Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 100 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.9295. Min Act: -0.1700
Data Index: 86147 (C4 (Web Text))
Max Activating Token Index: 43
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5857. Min Act: -0.1700
Data Index: 486172 (C4 (Web Text))
Max Activating Token Index: 790
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.6795. Min Act: -0.1700
Data Index: 620071 (C4 (Web Text))
Max Activating Token Index: 434
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5545. Min Act: -0.1700
Data Index: 992356 (C4 (Web Text))
Max Activating Token Index: 352
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5857. Min Act: -0.1700
Data Index: 372051 (C4 (Web Text))
Max Activating Token Index: 964
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5545. Min Act: -0.1700
Data Index: 191734 (C4 (Web Text))
Max Activating Token Index: 749
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5545. Min Act: -0.1700
Data Index: 931166 (C4 (Web Text))
Max Activating Token Index: 162
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5232. Min Act: -0.1700
Data Index: 72140 (C4 (Web Text))
Max Activating Token Index: 240
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5857. Min Act: -0.1700
Data Index: 1206124 (C4 (Web Text))
Max Activating Token Index: 945
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.6170. Min Act: -0.1700
Data Index: 819664 (C4 (Web Text))
Max Activating Token Index: 143
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5232. Min Act: -0.1700
Data Index: 350030 (C4 (Web Text))
Max Activating Token Index: 833
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5545. Min Act: -0.1700
Data Index: 902471 (C4 (Web Text))
Max Activating Token Index: 738
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5232. Min Act: -0.1700
Data Index: 1000470 (C4 (Web Text))
Max Activating Token Index: 882
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5545. Min Act: -0.1700
Data Index: 1137444 (C4 (Web Text))
Max Activating Token Index: 438
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5232. Min Act: -0.1700
Data Index: 165222 (C4 (Web Text))
Max Activating Token Index: 524
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5545. Min Act: -0.1700
Data Index: 882654 (C4 (Web Text))
Max Activating Token Index: 538
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5857. Min Act: -0.1700
Data Index: 884068 (C4 (Web Text))
Max Activating Token Index: 691
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5857. Min Act: -0.1700
Data Index: 1224393 (C4 (Web Text))
Max Activating Token Index: 503
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.5232. Min Act: -0.1700
Data Index: 1227913 (C4 (Web Text))
Max Activating Token Index: 293
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.9295. Min Range: -4.9295
Max Act: 4.4920. Min Act: -0.1700
Data Index: 417300 (C4 (Web Text))
Max Activating Token Index: 587
Click toggle to see full text
Truncated
Full Text #19