INDEX

Explanations

of

np_max-act · gemini-2.0-flash

The neuron fires on quantitative magnitude terms expressing proportions or amounts (e.g. “substantial,” “amount,” “large part,” “most,” “lost,” “less”).

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

urd

-0.07

 поруш

-0.06

 shader

-0.06

 buffering

-0.06

 replaced

-0.06

 '',↵

-0.06

 як

-0.06

DownLatch

-0.06

 bags

-0.06

.borrow

-0.06

POSITIVE LOGITS

 düzenlem

0.07

/ay

0.07

sek

0.07

�

0.07

 contempl

0.06

 date

0.06

/foo

0.06

 California

0.06

 brightest

0.06

KeyUp

0.06

Activations Density 0.028%

of

The neuron fires on quantitative magnitude terms expressing proportions or amounts (e.g. “substantial,” “amount,” “large part,” “most,” “lost,” “less”).

No Comments

No Known Activations