INDEX

Explanations

to be

np_max-act · gemini-2.0-flash

The neuron fires on phrasing that signals relevance—particularly the “is relevant to” construction.

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

BUT

-0.07

Sugar

-0.06

 while

-0.06

 when

-0.06

 whose

-0.06

extField

-0.06

how

-0.06

But

-0.06

who

-0.06

ata

-0.06

POSITIVE LOGITS

ync

0.07

 escap

0.06

 elasticity

0.06

描

0.06

{},↵

0.06

ind

0.06

지막

0.06

ึ้

0.06

cao

0.06

overflow

0.06

Activations Density 0.096%

to be

The neuron fires on phrasing that signals relevance—particularly the “is relevant to” construction.

No Comments

No Known Activations