INDEX

Explanations

scientific introductions

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

medical or scientific journal formatting with section headers and reference notations.

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

Triggered by @neilrathi

This neuron fires on the “dru/ g” prefix in pharmaceutical terms (i.e. the beginning of “drug…” tokens).

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

Negative Logits

gridColumn

-0.95

 egiten

-0.90

сери

-0.90

 nyeri

-0.90

 februari

-0.87

வில்

-0.86

retim

-0.85

Климат

-0.84

 Pizza

-0.82

Những

-0.81

POSITIVE LOGITS

paravant

0.98

</u>

0.88

 ayrıca

0.87

 inoltre

0.85

 asume

0.84

 Towels

0.83

giu

0.83

0.82

 Vaugh

0.82

kee

0.82

Activations Density 0.003%

scientific introductions

medical or scientific journal formatting with section headers and reference notations.

This neuron fires on the “dru/ g” prefix in pharmaceutical terms (i.e. the beginning of “drug…” tokens).

No Known Activations

scientific introductions

medical or scientific journal formatting with section headers and reference notations.

This neuron fires on the “dru/ g” prefix in pharmaceutical terms (i.e. the beginning of “drug…” tokens).

No Known Activations