INDEX

Explanations

intrigue

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

The neuron detects occurrences of the root “intrigu” in words like “intrigue,” “intriguing,” and “intrigued.”

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

mentions of intrigue—detecting the concept in noun, verb, and adjective forms indicating something is intriguing or someone is intrigued.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

the word "intriguing" or its variations (intrigue, intrigued).

oai_token-act-pair · claude-4-5-sonnet 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

雰

-3.00

羮

-2.70

).”

-2.59

注意事

-2.48

-2.47

 necessario

-2.42

with

-2.38

柽

-2.38

es

-2.36

actéristi

-2.36

POSITIVE LOGITS

2.59

翃

2.52

毳

2.23

 ใน

2.22

许久

2.20

2.13

medriver

2.08

な感じ

2.08

тій

2.03

 ofrecemos

1.99

Activations Density 0.001%

intrigue

The neuron detects occurrences of the root “intrigu” in words like “intrigue,” “intriguing,” and “intrigued.”

mentions of intrigue—detecting the concept in noun, verb, and adjective forms indicating something is intriguing or someone is intrigued.

the word "intriguing" or its variations (intrigue, intrigued).

No Comments

No Known Activations

intrigue

The neuron detects occurrences of the root “intrigu” in words like “intrigue,” “intriguing,” and “intrigued.”

mentions of intrigue—detecting the concept in noun, verb, and adjective forms indicating something is intriguing or someone is intrigued.

the word "intriguing" or its variations (intrigue, intrigued).

No Comments

No Known Activations