INDEX

Explanations

where to

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

The neuron fires strongly on occurrences of the question word “Where.”

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

the word "Where" at the beginning of questions.

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

questions that seek location or source information (location-seeking interrogatives at the start of a sentence).

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

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ことが多い

0.74

 menyelesaikan

0.66

觴

0.63

 solve

0.61

 раствор

0.61

ರನ್ನು

0.61

 existencia

0.59

を実現

0.59

が集

0.58

 solves

0.58

POSITIVE LOGITS

 στην

0.71

Đi

0.69

〢

0.68

 εδώ

0.68

fores

0.66

 nella

0.66

 else

0.64

hörigen

0.63

đi

0.63

けん

0.63

Activations Density 0.121%

where to

The neuron fires strongly on occurrences of the question word “Where.”

the word "Where" at the beginning of questions.

questions that seek location or source information (location-seeking interrogatives at the start of a sentence).

No Comments

No Known Activations

where to

The neuron fires strongly on occurrences of the question word “Where.”

the word "Where" at the beginning of questions.

questions that seek location or source information (location-seeking interrogatives at the start of a sentence).

No Comments

No Known Activations