Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1608 in Layer 0
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.8478. Min Act: -0.1700
Data Index: 325466 (C4 (Web Text))
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.8319. Min Act: -0.1700
Data Index: 557438 (C4 (Web Text))
Max Activating Token Index: 391
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.8001. Min Act: -0.1700
Data Index: 702643 (C4 (Web Text))
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.8001. Min Act: -0.1700
Data Index: 480789 (C4 (Web Text))
Max Activating Token Index: 67
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.8001. Min Act: -0.1700
Data Index: 105930 (C4 (Web Text))
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7842. Min Act: -0.1700
Data Index: 118039 (C4 (Web Text))
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7842. Min Act: -0.1700
Data Index: 979972 (C4 (Web Text))
Max Activating Token Index: 6
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7842. Min Act: -0.1700
Data Index: 1310103 (C4 (Web Text))
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7682. Min Act: -0.1700
Data Index: 883494 (C4 (Web Text))
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7682. Min Act: -0.1700
Data Index: 262002 (C4 (Web Text))
Max Activating Token Index: 950
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7682. Min Act: -0.1700
Data Index: 723767 (C4 (Web Text))
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 1481192 (Python Code)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 1318971 (C4 (Web Text))
Max Activating Token Index: 25
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 225948 (C4 (Web Text))
Max Activating Token Index: 907
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 57990 (C4 (Web Text))
Max Activating Token Index: 10
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 1039728 (C4 (Web Text))
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 928401 (C4 (Web Text))
Max Activating Token Index: 1020
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7523. Min Act: -0.1700
Data Index: 979101 (C4 (Web Text))
Max Activating Token Index: 5
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7363. Min Act: -0.1700
Data Index: 564036 (C4 (Web Text))
Max Activating Token Index: 7
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.8478. Min Range: -2.8478
Max Act: 2.7363. Min Act: -0.1700
Data Index: 1180060 (C4 (Web Text))
Max Activating Token Index: 10
Click toggle to see full text
Truncated
Full Text #19