INDEX

Explanations

classification codes and disease stages

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

alphanumeric classification codes or designations, such as zoning codes, hazard classifications, and disease staging.

oai_token-act-pair · claude-3-7-sonnet-20250219

Triggered by @neilrathi

The neuron detects formal property‐use classification labels (e.g. “Class A1 (Retail)”).

oai_token-act-pair · o4-mini

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

Negative Logits

 quoque

-1.61

ことになる

-1.54

the

-1.44

搧

-1.40

鳚

-1.40

 quidem

-1.40

 alicante

-1.39

of

-1.33

 薬局

-1.26

 supri

-1.26

POSITIVE LOGITS

 zbyt

1.73

ministrazione

1.35

 Вас

1.30

 séjours

1.28

 bahía

1.27

 membuat

1.25

 вместо

1.24

multicolumn

1.21

 style

1.20

 wichtig

1.20

Activations Density 0.085%

classification codes and disease stages

alphanumeric classification codes or designations, such as zoning codes, hazard classifications, and disease staging.

The neuron detects formal property‐use classification labels (e.g. “Class A1 (Retail)”).

No Known Activations

classification codes and disease stages

alphanumeric classification codes or designations, such as zoning codes, hazard classifications, and disease staging.

The neuron detects formal property‐use classification labels (e.g. “Class A1 (Retail)”).

No Known Activations