Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 950 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 3.2029. Min Range: -3.2029
Max Act: 3.2029. Min Act: -0.1700
Data Index: 1352891 (C4 (Web Text))
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.9188. Min Act: -0.1700
Data Index: 646634 (C4 (Web Text))
Max Activating Token Index: 947
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.9029. Min Act: -0.1700
Data Index: 1570857 (Python Code)
Max Activating Token Index: 316
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.9188. Min Act: -0.1700
Data Index: 224397 (C4 (Web Text))
Max Activating Token Index: 839
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8871. Min Act: -0.1700
Data Index: 550046 (C4 (Web Text))
Max Activating Token Index: 680
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8712. Min Act: -0.1699
Data Index: 33895 (C4 (Web Text))
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8871. Min Act: -0.1700
Data Index: 1647870 (Python Code)
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8871. Min Act: -0.1699
Data Index: 1459390 (Python Code)
Max Activating Token Index: 399
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.9188. Min Act: -0.1700
Data Index: 1112570 (C4 (Web Text))
Max Activating Token Index: 367
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8712. Min Act: -0.1700
Data Index: 298576 (C4 (Web Text))
Max Activating Token Index: 458
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8394. Min Act: -0.1700
Data Index: 420475 (C4 (Web Text))
Max Activating Token Index: 353
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.7757. Min Act: -0.1700
Data Index: 210220 (C4 (Web Text))
Max Activating Token Index: 487
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8394. Min Act: -0.1700
Data Index: 715701 (C4 (Web Text))
Max Activating Token Index: 630
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8871. Min Act: -0.1700
Data Index: 1290454 (C4 (Web Text))
Max Activating Token Index: 775
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.9029. Min Act: -0.1700
Data Index: 1607743 (Python Code)
Max Activating Token Index: 1023
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8235. Min Act: -0.1699
Data Index: 841634 (C4 (Web Text))
Max Activating Token Index: 736
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8871. Min Act: -0.1700
Data Index: 805902 (C4 (Web Text))
Max Activating Token Index: 818
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8553. Min Act: -0.1700
Data Index: 1645866 (Python Code)
Max Activating Token Index: 635
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8394. Min Act: -0.1699
Data Index: 266427 (C4 (Web Text))
Max Activating Token Index: 710
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 3.2029. Min Range: -3.2029
Max Act: 2.8235. Min Act: -0.1700
Data Index: 810948 (C4 (Web Text))
Max Activating Token Index: 1021
Click toggle to see full text
Truncated
Full Text #19