Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1990 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 5.1666. Min Range: -5.1666
Max Act: 5.1353. Min Act: -0.1700
Data Index: 215231 (C4 (Web Text))
Max Activating Token Index: 453
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 5.1666. Min Range: -5.1666
Max Act: 5.1041. Min Act: -0.1700
Data Index: 24389 (C4 (Web Text))
Max Activating Token Index: 415
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 5.1666. Min Range: -5.1666
Max Act: 5.1666. Min Act: -0.1700
Data Index: 1234774 (C4 (Web Text))
Max Activating Token Index: 443
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.9478. Min Act: -0.1700
Data Index: 451157 (C4 (Web Text))
Max Activating Token Index: 547
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.9791. Min Act: -0.1700
Data Index: 382793 (C4 (Web Text))
Max Activating Token Index: 972
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.9166. Min Act: -0.1700
Data Index: 466231 (C4 (Web Text))
Max Activating Token Index: 850
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.9166. Min Act: -0.1700
Data Index: 1247206 (C4 (Web Text))
Max Activating Token Index: 743
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.8853. Min Act: -0.1700
Data Index: 1352346 (C4 (Web Text))
Max Activating Token Index: 174
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.9478. Min Act: -0.1700
Data Index: 285395 (C4 (Web Text))
Max Activating Token Index: 145
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.7603. Min Act: -0.1700
Data Index: 319545 (C4 (Web Text))
Max Activating Token Index: 387
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6978. Min Act: -0.1700
Data Index: 326795 (C4 (Web Text))
Max Activating Token Index: 30
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.7291. Min Act: -0.1700
Data Index: 591920 (C4 (Web Text))
Max Activating Token Index: 744
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6666. Min Act: -0.1700
Data Index: 528284 (C4 (Web Text))
Max Activating Token Index: 162
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6666. Min Act: -0.1700
Data Index: 1061181 (C4 (Web Text))
Max Activating Token Index: 728
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6978. Min Act: -0.1700
Data Index: 1151556 (C4 (Web Text))
Max Activating Token Index: 735
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6041. Min Act: -0.1700
Data Index: 1185171 (C4 (Web Text))
Max Activating Token Index: 813
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6666. Min Act: -0.1700
Data Index: 1132569 (C4 (Web Text))
Max Activating Token Index: 783
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6353. Min Act: -0.1700
Data Index: 570604 (C4 (Web Text))
Max Activating Token Index: 98
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.6041. Min Act: -0.1700
Data Index: 140593 (C4 (Web Text))
Max Activating Token Index: 116
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 5.1666. Min Range: -5.1666
Max Act: 4.5728. Min Act: -0.1700
Data Index: 1114185 (C4 (Web Text))
Max Activating Token Index: 480
Click toggle to see full text
Truncated
Full Text #19