Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 262 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.8582. Min Act: -0.1700
Data Index: 927771 (C4 (Web Text))
Max Activating Token Index: 714
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.7957. Min Act: -0.1700
Data Index: 1416331 (Python Code)
Max Activating Token Index: 592
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.7644. Min Act: -0.1700
Data Index: 919799 (C4 (Web Text))
Max Activating Token Index: 486
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.7332. Min Act: -0.1700
Data Index: 1538477 (Python Code)
Max Activating Token Index: 591
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.7332. Min Act: -0.1700
Data Index: 1107140 (C4 (Web Text))
Max Activating Token Index: 831
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.7019. Min Act: -0.1700
Data Index: 1090288 (C4 (Web Text))
Max Activating Token Index: 715
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6707. Min Act: -0.1700
Data Index: 799658 (C4 (Web Text))
Max Activating Token Index: 524
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6394. Min Act: -0.1700
Data Index: 1383461 (Python Code)
Max Activating Token Index: 512
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.7019. Min Act: -0.1700
Data Index: 969046 (C4 (Web Text))
Max Activating Token Index: 472
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6707. Min Act: -0.1700
Data Index: 1144922 (C4 (Web Text))
Max Activating Token Index: 727
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6082. Min Act: -0.1700
Data Index: 78015 (C4 (Web Text))
Max Activating Token Index: 720
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6394. Min Act: -0.1700
Data Index: 1038726 (C4 (Web Text))
Max Activating Token Index: 184
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6082. Min Act: -0.1700
Data Index: 202669 (C4 (Web Text))
Max Activating Token Index: 885
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6082. Min Act: -0.1700
Data Index: 458357 (C4 (Web Text))
Max Activating Token Index: 851
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6082. Min Act: -0.1700
Data Index: 376809 (C4 (Web Text))
Max Activating Token Index: 600
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.5457. Min Act: -0.1700
Data Index: 144000 (C4 (Web Text))
Max Activating Token Index: 852
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6082. Min Act: -0.1700
Data Index: 1283257 (C4 (Web Text))
Max Activating Token Index: 432
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6394. Min Act: -0.1700
Data Index: 1448673 (Python Code)
Max Activating Token Index: 168
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.5769. Min Act: -0.1700
Data Index: 1532693 (Python Code)
Max Activating Token Index: 516
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.8582. Min Range: -4.8582
Max Act: 4.6394. Min Act: -0.1700
Data Index: 1584983 (Python Code)
Max Activating Token Index: 758
Click toggle to see full text
Truncated
Full Text #19