INDEX

Explanations

Anterior cruciate, Scanning electron, Attitudes

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

formatted tabular data with indentation and alignment in technical documents.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The main thing this neuron does is detect numeric sequences (digits and numbers) in 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

-1.79

 PLEASE

-1.57

чики

-1.55

—

-1.55

For

-1.55

oração

-1.52

 Whatever

-1.50

 with

-1.50

országban

-1.48

 menghadapi

-1.47

POSITIVE LOGITS

 хороший

1.59

 olvidado

1.52

jeje

1.51

🫃

1.49

咴

1.46

呔

1.45

躓

1.44

౨

1.43

咘

1.43

中です

1.42

Activations Density 0.004%

Anterior cruciate, Scanning electron, Attitudes

formatted tabular data with indentation and alignment in technical documents.

The main thing this neuron does is detect numeric sequences (digits and numbers) in the text.

No Comments

No Known Activations

Anterior cruciate, Scanning electron, Attitudes

formatted tabular data with indentation and alignment in technical documents.

The main thing this neuron does is detect numeric sequences (digits and numbers) in the text.

No Comments

No Known Activations