Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2015 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.2374. Min Act: -0.1700
Data Index: 216704 (C4 (Web Text))
Max Activating Token Index: 170
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.2061. Min Act: -0.1700
Data Index: 881781 (C4 (Web Text))
Max Activating Token Index: 262
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.1748. Min Act: -0.1700
Data Index: 1097468 (C4 (Web Text))
Max Activating Token Index: 67
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.1748. Min Act: -0.1700
Data Index: 805675 (C4 (Web Text))
Max Activating Token Index: 203
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.1748. Min Act: -0.1700
Data Index: 135776 (C4 (Web Text))
Max Activating Token Index: 59
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.1123. Min Act: -0.1700
Data Index: 197113 (C4 (Web Text))
Max Activating Token Index: 655
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0811. Min Act: -0.1700
Data Index: 1335328 (C4 (Web Text))
Max Activating Token Index: 888
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0811. Min Act: -0.1700
Data Index: 597083 (C4 (Web Text))
Max Activating Token Index: 68
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0811. Min Act: -0.1700
Data Index: 1020688 (C4 (Web Text))
Max Activating Token Index: 611
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0811. Min Act: -0.1700
Data Index: 432043 (C4 (Web Text))
Max Activating Token Index: 140
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0811. Min Act: -0.1700
Data Index: 843806 (C4 (Web Text))
Max Activating Token Index: 267
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0498. Min Act: -0.1700
Data Index: 1116471 (C4 (Web Text))
Max Activating Token Index: 133
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0811. Min Act: -0.1700
Data Index: 475384 (C4 (Web Text))
Max Activating Token Index: 668
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0498. Min Act: -0.1700
Data Index: 872515 (C4 (Web Text))
Max Activating Token Index: 749
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0185. Min Act: -0.1700
Data Index: 418420 (C4 (Web Text))
Max Activating Token Index: 230
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.2374. Min Range: -4.2374
Max Act: 3.9873. Min Act: -0.1700
Data Index: 503682 (C4 (Web Text))
Max Activating Token Index: 216
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.2374. Min Range: -4.2374
Max Act: 3.9873. Min Act: -0.1700
Data Index: 1117799 (C4 (Web Text))
Max Activating Token Index: 849
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0498. Min Act: -0.1700
Data Index: 870921 (C4 (Web Text))
Max Activating Token Index: 661
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0185. Min Act: -0.1700
Data Index: 785734 (C4 (Web Text))
Max Activating Token Index: 109
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.2374. Min Range: -4.2374
Max Act: 4.0498. Min Act: -0.1700
Data Index: 987605 (C4 (Web Text))
Max Activating Token Index: 126
Click toggle to see full text
Truncated
Full Text #19