INDEX

Explanations

connecting through

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

the word "through" followed by noun phrases, especially in descriptive or instructional contexts.

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

The neuron fires on numeric tokens—especially decimal values or measurements—within the text.

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

はこの

-1.31

 only

-1.23

總是

-1.23

 among

-1.17

雖

-1.15

 этот

-1.14

もともと

-1.13

饽

-1.12

忏

-1.12

ideon

-1.12

POSITIVE LOGITS

 through

2.02

Through

1.70

through

1.52

 prostřed

1.45

 através

1.43

 melalui

1.42

via

1.41

 через

1.39

༵

1.36

 Praze

1.34

Activations Density 0.220%

connecting through

the word "through" followed by noun phrases, especially in descriptive or instructional contexts.

The neuron fires on numeric tokens—especially decimal values or measurements—within the text.

No Comments

No Known Activations

connecting through

the word "through" followed by noun phrases, especially in descriptive or instructional contexts.

The neuron fires on numeric tokens—especially decimal values or measurements—within the text.

No Comments

No Known Activations