INDEX

Explanations

The neuron activates on occurrences of the word “invasive” (including as part of “non-invasive”).

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

noninvasive or invasive descriptors

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

The neuron identifies terms with the prefix "non-" followed by adjectives, particularly in medical and technical contexts.

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

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

えっ

-1.38

有两

-1.29

羕

-1.27

ⵌ

-1.25

iſter

-1.22

ޏ

-1.20

焌

-1.18

 estadounid

-1.16

ありますか

-1.15

 romántico

-1.14

POSITIVE LOGITS

of

1.30

："

1.11

1.09

↵

1.09

이드

1.05

Further

1.04

 They

1.03

de

1.03

1.01

es

0.98

Activations Density 0.062%

The neuron activates on occurrences of the word “invasive” (including as part of “non-invasive”).

noninvasive or invasive descriptors

The neuron identifies terms with the prefix "non-" followed by adjectives, particularly in medical and technical contexts.

No Comments

No Known Activations

The neuron activates on occurrences of the word “invasive” (including as part of “non-invasive”).

noninvasive or invasive descriptors

The neuron identifies terms with the prefix "non-" followed by adjectives, particularly in medical and technical contexts.

No Comments

No Known Activations