Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2022 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.9395. Min Act: -0.1699
Data Index: 798981 (C4 (Web Text))
Max Activating Token Index: 902
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8769. Min Act: -0.1700
Data Index: 843101 (C4 (Web Text))
Max Activating Token Index: 842
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8457. Min Act: -0.1700
Data Index: 1025983 (C4 (Web Text))
Max Activating Token Index: 983
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8769. Min Act: -0.1700
Data Index: 739553 (C4 (Web Text))
Max Activating Token Index: 846
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8769. Min Act: -0.1700
Data Index: 1243546 (C4 (Web Text))
Max Activating Token Index: 706
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8769. Min Act: -0.1700
Data Index: 683071 (C4 (Web Text))
Max Activating Token Index: 304
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8457. Min Act: -0.1700
Data Index: 238639 (C4 (Web Text))
Max Activating Token Index: 775
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8769. Min Act: -0.1700
Data Index: 1198796 (C4 (Web Text))
Max Activating Token Index: 614
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.7831. Min Act: -0.1700
Data Index: 697037 (C4 (Web Text))
Max Activating Token Index: 326
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8144. Min Act: -0.1700
Data Index: 89547 (C4 (Web Text))
Max Activating Token Index: 516
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8144. Min Act: -0.1700
Data Index: 421010 (C4 (Web Text))
Max Activating Token Index: 504
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8144. Min Act: -0.1700
Data Index: 566875 (C4 (Web Text))
Max Activating Token Index: 869
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8144. Min Act: -0.1700
Data Index: 916743 (C4 (Web Text))
Max Activating Token Index: 961
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.7831. Min Act: -0.1700
Data Index: 345792 (C4 (Web Text))
Max Activating Token Index: 379
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8144. Min Act: -0.1700
Data Index: 628783 (C4 (Web Text))
Max Activating Token Index: 344
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8457. Min Act: -0.1700
Data Index: 584537 (C4 (Web Text))
Max Activating Token Index: 870
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.8144. Min Act: -0.1700
Data Index: 1063022 (C4 (Web Text))
Max Activating Token Index: 914
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.7831. Min Act: -0.1700
Data Index: 1163672 (C4 (Web Text))
Max Activating Token Index: 845
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.7205. Min Act: -0.1700
Data Index: 349582 (C4 (Web Text))
Max Activating Token Index: 950
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 3.9395. Min Range: -3.9395
Max Act: 3.7205. Min Act: -0.1700
Data Index: 782848 (C4 (Web Text))
Max Activating Token Index: 976
Click toggle to see full text
Truncated
Full Text #19