Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 27 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.9526. Min Act: -0.1700
Data Index: 810226 (C4 (Web Text))
Max Activating Token Index: 329
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.9368. Min Act: -0.1700
Data Index: 661357 (C4 (Web Text))
Max Activating Token Index: 794
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.9526. Min Act: -0.1700
Data Index: 864563 (C4 (Web Text))
Max Activating Token Index: 498
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.9368. Min Act: -0.1700
Data Index: 1240624 (C4 (Web Text))
Max Activating Token Index: 492
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.9051. Min Act: -0.1700
Data Index: 483591 (C4 (Web Text))
Max Activating Token Index: 335
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8416. Min Act: -0.1700
Data Index: 428819 (C4 (Web Text))
Max Activating Token Index: 682
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8416. Min Act: -0.1700
Data Index: 440486 (C4 (Web Text))
Max Activating Token Index: 928
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8574. Min Act: -0.1700
Data Index: 1399219 (Python Code)
Max Activating Token Index: 827
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8256. Min Act: -0.1700
Data Index: 1101487 (C4 (Web Text))
Max Activating Token Index: 511
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8097. Min Act: -0.1700
Data Index: 409963 (C4 (Web Text))
Max Activating Token Index: 442
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.7938. Min Act: -0.1700
Data Index: 1268365 (C4 (Web Text))
Max Activating Token Index: 153
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.6660. Min Act: -0.1700
Data Index: 1388800 (Python Code)
Max Activating Token Index: 898
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8256. Min Act: -0.1700
Data Index: 1309270 (C4 (Web Text))
Max Activating Token Index: 234
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8574. Min Act: -0.1700
Data Index: 1357517 (C4 (Web Text))
Max Activating Token Index: 435
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.7938. Min Act: -0.1700
Data Index: 112814 (C4 (Web Text))
Max Activating Token Index: 799
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8733. Min Act: -0.1700
Data Index: 1232263 (C4 (Web Text))
Max Activating Token Index: 815
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.7779. Min Act: -0.1700
Data Index: 785972 (C4 (Web Text))
Max Activating Token Index: 385
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.7779. Min Act: -0.1700
Data Index: 1310390 (C4 (Web Text))
Max Activating Token Index: 719
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.8256. Min Act: -0.1700
Data Index: 1202415 (C4 (Web Text))
Max Activating Token Index: 633
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.9526. Min Range: -2.9526
Max Act: 2.7779. Min Act: -0.1700
Data Index: 1589335 (Python Code)
Max Activating Token Index: 663
Click toggle to see full text
Truncated
Full Text #19