge=lfs -text
131
+ pytorch_model-00353-of-00364.bin filter=lfs diff=lfs merge=lfs -text
132
+ pytorch_model-00181-of-00364.bin filter=lfs diff=lfs merge=lfs -text
133
+ pytorch_model-00208-of-00364.bin filter=lfs diff=lfs merge=lfs -text
134
+ pytorch_model-00220-of-00364.bin filter=lfs diff=lfs merge=lfs -text
135
+ pytorch_model-00233-of-00364.bin filter=lfs diff=lfs merge=lfs -text
136
+ pytorch_model-00106-of-00364.bin filter=lfs diff=lfs merge=lfs -text
137
+ pytorch_model-00364-of-00364.bin filter=lfs diff=lfs merge=lfs -text
138
+ pytorch_model-00083-of-00364.bin filter=lfs diff=lfs merge=lfs -text
139
+ pytorch_model-00127-of-00364.bin filter=lfs diff=lfs merge=lfs -text
140
+ pytorch_model-00066-of-00364.bin filter=lfs diff=lfs merge=lfs -text
141
+ pytorch_model-00091-of-00364.bin filter=lfs diff=lfs merge=lfs -text
142
+ pytorch_model-00123-of-00364.bin filter=lfs diff=lfs merge=lfs -text
143
+ pytorch_model-00124-of-00364.bin filter=lfs diff=lfs merge=lfs -text
144
+ pytorch_model-00182-of-00364.bin filter=lfs diff=lfs merge=lfs -text
145
+ pytorch_model-00217-of-00364.bin filter=lfs diff=lfs merge=lfs -text
146
+ pytorch_model-00249-of-00364.bin filter=lfs diff=lfs merge=lfs -text
147
+ pytorch_model-00286-of-00364.bin filter=lfs diff=lfs merge=lfs -text
148
+ pytorch_model-00009-of-00364.bin filter=lfs diff=lfs merge=lfs -text
149
+ pytorch_model-00036-of-00364.bin filter=lfs diff=lfs merge=lfs -text
150
+ pytorch_model-00055-of-00364.bin filter=lfs diff=lfs merge=lfs -text
151
+ pytorch_model-00119-of-00364.bin filter=lfs diff=lfs merge=lfs -text
152
+ pytorch_model-00345-of-00364.bin filter=lfs diff=lfs merge=lfs -text
153
+ pytorch_model-00287-of-00364.bin filter=lfs diff=lfs merge=lfs -text
154
+ pytorch_model-00005-of-00364.bin filter=lfs diff=lfs merge=lfs -text
155
+ pytorch_model-00059-of-00364.bin filter=lfs diff=lfs merge=lfs -text
156
+ pytorch_model-00069-of-00364.bin filter=lfs diff=lfs merge=lfs -text
157
+ pytorch_model-00110-of-00364.bin filter=lfs diff=lfs merge=lfs -text
158
+ pytorch_model-00172-of-00364.bin filter=lfs diff=lfs merge=lfs -text
159
+ pytorch_model-00216-of-00364.bin filter=lfs diff=lfs merge=lfs -text
160
+ pytorch_model-00228-of-00364.bin filter=lfs diff=lfs merge=lfs -text
161
+ pytorch_model-00236-of-00364.bin filter=lfs diff=lfs merge=lfs -text
162
+ pytorch_model-00026-of-00364.bin filter=lfs diff=lfs merge=lfs -text
163
+ pytorch_model-00071-of-00364.bin filter=lfs diff=lfs merge=lfs -text
164
+ pytorch_model-00104-of-00364.bin filter=lfs diff=lfs merge=lfs -text
165
+ pytorch_model-00150-of-00364.bin filter=lfs diff=lfs merge=lfs -text
166
+ pytorch_model-00271-of-00364.bin filter=lfs diff=lfs merge=lfs -text
167
+ pytorch_model-00348-of-00364.bin filter=lfs diff=lfs merge=lfs -text
168
+ pytorch_model-00107-of-00364.bin filter=lfs diff=lfs merge=lfs -text
169
+ pytorch_model-00108-of-00364.bin filter=lfs diff=lfs merge=lfs -text
170
+ pytorch_model-00227-of-00364.bin filter=lfs diff=lfs merge=lfs -text
171
+ pytorch_model-00237-of-00364.bin filter=lfs diff=lfs merge=lfs -text
172
+ pytorch_model-00012-of-00364.bin filter=lfs diff=lfs merge=lfs -text
173
+ pytorch_model-00015-of-00364.bin filter=lfs diff=lfs merge=lfs -text
174
+ pytorch_model-00043-of-00364.bin filter=lfs diff=lfs merge=lfs -text
175
+ pytorch_model-00088-of-00364.bin filter=lfs diff=lfs merge=lfs -text
176
+ pytorch_model-00292-of-00364.bin filter=lfs diff=lfs merge=lfs -text
177
+ pytorch_model-00347-of-00364.bin filter=lfs diff=lfs merge=lfs -text
178
+ pytorch_model-00274-of-00364.bin filter=lfs diff=lfs merge=lfs -text
179
+ pytorch_model-00018-of-00364.bin filter=lfs diff=lfs merge=lfs -text
180
+ pytorch_model-00072-of-00364.bin filter=lfs diff=lfs merge=lfs -text
181
+ pytorch_model-00101-of-00364.bin filter=lfs diff=lfs merge=lfs -text
182
+ pytorch_model-00238-of-00364.bin filter=lfs diff=lfs merge=lfs -text
183
+ pytorch_model-00129-of-00364.bin filter=lfs diff=lfs merge=lfs -text
184
+ pytorch_model-00140-of-00364.bin filter=lfs diff=lfs merge=lfs -text
185
+ pytorch_model-00161-of-00364.bin filter=lfs diff=lfs merge=lfs -text
186
+ pytorch_model-00316-of-00364.bin filter=lfs diff=lfs merge=lfs -text
187
+ pytorch_model-00014-of-00364.bin filter=lfs diff=lfs merge=lfs -text
188
+ pytorch_model-00025-of-00364.bin filter=lfs diff=lfs merge=lfs -text
189
+ pytorch_model-00053-of-00364.bin filter=lfs diff=lfs merge=lfs -text
190
+ pytorch_model-00075-of-00364.bin filter=lfs diff=lfs merge=lfs -text
191
+ pytorch_model-00320-of-00364.bin filter=lfs diff=lfs merge=lfs -text
192
+ pytorch_model-00354-of-00364.bin filter=lfs diff=lfs merge=lfs -text
193
+ pytorch_model-00116-of-00364.bin filter=lfs diff=lfs merge=lfs -text
194
+ pytorch_model-00117-of-00364.bin filter=lfs diff=lfs merge=lfs -text
195
+ pytorch_model-00328-of-00364.bin filter=lfs diff=lfs merge=lfs -text
196
+ pytorch_model-00342-of-00364.bin filter=lfs diff=lfs merge=lfs -text
197
+ pytorch_model-00261-of-00364.bin filter=lfs diff=lfs merge=lfs -text
198
+ pytorch_model-00046-of-00364.bin filter=lfs diff=lfs merge=lfs -text
199
+ pytorch_model-00145-of-00364.bin filter=lfs diff=lfs merge=lfs -text
200
+ pytorch_model-00188-of-00364.bin filter=lfs diff=lfs merge=lfs -text
201
+ pytorch_model-00212-of-00364.bin filter=lfs diff=lfs merge=lfs -text
202
+ pytorch_model-00147-of-00364.bin filter=lfs diff=lfs merge=lfs -text
203
+ pytorch_model-00285-of-00364.bin filter=lfs diff=lfs merge=lfs -text
204
+ pytorch_model-00319-of-00364.bin filter=lfs diff=lfs merge=lfs -text
205
+ pytorch_model-00337-of-00364.bin filter=lfs diff=lfs merge=lfs -text
206
+ pytorch_model-00027-of-00364.bin filter=lfs diff=lfs merge=lfs -text
207
+ pytorch_model-00034-of-00364.bin filter=lfs diff=lfs merge=lfs -text
208
+ pytorch_model-00103-of-00364.bin filter=lfs diff=lfs merge=lfs -text
209
+ pytorch_model-00115-of-00364.bin filter=lfs diff=lfs merge=lfs -text
210
+ pytorch_model-00207-of-00364.bin filter=lfs diff=lfs merge=lfs -text
211
+ pytorch_model-00298-of-00364.bin filter=lfs diff=lfs merge=lfs -text
212
+ pytorch_model-00341-of-00364.bin filter=lfs diff=lfs merge=lfs -text
213
+ pytorch_model-00350-of-00364.bin filter=lfs diff=lfs merge=lfs -text
214
+ pytorch_model-00077-of-00364.bin filter=lfs diff=lfs merge=lfs -text
215
+ pytorch_model-00109-of-00364.bin filter=lfs diff=lfs merge=lfs -text
216
+ pytorch_model-00175-of-00364.bin filter=lfs diff=lfs merge=lfs -text
217
+ pytorch_model-00178-of-00364.bin filter=lfs diff=lfs merge=lfs -text
218
+ pytorch_model-00239-of-00364.bin filter=lfs diff=lfs merge=lfs -text
219
+ pytorch_model-00245-of-00364.bin filter=lfs diff=lfs merge=lfs -text
220
+ pytorch_model-00250-of-00364.bin filter=lfs diff=lfs merge=lfs -text
221
+ pytorch_model-00334-of-00364.bin filter=lfs diff=lfs merge=lfs -text
222
+ pytorch_model-00023-of-00364.bin filter=lfs diff=lfs merge=lfs -text
223
+ pytorch_model-00065-of-00364.bin filter=lfs diff=lfs merge=lfs -text
224
+ pytorch_model-00099-of-00364.bin filter=lfs diff=lfs merge=lfs -text
225
+ pytorch_model-00198-of-00364.bin filter=lfs diff=lfs merge=lfs -text
226
+ pytorch_model-00356-of-00364.bin filter=lfs diff=lfs merge=lfs -text
227
+ pytorch_model-00187-of-00364.bin filter=lfs diff=lfs merge=lfs -text
228
+ pytorch_model-00211-of-00364.bin filter=lfs diff=lfs merge=lfs -text
229
+ pytorch_model-00253-of-00364.bin filter=lfs diff=lfs merge=lfs -text
230
+ pytorch_model-00013-of-00364.bin filter=lfs diff=lfs merge=lfs -text
231
+ pytorch_model-00138-of-00364.bin filter=lfs diff=lfs merge=lfs -text
232
+ pytorch_model-00143-of-00364.bin filter=lfs diff=lfs merge=lfs -text
233
+ pytorch_model-00152-of-00364.bin filter=lfs diff=lfs merge=lfs -text
234
+ pytorch_model-00322-of-00364.bin filter=lfs diff=lfs merge=lfs -text
235
+ pytorch_model-00351-of-00364.bin filter=lfs diff=lfs merge=lfs -text
236
+ pytorch_model-00052-of-00364.bin filter=lfs diff=lfs merge=lfs -text
237
+ pytorch_model-00154-of-00364.bin filter=lfs diff=lfs merge=lfs -text
238
+ pytorch_model-00155-of-00364.bin filter=lfs diff=lfs merge=lfs -text
239
+ pytorch_model-00246-of-00364.bin filter=lfs diff=lfs merge=lfs -text
240
+ pytorch_model-00290-of-00364.bin filter=lfs diff=lfs merge=lfs -text
241
+ pytorch_model-00064-of-00364.bin filter=lfs diff=lfs merge=lfs -text
242
+ pytorch_model-00098-of-00364.bin filter=lfs diff=lfs merge=lfs -text
243
+ pytorch_model-00162-of-00364.bin filter=lfs diff=lfs merge=lfs -text
244
+ pytorch_model-00265-of-00364.bin filter=lfs diff=lfs merge=lfs -text
245
+ pytorch_model-00177-of-00364.bin filter=lfs diff=lfs merge=lfs -text
246
+ pytorch_model-00232-of-00364.bin filter=lfs diff=lfs merge=lfs -text
247
+ pytorch_model-00038-of-00364.bin filter=lfs diff=lfs merge=lfs -text
248
+ pytorch_model-00048-of-00364.bin filter=lfs diff=lfs merge=lfs -text
249
+ pytorch_model-00141-of-00364.bin filter=lfs diff=lfs merge=lfs -text
250
+ pytorch_model-00166-of-00364.bin filter=lfs diff=lfs merge=lfs -text
251
+ pytorch_model-00193-of-00364.bin filter=lfs diff=lfs merge=lfs -text
252
+ pytorch_model-00263-of-00364.bin filter=lfs diff=lfs merge=lfs -text
253
+ pytorch_model-00273-of-00364.bin filter=lfs diff=lfs merge=lfs -text
254
+ pytorch_model-00045-of-00364.bin filter=lfs diff=lfs merge=lfs -text
255
+ pytorch_model-00094-of-00364.bin filter=lfs diff=lfs merge=lfs -text
256
+ pytorch_model-00135-of-00364.bin filter=lfs diff=lfs merge=lfs -text
257
+ pytorch_model-00167-of-00364.bin filter=lfs diff=lfs merge=lfs -text
258
+ pytorch_model-00191-of-00364.bin filter=lfs diff=lfs merge=lfs -text
259
+ pytorch_model-00231-of-00364.bin filter=lfs diff=lfs merge=lfs -text
260
+ pytorch_model-00264-of-00364.bin filter=lfs diff=lfs merge=lfs -text
261
+ pytorch_model-00277-of-00364.bin filter=lfs diff=lfs merge=lfs -text
262
+ pytorch_model-00022-of-00364.bin filter=lfs diff=lfs merge=lfs -text
263
+ pytorch_model-00074-of-00364.bin filter=lfs diff=lfs merge=lfs -text
264
+ pytorch_model-00078-of-00364.bin filter=lfs diff=lfs merge=lfs -text
265
+ pytorch_model-00114-of-00364.bin filter=lfs diff=lfs merge=lfs -text
266
+ pytorch_model-00280-of-00364.bin filter=lfs diff=lfs merge=lfs -text
267
+ pytorch_model-00343-of-00364.bin filter=lfs diff=lfs merge=lfs -text
268
+ pytorch_model-00304-of-00364.bin filter=lfs diff=lfs merge=lfs -text
269
+ pytorch_model-00126-of-00364.bin filter=lfs diff=lfs merge=lfs -text
270
+ pytorch_model-00149-of-00364.bin filter=lfs diff=lfs merge=lfs -text
271
+ pytorch_model-00184-of-00364.bin filter=lfs diff=lfs merge=lfs -text
272
+ pytorch_model-00240-of-00364.bin filter=lfs diff=lfs merge=lfs -text
273
+ pytorch_model-00049-of-00364.bin filter=lfs diff=lfs merge=lfs -text
274
+ pytorch_model-00168-of-00364.bin filter=lfs diff=lfs merge=lfs -text
275
+ pytorch_model-00266-of-00364.bin filter=lfs diff=lfs merge=lfs -text
276
+ pytorch_model-00294-of-00364.bin filter=lfs diff=lfs merge=lfs -text
277
+ pytorch_model-00308-of-00364.bin filter=lfs diff=lfs merge=lfs -text
278
+ pytorch_model-00352-of-00364.bin filter=lfs diff=lfs merge=lfs -text
279
+ pytorch_model-00079-of-00364.bin filter=lfs diff=lfs merge=lfs -text
280
+ pytorch_model-00090-of-00364.bin filter=lfs diff=lfs merge=lfs -text
281
+ pytorch_model-00153-of-00364.bin filter=lfs diff=lfs merge=lfs -text
282
+ pytorch_model-00225-of-00364.bin filter=lfs diff=lfs merge=lfs -text
283
+ pytorch_model-00355-of-00364.bin filter=lfs diff=lfs merge=lfs -text
284
+ pytorch_model-00340-of-00364.bin filter=lfs diff=lfs merge=lfs -text
285
+ pytorch_model-00007-of-00364.bin filter=lfs diff=lfs merge=lfs -text
286
+ pytorch_model-00040-of-00364.bin filter=lfs diff=lfs merge=lfs -text
287
+ pytorch_model-00042-of-00364.bin filter=lfs diff=lfs merge=lfs -text
288
+ pytorch_model-00112-of-00364.bin filter=lfs diff=lfs merge=lfs -text
289
+ pytorch_model-00148-of-00364.bin filter=lfs diff=lfs merge=lfs -text
290
+ pytorch_model-00262-of-00364.bin filter=lfs diff=lfs merge=lfs -text
291
+ pytorch_model-00309-of-00364.bin filter=lfs diff=lfs merge=lfs -text
292
+ pytorch_model-00303-of-00364.bin filter=lfs diff=lfs merge=lfs -text
293
+ pytorch_model-00028-of-00364.bin filter=lfs diff=lfs merge=lfs -text
294
+ pytorch_model-00092-of-00364.bin filter=lfs diff=lfs merge=lfs -text
295
+ pytorch_model-00102-of-00364.bin filter=lfs diff=lfs merge=lfs -text
296
+ pytorch_model-00165-of-00364.bin filter=lfs diff=lfs merge=lfs -text
297
+ pytorch_model-00054-of-00364.bin filter=lfs diff=lfs merge=lfs -text
298
+ pytorch_model-00190-of-00364.bin filter=lfs diff=lfs merge=lfs -text
299
+ pytorch_model-00325-of-00364.bin filter=lfs diff=lfs merge=lfs -text
300
+ pytorch_model-00118-of-00364.bin filter=lfs diff=lfs merge=lfs -text
301
+ pytorch_model-00144-of-00364.bin filter=lfs diff=lfs merge=lfs -text
302
+ pytorch_model-00279-of-00364.bin filter=lfs diff=lfs merge=lfs -text
303
+ pytorch_model-00301-of-00364.bin filter=lfs diff=lfs merge=lfs -text
304
+ pytorch_model-00134-of-00364.bin filter=lfs diff=lfs merge=lfs -text
305
+ pytorch_model-00218-of-00364.bin filter=lfs diff=lfs merge=lfs -text
306
+ pytorch_model-00327-of-00364.bin filter=lfs diff=lfs merge=lfs -text
307
+ pytorch_model-00332-of-00364.bin filter=lfs diff=lfs merge=lfs -text
308
+ pytorch_model-00002-of-00364.bin filter=lfs diff=lfs merge=lfs -text
309
+ pytorch_model-00004-of-00364.bin filter=lfs diff=lfs merge=lfs -text
310
+ pytorch_model-00085-of-00364.bin filter=lfs diff=lfs merge=lfs -text
311
+ pytorch_model-00095-of-00364.bin filter=lfs diff=lfs merge=lfs -text
312
+ pytorch_model-00362-of-00364.bin filter=lfs diff=lfs merge=lfs -text
313
+ pytorch_model-00209-of-00364.bin filter=lfs diff=lfs merge=lfs -text
314
+ pytorch_model-00256-of-00364.bin filter=lfs diff=lfs merge=lfs -text
315
+ pytorch_model-00281-of-00364.bin filter=lfs diff=lfs merge=lfs -text
316
+ pytorch_model-00349-of-00364.bin filter=lfs diff=lfs merge=lfs -text
317
+ pytorch_model-00030-of-00364.bin filter=lfs diff=lfs merge=lfs -text
318
+ pytorch_model-00035-of-00364.bin filter=lfs diff=lfs merge=lfs -text
319
+ pytorch_model-00039-of-00364.bin filter=lfs diff=lfs merge=lfs -text
320
+ pytorch_model-00192-of-00364.bin filter=lfs diff=lfs merge=lfs -text
321
+ pytorch_model-00312-of-00364.bin filter=lfs diff=lfs merge=lfs -text
322
+ pytorch_model-00338-of-00364.bin filter=lfs diff=lfs merge=lfs -text
323
+ pytorch_model-00189-of-00364.bin filter=lfs diff=lfs merge=lfs -text
324
+ pytorch_model-00204-of-00364.bin filter=lfs diff=lfs merge=lfs -text
325
+ pytorch_model-00213-of-00364.bin filter=lfs diff=lfs merge=lfs -text
326
+ pytorch_model-00306-of-00364.bin filter=lfs diff=lfs merge=lfs -text
327
+ pytorch_model-00248-of-00364.bin filter=lfs diff=lfs merge=lfs -text
328
+ pytorch_model-00272-of-00364.bin filter=lfs diff=lfs merge=lfs -text
329
+ pytorch_model-00044-of-00364.bin filter=lfs diff=lfs merge=lfs -text
330
+ pytorch_model-00137-of-00364.bin filter=lfs diff=lfs merge=lfs -text
331
+ pytorch_model-00201-of-00364.bin filter=lfs diff=lfs merge=lfs -text
332
+ pytorch_model-00205-of-00364.bin filter=lfs diff=lfs merge=lfs -text
333
+ pytorch_model-00229-of-00364.bin filter=lfs diff=lfs merge=lfs -text
334
+ pytorch_model-00259-of-00364.bin filter=lfs diff=lfs merge=lfs -text
335
+ pytorch_model-00269-of-00364.bin filter=lfs diff=lfs merge=lfs -text
336
+ pytorch_model-00093-of-00364.bin filter=lfs diff=lfs merge=lfs -text
337
+ pytorch_model-00100-of-00364.bin filter=lfs diff=lfs merge=lfs -text
338
+ pytorch_model-00159-of-00364.bin filter=lfs diff=lfs merge=lfs -text
339
+ pytorch_model-00196-of-00364.bin filter=lfs diff=lfs merge=lfs -text
340
+ pytorch_model-00357-of-00364.bin filter=lfs diff=lfs merge=lfs -text
341
+ pytorch_model-00020-of-00364.bin filter=lfs diff=lfs merge=lfs -text
342
+ pytorch_model-00061-of-00364.bin filter=lfs diff=lfs merge=lfs -text
343
+ pytorch_model-00087-of-00364.bin filter=lfs diff=lfs merge=lfs -text
344
+ pytorch_model-00288-of-00364.bin filter=lfs diff=lfs merge=lfs -text
345
+ pytorch_model-00289-of-00364.bin filter=lfs diff=lfs merge=lfs -text
346
+ pytorch_model-00346-of-00364.bin filter=lfs diff=lfs merge=lfs -text
347
+ pytorch_model-00060-of-00364.bin filter=lfs diff=lfs merge=lfs -text
348
+ pytorch_model-00131-of-00364.bin filter=lfs diff=lfs merge=lfs -text
349
+ pytorch_model-00171-of-00364.bin filter=lfs diff=lfs merge=lfs -text
350
+ pytorch_model-00226-of-00364.bin filter=lfs diff=lfs merge=lfs -text
351
+ pytorch_model-00326-of-00364.bin filter=lfs diff=lfs merge=lfs -text
352
+ pytorch_model-00062-of-00364.bin filter=lfs diff=lfs merge=lfs -text
353
+ pytorch_model-00235-of-00364.bin filter=lfs diff=lfs merge=lfs -text
354
+ pytorch_model-00282-of-00364.bin filter=lfs diff=lfs merge=lfs -text
355
+ pytorch_model-00296-of-00364.bin filter=lfs diff=lfs merge=lfs -text
356
+ pytorch_model-00032-of-00364.bin filter=lfs diff=lfs merge=lfs -text
357
+ pytorch_model-00215-of-00364.bin filter=lfs diff=lfs merge=lfs -text
358
+ pytorch_model-00363-of-00364.bin filter=lfs diff=lfs merge=lfs -text
359
+ pytorch_model-00202-of-00364.bin filter=lfs diff=lfs merge=lfs -text
360
+ pytorch_model-00221-of-00364.bin filter=lfs diff=lfs merge=lfs -text
361
+ pytorch_model-00284-of-00364.bin filter=lfs diff=lfs merge=lfs -text
362
+ pytorch_model-00311-of-00364.bin filter=lfs diff=lfs merge=lfs -text
363
+ pytorch_model-00047-of-00364.bin filter=lfs diff=lfs merge=lfs -text
364
+ pytorch_model-00157-of-00364.bin filter=lfs diff=lfs merge=lfs -text
365
+ pytorch_model-00164-of-00364.bin filter=lfs diff=lfs merge=lfs -text
366
+ pytorch_model-00176-of-00364.bin filter=lfs diff=lfs merge=lfs -text
367
+ pytorch_model-00361-of-00364.bin filter=lfs diff=lfs merge=lfs -text
368
+ pytorch_model-00302-of-00364.bin filter=lfs diff=lfs merge=lfs -text
369
+ pytorch_model-00180-of-00364.bin filter=lfs diff=lfs merge=lfs -text
370
+ pytorch_model-00206-of-00364.bin filter=lfs diff=lfs merge=lfs -text
371
+ pytorch_model-00241-of-00364.bin filter=lfs diff=lfs merge=lfs -text
372
+ pytorch_model-00251-of-00364.bin filter=lfs diff=lfs merge=lfs -text
373
+ pytorch_model-00031-of-00364.bin filter=lfs diff=lfs merge=lfs -text
374
+ pytorch_model-00068-of-00364.bin filter=lfs diff=lfs merge=lfs -text
375
+ pytorch_model-00179-of-00364.bin filter=lfs diff=lfs merge=lfs -text
376
+ pytorch_model-00344-of-00364.bin filter=lfs diff=lfs merge=lfs -text
377
+ pytorch_model-00278-of-00364.bin filter=lfs diff=lfs merge=lfs -text
378
+ pytorch_model-00096-of-00364.bin filter=lfs diff=lfs merge=lfs -text
379
+ pytorch_model-00121-of-00364.bin filter=lfs diff=lfs merge=lfs -text
380
+ pytorch_model-00142-of-00364.bin filter=lfs diff=lfs merge=lfs -text
381
+ pytorch_model-00258-of-00364.bin filter=lfs diff=lfs merge=lfs -text
382
+ pytorch_model-00041-of-00364.bin filter=lfs diff=lfs merge=lfs -text
383
+ pytorch_model-00186-of-00364.bin filter=lfs diff=lfs merge=lfs -text
384
+ pytorch_model-00333-of-00364.bin filter=lfs diff=lfs merge=lfs -text
385
+ pytorch_model-00001-of-00364.bin filter=lfs diff=lfs merge=lfs -text
386
+ pytorch_model-00021-of-00364.bin filter=lfs diff=lfs merge=lfs -text
387
+ pytorch_model-00024-of-00364.bin filter=lfs diff=lfs merge=lfs -text
388
+ pytorch_model-00029-of-00364.bin filter=lfs diff=lfs merge=lfs -text
389
+ pytorch_model-00073-of-00364.bin filter=lfs diff=lfs merge=lfs -text
390
+ pytorch_model-00097-of-00364.bin filter=lfs diff=lfs merge=lfs -text
391
+ pytorch_model-00291-of-00364.bin filter=lfs diff=lfs merge=lfs -text
392
+ pytorch_model-00195-of-00364.bin filter=lfs diff=lfs merge=lfs -text
393
+ pytorch_model-00252-of-00364.bin filter=lfs diff=lfs merge=lfs -text
394
+ pytorch_model-00318-of-00364.bin filter=lfs diff=lfs merge=lfs -text
395
+ pytorch_model-00076-of-00364.bin filter=lfs diff=lfs merge=lfs -text
396
+ pytorch_model-00082-of-00364.bin filter=lfs diff=lfs merge=lfs -text
397
+ pytorch_model-00130-of-00364.bin filter=lfs diff=lfs merge=lfs -text
398
+ pytorch_model-00254-of-00364.bin filter=lfs diff=lfs merge=lfs -text
399
+ pytorch_model.bin.index.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+
5
+ tags:
6
+ - text2text-generation
7
+
8
+ widget:
9
+ - text: "The <extra_id_0> walks in <extra_id_1> park"
10
+ example_title: "Masked Language Modeling"
11
+
12
+ datasets:
13
+ - c4
14
+
15
+ inference: false
16
+ license: apache-2.0
17
+ ---
18
+
19
+ # Model Card for Switch Transformers C - 2048 experts (1.6T parameters for 3.1 TB)
20
+
21
+ ![model image](https://cdn-uploads.huggingface.co/production/uploads/1666966931908-62441d1d9fdefb55a0b7d12c.png)
22
+
23
+ # Table of Contents
24
+
25
+ 0. [TL;DR](#TL;DR)
26
+ 1. [Model Details](#model-details)
27
+ 2. [Usage](#usage)
28
+ 3. [Uses](#uses)
29
+ 4. [Bias, Risks, and Limitations](#bias-risks-and-limitations)
30
+ 5. [Training Details](#training-details)
31
+ 6. [Evaluation](#evaluation)
32
+ 7. [Environmental Impact](#environmental-impact)
33
+ 8. [Citation](#citation)
34
+ 9. [Model Card Authors](#model-card-authors)
35
+
36
+ # TL;DR
37
+
38
+ Switch Transformers is a Mixture of Experts (MoE) model trained on Masked Language Modeling (MLM) task. The model architecture is similar to the classic T5, but with the Feed Forward layers replaced by the Sparse MLP layers containing "experts" MLP. According to the [original paper](https://arxiv.org/pdf/2101.03961.pdf) the model enables faster training (scaling properties) while being better than T5 on fine-tuned tasks.
39
+ As mentioned in the first few lines of the abstract :
40
+ > we advance the current scale of language models by pre-training up to trillion parameter models on the “Colossal Clean Crawled Corpus”, and achieve a 4x speedup over the T5-XXL model.
41
+
42
+ **Disclaimer**: Content from **this** model card has been written by the Hugging Face team, and parts of it were copy pasted from the [original paper](https://arxiv.org/pdf/2101.03961.pdf).
43
+
44
+ # Model Details
45
+
46
+ ## Model Description
47
+
48
+
49
+ - **Model type:** Language model
50
+ - **Language(s) (NLP):** English
51
+ - **License:** Apache 2.0
52
+ - **Related Models:** [All FLAN-T5 Checkpoints](https://huggingface.co/models?search=switch)
53
+ - **Original Checkpoints:** [All Original FLAN-T5 Checkpoints](https://github.com/google-research/t5x/blob/main/docs/models.md#mixture-of-experts-moe-checkpoints)
54
+ - **Resources for more information:**
55
+ - [Research paper](https://arxiv.org/pdf/2101.03961.pdf)
56
+ - [GitHub Repo](https://github.com/google-research/t5x)
57
+ - [Hugging Face Switch Transformers Docs (Similar to T5) ](https://huggingface.co/docs/transformers/model_doc/switch_transformers)
58
+
59
+ # Usage
60
+
61
+ Note that these checkpoints has been trained on Masked-Language Modeling (MLM) task. Therefore the checkpoints are not "ready-to-use" for downstream tasks. You may want to check `FLAN-T5` for running fine-tuned weights or fine-tune your own MoE following [this notebook](https://colab.research.google.com/drive/1aGGVHZmtKmcNBbAwa9hbu58DDpIuB5O4?usp=sharing)
62
+
63
+ Find below some example scripts on how to use the model in `transformers` - bear in mind that the model is **extremely** large, so you may consider using disk offload from `accelerate`:
64
+
65
+ ## Using the Pytorch model
66
+
67
+ ### Running the model on a CPU
68
+
69
+ <details>
70
+ <summary> Click to expand </summary>
71
+
72
+ ```python
73
+ # pip install accelerate
74
+ from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
75
+
76
+ tokenizer = AutoTokenizer.from_pretrained("google/switch-c-2048")
77
+ model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-c-2048", device_map="auto", offload_folder=<OFFLOAD_FOLDER>)
78
+
79
+ input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
80
+ input_ids = tokenizer(input_text, return_tensors="pt").input_ids
81
+
82
+ outputs = model.generate(input_ids)
83
+ print(tokenizer.decode(outputs[0]))
84
+ >>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
85
+ ```
86
+
87
+ </details>
88
+
89
+ ### Running the model on a GPU
90
+
91
+ <details>
92
+ <summary> Click to expand </summary>
93
+
94
+ ```python
95
+ # pip install accelerate
96
+ from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
97
+
98
+ tokenizer = AutoTokenizer.from_pretrained("google/switch-c-2048")
99
+ model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-c-2048", device_map="auto", offload_folder=<OFFLOAD_FOLDER>)
100
+
101
+ input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
102
+ input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(0)
103
+
104
+ outputs = model.generate(input_ids)
105
+ print(tokenizer.decode(outputs[0]))
106
+ >>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
107
+ ```
108
+
109
+ </details>
110
+
111
+ ### Running the model on a GPU using different precisions
112
+
113
+ #### BF16
114
+
115
+ <details>
116
+ <summary> Click to expand </summary>
117
+
118
+ ```python
119
+ # pip install accelerate
120
+ from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
121
+
122
+ tokenizer = AutoTokenizer.from_pretrained("google/switch-c-2048")
123
+ model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-c-2048", device_map="auto", torch_dtype=torch.bfloat16, offload_folder=<OFFLOAD_FOLDER>)
124
+
125
+ input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
126
+ input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(0)
127
+
128
+ outputs = model.generate(input_ids)
129
+ print(tokenizer.decode(outputs[0]))
130
+ >>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
131
+ ```
132
+
133
+ </details>
134
+
135
+ #### INT8
136
+
137
+ <details>
138
+ <summary> Click to expand </summary>
139
+
140
+ ```python
141
+ # pip install bitsandbytes accelerate
142
+ from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
143
+
144
+ tokenizer = AutoTokenizer.from_pretrained("google/switch-c-2048")
145
+ model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-c-2048", device_map="auto", offload_folder=<OFFLOAD_FOLDER>)
146
+
147
+ input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
148
+ input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(0)
149
+
150
+ outputs = model.generate(input_ids)
151
+ print(tokenizer.decode(outputs[0]))
152
+ >>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
153
+ ```
154
+
155
+ </details>
156
+
157
+ # Uses
158
+
159
+ ## Direct Use and Downstream Use
160
+
161
+ See the [research paper](https://arxiv.org/pdf/2101.03961.pdf) for further details.
162
+
163
+ ## Out-of-Scope Use
164
+
165
+ More information needed.
166
+
167
+ # Bias, Risks, and Limitations
168
+
169
+ More information needed.
170
+
171
+ ## Ethical considerations and risks
172
+
173
+ More information needed.
174
+
175
+ ## Known Limitations
176
+
177
+ More information needed.
178
+
179
+ ## Sensitive Use:
180
+
181
+ More information needed.
182
+
183
+ # Training Details
184
+
185
+ ## Training Data
186
+
187
+ The model was trained on a Masked Language Modeling task, on Colossal Clean Crawled Corpus (C4) dataset, following the same procedure as `T5`.
188
+
189
+
190
+ ## Training Procedure
191
+
192
+ According to the model card from the [original paper](https://arxiv.org/pdf/2101.03961.pdf) the model has been trained on TPU v3 or TPU v4 pods, using [`t5x`](https://github.com/google-research/t5x) codebase together with [`jax`](https://github.com/google/jax).
193
+
194
+
195
+ # Evaluation
196
+
197
+ ## Testing Data, Factors & Metrics
198
+
199
+ The authors evaluated the model on various tasks and compared the results against T5. See the table below for some quantitative evaluation:
200
+ ![image.png](https://s3.amazonaws.com/moonup/production/uploads/1666967660372-62441d1d9fdefb55a0b7d12c.png)
201
+ For full details, please check the [research paper](https://arxiv.org/pdf/2101.03961.pdf).
202
+
203
+ ## Results
204
+
205
+ For full results for Switch Transformers, see the [research paper](https://arxiv.org/pdf/2101.03961.pdf), Table 5.
206
+
207
+ # Environmental Impact
208
+
209
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
210
+
211
+ - **Hardware Type:** Google Cloud TPU Pods - TPU v3 or TPU v4 | Number of chips ≥ 4.
212
+ - **Hours used:** More information needed
213
+ - **Cloud Provider:** GCP
214
+ - **Compute Region:** More information needed
215
+ - **Carbon Emitted:** More information needed
216
+
217
+ # Citation
218
+
219
+ **BibTeX:**
220
+
221
+ ```bibtex
222
+ @misc{https://doi.org/10.48550/arxiv.2101.03961,
223
+ doi = {10.48550/ARXIV.2101.03961},
224
+
225
+ url = {https://arxiv.org/abs/2101.03961},
226
+
227
+ author = {Fedus, William and Zoph, Barret and Shazeer, Noam},
228
+
229
+ keywords = {Machine Learning (cs.LG), Artificial Intelligence (cs.AI), FOS: Computer and information sciences, FOS: Computer and information sciences},
230
+
231
+ title = {Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity},
232
+
233
+ publisher = {arXiv},
234
+
235
+ year = {2021},
236
+
237
+ copyright = {arXiv.org perpetual, non-exclusive license}
238
+ }
239
+
240
+ ```
config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_router_probs": false,
3
+ "d_ff": 6144,
4
+ "d_kv": 64,
5
+ "d_model": 2080,
6
+ "decoder_sparse_step": 0,
7
+ "dense_act_fn": "relu",
8
+ "dropout_rate": 0.1,
9
+ "encoder_sparse_step": 1,
10
+ "eos_token_id": 1,
11
+ "expert_capacity": 64,
12
+ "feed_forward_proj": "relu",
13
+ "initializer_factor": 1.0,
14
+ "decoder_start_token_id": 0,
15
+ "is_encoder_decoder": true,
16
+ "is_full_sparse": true,
17
+ "is_gated_act": false,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "switch_transformers",
20
+ "num_decoder_layers": 12,
21
+ "num_experts": 2048,
22
+ "num_heads": 30,
23
+ "num_layers": 15,
24
+ "num_sparse_decoder_layers": 15,
25
+ "num_sparse_encoder_layers": 15,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "router_aux_loss_coef": 0.001,
30
+ "router_bias": false,
31
+ "router_dtype": "float32",
32
+ "router_ignore_padding_tokens": false,
33
+ "router_jitter_noise": 0.01,
34
+ "router_type": "tokens_masked",
35
+ "router_z_loss_coef": 0.001,
36
+ "transformers_version": "4.25.0.dev0",
37
+ "use_cache": true,
38
+ "vocab_size": 32128
39
+ }
pytorch_model-00001-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:212feea8f74910ead498340582f6c5cf28bbd085bc4de02cb934c6d9a4ef5171
3
+ size 9968159397
pytorch_model-00002-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11a14c860519e4d2f7621b90f3fde1907e1c375fc0fc9e413d6d5f1f7b3236d2
3
+ size 9993718895
pytorch_model-00003-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02b520d22bead67c388b455e5c5efee6229ceeffb1c24073467f69b3f6304a61
3
+ size 9993719087
pytorch_model-00004-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22bbc83df24248e41b50f0bf5bdb5b65ce7b2b02fd97c8265ac90b4b98c71a1a
3
+ size 9993719279
pytorch_model-00005-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bb4198c8c9b86c92e1ed24402bd802fb416ac4faabb8fd0fb63dec63d7acb01
3
+ size 9993719279
pytorch_model-00006-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:085331026ed82f956cd5b0dd3654267ab7187345679f7359643ad0e27975f546
3
+ size 2402580619
pytorch_model-00007-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e80b277903f7067eb9e974678928466d2782af413fcaccac1d6ead003d02a19
3
+ size 9993718767
pytorch_model-00008-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a45a6c7a545c5ff587cad8f6f74843857db1d8e4142a765a2d04a0d0038b84ad
3
+ size 9993718895
pytorch_model-00009-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c70bcbceee215179dfa6f99f649429c331808a06322989c093ff3f32b2093ec6
3
+ size 9993719087
pytorch_model-00010-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96afefecd70051f557923efb0d39d0ef3bac8edadde5a2a0fa93248349ac5b97
3
+ size 9993719279
pytorch_model-00011-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d662a7e8b4589c6eec94190e4987da904bbe53a3d0c7ebb8d0f4e0d46ebedf2
3
+ size 9993719279
pytorch_model-00012-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87b36ceac3215f48f60eeaf50e15c6c67ab3dd7bf3f243779022340a7f26f6be
3
+ size 2377021198
pytorch_model-00013-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a46ce7b6dc418c0e3400d949ca67f67c6b25d1e137c993e36badefb70fe892ce
3
+ size 9917040593
pytorch_model-00014-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:611141dd22df27261b0b53055ac94e1973d99e1506dfe0712c6bdd59de9c4ce7
3
+ size 9993718895
pytorch_model-00015-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bd91e66c6f82d8c96347035d8733ad3972979f5c5dc9dcd3886d629a3526ca5
3
+ size 9993719087
pytorch_model-00016-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7166f9b9c6ae6de36908e59e5744ee991ecaf86b0d2769e9aea16b8eaefe4a0
3
+ size 9993719279
pytorch_model-00017-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5128105f58a1840c4c4217d93f73a0f3ffc30c2277f826d0c96ac220128437c
3
+ size 9993719279
pytorch_model-00018-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:492e5879646d6a079b943a74542fc2535f3168b0fddf6777c38baae426914874
3
+ size 2453699333
pytorch_model-00019-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1745a7964095195a9a8aa8e4979829d0e737cf939ffed7aaedce6d4454d56df
3
+ size 9993718767
pytorch_model-00020-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fcae612706c038dbb74c08ca65677c79db116fe3a04e4ad057d46a07e64e45a
3
+ size 9993718895
pytorch_model-00021-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2728c51fb6b1b3aa1e8329b6e788526489fd8f059595d02f87b87f4206aaca4
3
+ size 9993719087
pytorch_model-00022-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cf006aad65c14df872a9a2f552fe48f66f724c16a54317a6e8e11b39e70f4ef
3
+ size 9993719279
pytorch_model-00023-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e39d81b46378399d4dfabbf96cfe8e418595d8d2d5499918554fd1118fba99e
3
+ size 9993719279
pytorch_model-00024-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21480e4d2be587d3cecfba1593d71018c501cca90dee0959cf3f4a2336b5d7f5
3
+ size 2377021198
pytorch_model-00025-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b023761034d75535f24001ec8d81fa21b6cb919c453779d20e2e5b66c464007
3
+ size 9917040977
pytorch_model-00026-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aae30b1ca1c39257a1e7e59946095e897d6cc5653f6b9b1ce7b1dcf3829b35d0
3
+ size 9993719279
pytorch_model-00027-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3953a7c918e04d97f8d5233ca43b9d9113b43e7e84980e7fa0d453a2b0832cad
3
+ size 9993719471
pytorch_model-00028-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8dfc4a767e8a66d8537eb9f70dee908cea111e53ad945cf96b995a65e83427a4
3
+ size 9993719663
pytorch_model-00029-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b34aa8f7b2fe822cc30f9a5d065a4fe266c1d259916c7ade832d69b0597168ad
3
+ size 9993719663
pytorch_model-00030-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1d2c26f3d0ca4bf2b5fa17a6b37d4d83d48da9ba47423f827b67da90f768aef
3
+ size 2453699461
pytorch_model-00031-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f2e408b3e1b3dd8546f3d937f19402d82dc879e00bc5c2be7a7769ae4690b54
3
+ size 9993719151
pytorch_model-00032-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d917be10773d8bea3670e756206b6d61a1fd9b9edf239c70b074d2aa92a68f3
3
+ size 9993719279
pytorch_model-00033-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83f940ec9ea9dce0a5f699359641c84ef22be86e021e36b2d250147330775af6
3
+ size 9993719471
pytorch_model-00034-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39525430340f1672ec2f1302c36cbee6afcb54c7a2722ed39e4c539774e8243f
3
+ size 9993719663
pytorch_model-00035-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0cfa1a26d1389ecd792d82485192a638cb3347e638927c83d29d790304c4fde
3
+ size 9993719663
pytorch_model-00036-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9132777965dd46d6d26af1c3bff3623c2c02601f1b76f8da0d982faa1061a25
3
+ size 2377021326
pytorch_model-00037-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dea84bfc73ed37b76bb7e8ecef0ac063ebb3c26868f8dc4a42a460114ba8a56f
3
+ size 9917040977
pytorch_model-00038-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd109df090d17103f0db651b4dc2520fd2ad9c67edf27b406d4a73023a7c69c8
3
+ size 9993719279
pytorch_model-00039-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73cb50d63103e1830d31f8480effc1d10575ee52f339868ade779aab96d72fd5
3
+ size 9993719471
pytorch_model-00040-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c40acf2fbe03660e649586aa394d44cf4a1d883a07104cdcc0e73eb9d8a23575
3
+ size 9993719663
pytorch_model-00041-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:110279dc20be0cde4259cfde82c3638c5aa332bb862cdab27be4dd665d0c84c5
3
+ size 9993719663
pytorch_model-00042-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5feadd84287d116523c5ca6dc22a1d9f97f117822279d64d5be00f5ef75cb889
3
+ size 2453699461
pytorch_model-00043-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b46ef605e8c151e38292202685d7031cd9c33c0c0f7410ebe1dfccdfb1bce49
3
+ size 9993719151
pytorch_model-00044-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56fa93996262c8f8b890a3ecbca8b476a6bf70e1e70e5604ab1a0822b0932e42
3
+ size 9993719279
pytorch_model-00045-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d70199fe6aa97876a4ca32080c43cc9f1f93cc9467c82af2825e08bc90d05c3
3
+ size 9993719471
pytorch_model-00046-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b55d235cdc67cc4f0252a3f58febf3b42c8babd10fdd72746f9750fde700901
3
+ size 9993719663
pytorch_model-00047-of-00364.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b437241e6ea22558e9f7d0bca3c60615f326e1f0a107207464cf8468ef895829
3
+ size 9993719663