INDEX

Explanations

The neuron activates on numerical tokens—years, measurements, large numbers or decimal values.

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

passages about organizations, institutions, or people with founding dates or career milestones in the late 20th century (particularly 1970s-2010s).

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

references to concrete factual details and named entities, such as dates, quantities, organizations, titles, places, and life or career milestones.

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

 সপ্তাহের

0.66

বারের

0.63

今まで

0.61

 prochains

0.61

 vagy

0.60

antiti

0.60

 lesz

0.60

 যতই

0.59

ått

0.59

<unused6>

0.59

POSITIVE LOGITS

 בשנת

3.45

 vuonna

3.13

ในปี

3.06

 năm

1.82

 летом

1.80

 circa

1.80

 ২০১৩

1.78

 سنة

1.77

 tahun

1.73

 през

1.71

Activations Density 1.242%

The neuron activates on numerical tokens—years, measurements, large numbers or decimal values.

passages about organizations, institutions, or people with founding dates or career milestones in the late 20th century (particularly 1970s-2010s).

references to concrete factual details and named entities, such as dates, quantities, organizations, titles, places, and life or career milestones.

No Comments

No Known Activations

The neuron activates on numerical tokens—years, measurements, large numbers or decimal values.

passages about organizations, institutions, or people with founding dates or career milestones in the late 20th century (particularly 1970s-2010s).

references to concrete factual details and named entities, such as dates, quantities, organizations, titles, places, and life or career milestones.

No Comments

No Known Activations