Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2010 in Layer 0
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.9049. Min Act: -0.1700
Data Index: 162341 (C4 (Web Text))
Max Activating Token Index: 31
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8573. Min Act: -0.1700
Data Index: 959647 (C4 (Web Text))
Max Activating Token Index: 285
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8414. Min Act: -0.1700
Data Index: 463744 (C4 (Web Text))
Max Activating Token Index: 525
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8255. Min Act: -0.1700
Data Index: 187845 (C4 (Web Text))
Max Activating Token Index: 772
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8255. Min Act: -0.1700
Data Index: 627314 (C4 (Web Text))
Max Activating Token Index: 987
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8255. Min Act: -0.1700
Data Index: 406916 (C4 (Web Text))
Max Activating Token Index: 196
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8255. Min Act: -0.1700
Data Index: 500221 (C4 (Web Text))
Max Activating Token Index: 576
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.8096. Min Act: -0.1700
Data Index: 903946 (C4 (Web Text))
Max Activating Token Index: 471
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7937. Min Act: -0.1700
Data Index: 922996 (C4 (Web Text))
Max Activating Token Index: 983
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7618. Min Act: -0.1700
Data Index: 1238980 (C4 (Web Text))
Max Activating Token Index: 809
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7937. Min Act: -0.1700
Data Index: 1284762 (C4 (Web Text))
Max Activating Token Index: 478
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7937. Min Act: -0.1700
Data Index: 613440 (C4 (Web Text))
Max Activating Token Index: 897
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7777. Min Act: -0.1700
Data Index: 376408 (C4 (Web Text))
Max Activating Token Index: 236
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7777. Min Act: -0.1700
Data Index: 568333 (C4 (Web Text))
Max Activating Token Index: 63
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7777. Min Act: -0.1700
Data Index: 1028510 (C4 (Web Text))
Max Activating Token Index: 15
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7618. Min Act: -0.1700
Data Index: 530621 (C4 (Web Text))
Max Activating Token Index: 74
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7458. Min Act: -0.1700
Data Index: 248529 (C4 (Web Text))
Max Activating Token Index: 782
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7618. Min Act: -0.1700
Data Index: 848814 (C4 (Web Text))
Max Activating Token Index: 49
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7618. Min Act: -0.1700
Data Index: 923423 (C4 (Web Text))
Max Activating Token Index: 42
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.9049. Min Range: -2.9049
Max Act: 2.7299. Min Act: -0.1700
Data Index: 491688 (C4 (Web Text))
Max Activating Token Index: 705
Click toggle to see full text
Truncated
Full Text #19