INDEX

Explanations

function calls and updates

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

The neuron primarily responds to literal numbers (numeric digits) in the text.

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

lines of code that contain method calls with `self` or object references followed by method invocations.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/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

 konsep

-1.45

 kosme

-1.41

 meil

-1.38

 rése

-1.37

 amerik

-1.33

 aéro

-1.28

 kalori

-1.27

 sade

-1.27

 foton

-1.27

 koordin

-1.25

POSITIVE LOGITS

 make

1.40

 then

1.36

 will

1.27

 changed

1.26

and

1.26

 became

1.23

as

1.20

set

1.20

 decide

1.17

 ended

1.16

Activations Density 0.005%

function calls and updates

The neuron primarily responds to literal numbers (numeric digits) in the text.

lines of code that contain method calls with `self` or object references followed by method invocations.

No Comments

No Known Activations

function calls and updates

The neuron primarily responds to literal numbers (numeric digits) in the text.

lines of code that contain method calls with `self` or object references followed by method invocations.

No Comments

No Known Activations