INDEX

Explanations

specific column medic recent

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

first words or beginnings of statements in question-answer forums.

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

This neuron fires on tokens that begin a new line or block—i.e. the very first non‐whitespace word or symbol at the start of a line.

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

Not in Any Lists

No Comments

Negative Logits



-1.48

 provides

-1.44

Rumah

-1.41

Suka

-1.38

我们的

-1.38

تاریخ

-1.38

 szczeg

-1.37

Paket

-1.36

ၜ

-1.35

Groet

-1.34

POSITIVE LOGITS

</strong>

2.20

</b>

1.80

 Either

1.59

 podendo

1.52



1.51

…

1.49

to

1.48

could

1.48

These

1.48

Activations Density 0.071%

specific column medic recent

first words or beginnings of statements in question-answer forums.

This neuron fires on tokens that begin a new line or block—i.e. the very first non‐whitespace word or symbol at the start of a line.

No Comments

No Known Activations

specific column medic recent

first words or beginnings of statements in question-answer forums.

This neuron fires on tokens that begin a new line or block—i.e. the very first non‐whitespace word or symbol at the start of a line.

No Comments

No Known Activations