Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1994 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.2563. Min Act: -0.1700
Data Index: 841904 (C4 (Web Text))
Max Activating Token Index: 147
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.2248. Min Act: -0.1700
Data Index: 350430 (C4 (Web Text))
Max Activating Token Index: 449
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.1304. Min Act: -0.1700
Data Index: 80209 (C4 (Web Text))
Max Activating Token Index: 270
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0989. Min Act: -0.1700
Data Index: 635868 (C4 (Web Text))
Max Activating Token Index: 681
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.1146. Min Act: -0.1700
Data Index: 41814 (C4 (Web Text))
Max Activating Token Index: 351
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0989. Min Act: -0.1700
Data Index: 1282102 (C4 (Web Text))
Max Activating Token Index: 371
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.1461. Min Act: -0.1700
Data Index: 822599 (C4 (Web Text))
Max Activating Token Index: 531
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0989. Min Act: -0.1700
Data Index: 843557 (C4 (Web Text))
Max Activating Token Index: 362
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.1146. Min Act: -0.1700
Data Index: 955785 (C4 (Web Text))
Max Activating Token Index: 338
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0831. Min Act: -0.1700
Data Index: 1122688 (C4 (Web Text))
Max Activating Token Index: 646
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0989. Min Act: -0.1700
Data Index: 1317814 (C4 (Web Text))
Max Activating Token Index: 594
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.1146. Min Act: -0.1700
Data Index: 679132 (C4 (Web Text))
Max Activating Token Index: 412
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0989. Min Act: -0.1700
Data Index: 855567 (C4 (Web Text))
Max Activating Token Index: 311
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0989. Min Act: -0.1700
Data Index: 59283 (C4 (Web Text))
Max Activating Token Index: 9
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.1146. Min Act: -0.1700
Data Index: 1254894 (C4 (Web Text))
Max Activating Token Index: 263
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0831. Min Act: -0.1700
Data Index: 64091 (C4 (Web Text))
Max Activating Token Index: 306
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0357. Min Act: -0.1700
Data Index: 146409 (C4 (Web Text))
Max Activating Token Index: 191
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0673. Min Act: -0.1700
Data Index: 419272 (C4 (Web Text))
Max Activating Token Index: 552
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0831. Min Act: -0.1700
Data Index: 55274 (C4 (Web Text))
Max Activating Token Index: 454
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 3.2563. Min Range: -3.2563
Max Act: 3.0831. Min Act: -0.1700
Data Index: 200381 (C4 (Web Text))
Max Activating Token Index: 63
Click toggle to see full text
Truncated
Full Text #19