INDEX

Explanations

UC Berkeley

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

The neuron is detecting named entities—especially proper nouns for places and institutions (e.g. “Berkeley,” “University,” “Center,” etc.).

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

mentions of Berkeley, especially UC Berkeley and its affiliated institutions and local entities.

oai_token-act-pair · gpt-5

Triggered by @jyhe0408

references to Berkeley (the city or university) and related institutional or organizational contexts.

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

Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_34/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

riscoll

-0.80

kwds

-0.79

 שכ

-0.79

来说

-0.76

ciatore

-0.74

來說

-0.73

ác

-0.73

ara

-0.73

verture

-0.73

asibility

-0.73

POSITIVE LOGITS

 Berkeley

1.48

Berkeley

1.27

UCB

1.13

berkeley

1.08

 BERKELEY

0.94

 Haas

0.90

Oval

0.88

 encampment

0.82

Nuke

0.81

California

0.80

Activations Density 0.026%

UC Berkeley

The neuron is detecting named entities—especially proper nouns for places and institutions (e.g. “Berkeley,” “University,” “Center,” etc.).

mentions of Berkeley, especially UC Berkeley and its affiliated institutions and local entities.

references to Berkeley (the city or university) and related institutional or organizational contexts.

No Known Activations