Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 905 in Layer 3
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 6.5179. Min Range: -6.5179
Max Act: 6.5179. Min Act: -0.1700
Data Index: 1409952 (Python Code)
Max Activating Token Index: 42
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 6.5179. Min Range: -6.5179
Max Act: 6.2991. Min Act: -0.1700
Data Index: 331379 (C4 (Web Text))
Max Activating Token Index: 794
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 6.5179. Min Range: -6.5179
Max Act: 6.2679. Min Act: -0.1700
Data Index: 1423527 (Python Code)
Max Activating Token Index: 702
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 6.5179. Min Range: -6.5179
Max Act: 6.0179. Min Act: -0.1700
Data Index: 723495 (C4 (Web Text))
Max Activating Token Index: 469
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 6.5179. Min Range: -6.5179
Max Act: 6.0179. Min Act: -0.1700
Data Index: 1045383 (C4 (Web Text))
Max Activating Token Index: 143
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.9866. Min Act: -0.1700
Data Index: 563809 (C4 (Web Text))
Max Activating Token Index: 904
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.9866. Min Act: -0.1700
Data Index: 526329 (C4 (Web Text))
Max Activating Token Index: 780
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.9241. Min Act: -0.1700
Data Index: 1381867 (Python Code)
Max Activating Token Index: 934
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.8616. Min Act: -0.1700
Data Index: 1000989 (C4 (Web Text))
Max Activating Token Index: 834
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.8304. Min Act: -0.1700
Data Index: 556615 (C4 (Web Text))
Max Activating Token Index: 614
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.7991. Min Act: -0.1700
Data Index: 1370096 (Python Code)
Max Activating Token Index: 913
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.7991. Min Act: -0.1700
Data Index: 457880 (C4 (Web Text))
Max Activating Token Index: 504
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.8616. Min Act: -0.1700
Data Index: 657679 (C4 (Web Text))
Max Activating Token Index: 151
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.7054. Min Act: -0.1700
Data Index: 1398767 (Python Code)
Max Activating Token Index: 954
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.6741. Min Act: -0.1700
Data Index: 1388038 (Python Code)
Max Activating Token Index: 652
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.8304. Min Act: -0.1700
Data Index: 1146561 (C4 (Web Text))
Max Activating Token Index: 658
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.7679. Min Act: -0.1700
Data Index: 1312830 (C4 (Web Text))
Max Activating Token Index: 860
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.6741. Min Act: -0.1700
Data Index: 636145 (C4 (Web Text))
Max Activating Token Index: 250
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.7054. Min Act: -0.1700
Data Index: 1459571 (Python Code)
Max Activating Token Index: 893
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 6.5179. Min Range: -6.5179
Max Act: 5.7054. Min Act: -0.1700
Data Index: 1374768 (Python Code)
Max Activating Token Index: 1015
Click toggle to see full text
Truncated
Full Text #19