INDEX

Explanations

medical and technical terms

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

medical and clinical terminology, particularly disease conditions and medical classifications.

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

The neuron consistently fires on longer descriptive adjectives—especially those ending in -ic, -ical, -ive, etc.—that qualify or characterize concepts.

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

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

and

-1.17

こいつ

-1.10

 canola

-1.06

ttin

-1.03

 дѣ

-0.99

plained

-0.98

 centrifug

-0.97

ueba

-0.97

ike

-0.97

hotos

-0.96

POSITIVE LOGITS

 time

1.30

of

1.13

all

1.09

 level

1.09

 over

1.02

so

1.02

 when

0.98

as

0.98

 give

0.96

 dazzling

0.96

Activations Density 0.103%

medical and technical terms

medical and clinical terminology, particularly disease conditions and medical classifications.

The neuron consistently fires on longer descriptive adjectives—especially those ending in -ic, -ical, -ive, etc.—that qualify or characterize concepts.

No Comments

No Known Activations

medical and technical terms

medical and clinical terminology, particularly disease conditions and medical classifications.

The neuron consistently fires on longer descriptive adjectives—especially those ending in -ic, -ical, -ive, etc.—that qualify or characterize concepts.

No Comments

No Known Activations

medical and technical terms

medical and clinical terminology, particularly disease conditions and medical classifications.

The neuron consistently fires on longer descriptive adjectives—especially those ending in ­-ic, ­-ical, ­-ive, etc.—that qualify or characterize concepts.

No Comments

No Known Activations

medical and technical terms

medical and clinical terminology, particularly disease conditions and medical classifications.

The neuron consistently fires on longer descriptive adjectives—especially those ending in ­-ic, ­-ical, ­-ive, etc.—that qualify or characterize concepts.

No Comments

No Known Activations

The neuron consistently fires on longer descriptive adjectives—especially those ending in -ic, -ical, -ive, etc.—that qualify or characterize concepts.

The neuron consistently fires on longer descriptive adjectives—especially those ending in -ic, -ical, -ive, etc.—that qualify or characterize concepts.