INDEX

Explanations

justifications or support

np_max-act · gemini-2.0-flash

The neuron fires on abstract, formal nouns (e.g. “expense,” “basis”) typically introducing “of”‐phrases in technical or legal‐style writing.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.07

 Presbyterian

-0.07

 odense

-0.07

 Southern

-0.07

would

-0.07

odo

-0.07

 profound

-0.07

 consulting

-0.07

lucent

-0.06

high

-0.06

POSITIVE LOGITS

 behalf

0.08

 través

0.08

//{↵

0.07

(username

0.07

vs

0.06

 através

0.06

:item

0.06

Thumb

0.06

'name

0.06

 dims

0.06

Activations Density 0.095%

justifications or support

The neuron fires on abstract, formal nouns (e.g. “expense,” “basis”) typically introducing “of”‐phrases in technical or legal‐style writing.

No Comments

No Known Activations

justifications or support

The neuron fires on abstract, formal nouns (e.g. “expense,” “basis”) typically introducing “of”‐phrases in technical or legal‐style writing.

No Comments

No Known Activations