INDEX

Explanations

medical and technical terms

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

technical terminology related to specific domains, particularly software libraries, medical scales, and assessment tools.

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

The neuron fires on capitalized content words that are part of proper names, titles, and acronyms (e.g. names of tools, questionnaires, scales, features, or studies).

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

 hongos

-1.59

 couverte

-1.55

colori

-1.55

に行って

-1.55

 uñas

-1.53

で使

-1.51

 feroit

-1.51

 faldas

-1.50

 terminée

-1.48

 tomba

-1.48

POSITIVE LOGITS

in

1.71

1.62

1.55

1.53

for

1.49

1.45

それも

1.43

1.42

つまり

1.41

Activations Density 0.221%

medical and technical terms

technical terminology related to specific domains, particularly software libraries, medical scales, and assessment tools.

The neuron fires on capitalized content words that are part of proper names, titles, and acronyms (e.g. names of tools, questionnaires, scales, features, or studies).

No Comments

No Known Activations

medical and technical terms

technical terminology related to specific domains, particularly software libraries, medical scales, and assessment tools.

The neuron fires on capitalized content words that are part of proper names, titles, and acronyms (e.g. names of tools, questionnaires, scales, features, or studies).

No Comments

No Known Activations