Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 493 in Layer 0
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.6508. Min Act: -0.1700
Data Index: 728478 (C4 (Web Text))
Max Activating Token Index: 14
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.6348. Min Act: -0.1700
Data Index: 1071462 (C4 (Web Text))
Max Activating Token Index: 627
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5865. Min Act: -0.1700
Data Index: 200980 (C4 (Web Text))
Max Activating Token Index: 384
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.6026. Min Act: -0.1700
Data Index: 176947 (C4 (Web Text))
Max Activating Token Index: 305
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5704. Min Act: -0.1700
Data Index: 1235836 (C4 (Web Text))
Max Activating Token Index: 621
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5704. Min Act: -0.1700
Data Index: 933496 (C4 (Web Text))
Max Activating Token Index: 32
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5543. Min Act: -0.1700
Data Index: 242454 (C4 (Web Text))
Max Activating Token Index: 541
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5543. Min Act: -0.1700
Data Index: 501375 (C4 (Web Text))
Max Activating Token Index: 212
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5543. Min Act: -0.1700
Data Index: 829469 (C4 (Web Text))
Max Activating Token Index: 645
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5381. Min Act: -0.1700
Data Index: 1379545 (Python Code)
Max Activating Token Index: 742
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5543. Min Act: -0.1700
Data Index: 949336 (C4 (Web Text))
Max Activating Token Index: 835
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5381. Min Act: -0.1700
Data Index: 278135 (C4 (Web Text))
Max Activating Token Index: 182
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5220. Min Act: -0.1700
Data Index: 1334165 (C4 (Web Text))
Max Activating Token Index: 851
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5220. Min Act: -0.1700
Data Index: 940177 (C4 (Web Text))
Max Activating Token Index: 179
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5220. Min Act: -0.1700
Data Index: 231233 (C4 (Web Text))
Max Activating Token Index: 636
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5220. Min Act: -0.1700
Data Index: 58824 (C4 (Web Text))
Max Activating Token Index: 632
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5220. Min Act: -0.1700
Data Index: 1382252 (Python Code)
Max Activating Token Index: 804
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.5058. Min Act: -0.1700
Data Index: 976811 (C4 (Web Text))
Max Activating Token Index: 723
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.4896. Min Act: -0.1700
Data Index: 676486 (C4 (Web Text))
Max Activating Token Index: 935
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.6508. Min Range: -2.6508
Max Act: 2.4734. Min Act: -0.1700
Data Index: 858013 (C4 (Web Text))
Max Activating Token Index: 925
Click toggle to see full text
Truncated
Full Text #19