Model: GELU Model: 2 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2023 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')
Text #0
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.7555. Min Act: -0.1700
Data Index: 112884 (C4 (Web Text))
Max Activating Token Index: 211
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.6930. Min Act: -0.1700
Data Index: 516062 (C4 (Web Text))
Max Activating Token Index: 46
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.6461. Min Act: -0.1700
Data Index: 21642 (C4 (Web Text))
Max Activating Token Index: 118
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.5680. Min Act: -0.1700
Data Index: 451852 (C4 (Web Text))
Max Activating Token Index: 876
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.5680. Min Act: -0.1700
Data Index: 496419 (C4 (Web Text))
Max Activating Token Index: 794
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.5055. Min Act: -0.1700
Data Index: 1049210 (C4 (Web Text))
Max Activating Token Index: 510
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4742. Min Act: -0.1700
Data Index: 947665 (C4 (Web Text))
Max Activating Token Index: 130
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4586. Min Act: -0.1700
Data Index: 949892 (C4 (Web Text))
Max Activating Token Index: 40
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4430. Min Act: -0.1700
Data Index: 1087585 (C4 (Web Text))
Max Activating Token Index: 169
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4273. Min Act: -0.1700
Data Index: 1334577 (C4 (Web Text))
Max Activating Token Index: 99
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4586. Min Act: -0.1700
Data Index: 681547 (C4 (Web Text))
Max Activating Token Index: 339
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4117. Min Act: -0.1700
Data Index: 485296 (C4 (Web Text))
Max Activating Token Index: 147
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.4273. Min Act: -0.1700
Data Index: 781500 (C4 (Web Text))
Max Activating Token Index: 132
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3961. Min Act: -0.1700
Data Index: 190220 (C4 (Web Text))
Max Activating Token Index: 42
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3805. Min Act: -0.1700
Data Index: 18644 (C4 (Web Text))
Max Activating Token Index: 78
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3961. Min Act: -0.1700
Data Index: 160619 (C4 (Web Text))
Max Activating Token Index: 242
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3961. Min Act: -0.1700
Data Index: 317024 (C4 (Web Text))
Max Activating Token Index: 424
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3805. Min Act: -0.1700
Data Index: 461335 (C4 (Web Text))
Max Activating Token Index: 14
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3805. Min Act: -0.1700
Data Index: 296661 (C4 (Web Text))
Max Activating Token Index: 519
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.7555. Min Range: -4.7555
Max Act: 4.3648. Min Act: -0.1700
Data Index: 35653 (C4 (Web Text))
Max Activating Token Index: 26
Click toggle to see full text
Truncated
Full Text #19