INDEX

Explanations

the disease

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

mentions of diseases or medical conditions within scientific/medical text.

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

The neuron is acting as a “high-frequency token” detector, responding strongly to common function words and punctuation (e.g. “the,” “is,” “a,” commas, parentheses) that occur ubiquitously in biomedical‐style text.

oai_token-act-pair · o4-mini 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

Not in Any Lists

No Comments

Negative Logits

ﺰ

-0.77

ylim

-0.76

Abilities

-0.76

ariatric

-0.76

anı

-0.73

 virkelig

-0.73

ſe

-0.73

giti

-0.72

legung

-0.71

ългар

-0.71

POSITIVE LOGITS

 disease

5.41

disease

3.80

 condition

3.50

 Disease

3.38

 illness

3.36

 DISEASE

3.22

Disease

3.11

 disorder

2.98

 diseases

2.86

 enfermedad

2.67

Activations Density 0.289%

the disease

mentions of diseases or medical conditions within scientific/medical text.

The neuron is acting as a “high-frequency token” detector, responding strongly to common function words and punctuation (e.g. “the,” “is,” “a,” commas, parentheses) that occur ubiquitously in biomedical‐style text.

No Comments

No Known Activations

the disease

mentions of diseases or medical conditions within scientific/medical text.

The neuron is acting as a “high-frequency token” detector, responding strongly to common function words and punctuation (e.g. “the,” “is,” “a,” commas, parentheses) that occur ubiquitously in biomedical‐style text.

No Comments

No Known Activations