Model: SoLU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1644 in Layer 3
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')
Text #0
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.6224. Min Act: -0.0002
Data Index: 125913 (C4 (Web Text))
Max Activating Token Index: 969
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.4314. Min Act: -0.0002
Data Index: 1451502 (Python Code)
Max Activating Token Index: 338
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.2791. Min Act: -0.0002
Data Index: 864533 (C4 (Web Text))
Max Activating Token Index: 349
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.1911. Min Act: -0.0001
Data Index: 550086 (C4 (Web Text))
Max Activating Token Index: 269
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.1375. Min Act: -0.0002
Data Index: 611699 (C4 (Web Text))
Max Activating Token Index: 577
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.1040. Min Act: -0.0001
Data Index: 39131 (C4 (Web Text))
Max Activating Token Index: 859
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0792. Min Act: -0.0001
Data Index: 603602 (C4 (Web Text))
Max Activating Token Index: 649
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.1208. Min Act: -0.0001
Data Index: 853160 (C4 (Web Text))
Max Activating Token Index: 543
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.1300. Min Act: -0.0002
Data Index: 37723 (C4 (Web Text))
Max Activating Token Index: 416
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.1162. Min Act: -0.0001
Data Index: 760539 (C4 (Web Text))
Max Activating Token Index: 586
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0792. Min Act: -0.0002
Data Index: 1303811 (C4 (Web Text))
Max Activating Token Index: 407
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0912. Min Act: -0.0001
Data Index: 340423 (C4 (Web Text))
Max Activating Token Index: 456
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0600. Min Act: -0.0001
Data Index: 300445 (C4 (Web Text))
Max Activating Token Index: 903
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0468. Min Act: -0.0001
Data Index: 910528 (C4 (Web Text))
Max Activating Token Index: 988
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0873. Min Act: -0.0001
Data Index: 1280694 (C4 (Web Text))
Max Activating Token Index: 368
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.6224. Min Range: -1.6224
Max Act: 0.9617. Min Act: -0.0002
Data Index: 1492579 (Python Code)
Max Activating Token Index: 189
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0922. Min Act: -0.0001
Data Index: 856055 (C4 (Web Text))
Max Activating Token Index: 576
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0529. Min Act: -0.0001
Data Index: 88080 (C4 (Web Text))
Max Activating Token Index: 776
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.6224. Min Range: -1.6224
Max Act: 0.9083. Min Act: -0.0001
Data Index: 687809 (C4 (Web Text))
Max Activating Token Index: 665
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.6224. Min Range: -1.6224
Max Act: 1.0486. Min Act: -0.0001
Data Index: 1112972 (C4 (Web Text))
Max Activating Token Index: 247
Click toggle to see full text
Truncated
Full Text #19