INDEX

Explanations

executing code and file operations

np_acts-logits-general · gemini-2.5-flash-lite

periods that follow method calls or object property access in code.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

The neuron fires on the dot “.” that signals a method or attribute access (object.member) in code.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

of

-1.73

who

-1.67

 their

-1.58

–

-1.41

 each

-1.34

mathrm

-1.33

-1.26

 każdy

-1.22

 kotlinx

-1.21

每次

-1.20

POSITIVE LOGITS

素晴

1.42

菇

1.37

untergang

1.37

叹息

1.34

棰

1.32

kilometer

1.30

collant

1.28

וקה

1.24

奇跡

1.23

ihnachts

1.22

Activations Density 0.026%

executing code and file operations

periods that follow method calls or object property access in code.

The neuron fires on the dot “.” that signals a method or attribute access (object.member) in code.

No Comments

No Known Activations

executing code and file operations

periods that follow method calls or object property access in code.

The neuron fires on the dot “.” that signals a method or attribute access (object.member) in code.

No Comments

No Known Activations