Model: SoLU Model: 10 Layers, 5120 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 3638 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')
Text #0
Max Range: 3.4750. Min Range: -3.4750
Max Act: 3.4750. Min Act: -0.0001
Data Index: 425575 (C4 (Web Text))
Max Activating Token Index: 175
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 3.4750. Min Range: -3.4750
Max Act: 3.3115. Min Act: -0.0001
Data Index: 692323 (C4 (Web Text))
Max Activating Token Index: 605
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 3.4750. Min Range: -3.4750
Max Act: 2.4243. Min Act: -0.0001
Data Index: 332603 (C4 (Web Text))
Max Activating Token Index: 65
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 3.4750. Min Range: -3.4750
Max Act: 2.5880. Min Act: -0.0001
Data Index: 914428 (C4 (Web Text))
Max Activating Token Index: 472
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.9896. Min Act: -0.0001
Data Index: 910994 (C4 (Web Text))
Max Activating Token Index: 781
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 3.4750. Min Range: -3.4750
Max Act: 2.2972. Min Act: -0.0001
Data Index: 1052492 (C4 (Web Text))
Max Activating Token Index: 837
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 3.4750. Min Range: -3.4750
Max Act: 2.1253. Min Act: -0.0001
Data Index: 1613142 (Python Code)
Max Activating Token Index: 477
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 3.4750. Min Range: -3.4750
Max Act: 2.3230. Min Act: -0.0001
Data Index: 577933 (C4 (Web Text))
Max Activating Token Index: 309
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.9608. Min Act: -0.0001
Data Index: 1431309 (Python Code)
Max Activating Token Index: 427
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.8089. Min Act: -0.0001
Data Index: 111867 (C4 (Web Text))
Max Activating Token Index: 489
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.7894. Min Act: -0.0001
Data Index: 1578289 (Python Code)
Max Activating Token Index: 841
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.7928. Min Act: -0.0001
Data Index: 1507298 (Python Code)
Max Activating Token Index: 725
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.8793. Min Act: -0.0001
Data Index: 1118602 (C4 (Web Text))
Max Activating Token Index: 733
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.8646. Min Act: -0.0001
Data Index: 804854 (C4 (Web Text))
Max Activating Token Index: 1005
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.8435. Min Act: -0.0001
Data Index: 60988 (C4 (Web Text))
Max Activating Token Index: 156
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.6773. Min Act: -0.0001
Data Index: 1600110 (Python Code)
Max Activating Token Index: 758
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.5319. Min Act: -0.0001
Data Index: 608013 (C4 (Web Text))
Max Activating Token Index: 753
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.7395. Min Act: -0.0001
Data Index: 626732 (C4 (Web Text))
Max Activating Token Index: 1015
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.7288. Min Act: -0.0001
Data Index: 239320 (C4 (Web Text))
Max Activating Token Index: 936
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 3.4750. Min Range: -3.4750
Max Act: 1.7143. Min Act: -0.0001
Data Index: 146701 (C4 (Web Text))
Max Activating Token Index: 338
Click toggle to see full text
Truncated
Full Text #19