INDEX

Explanations

ological

np_max-act · gemini-2.0-flash

This neuron detects the word “methodological” (especially as part of section headings like “Methodological considerations”).

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

 uncomfortable

-0.07

one

-0.07

Apart

-0.06

 معرفی

-0.06

pton

-0.06

jian

-0.06

วไป

-0.06

enames

-0.05

toHaveBeenCalledTimes

-0.05

 같은

-0.05

POSITIVE LOGITS

 chuck

0.07

 encompass

0.07

 mapView

0.07

 meinen

0.07

BlockSize

0.07

 methodology

0.07

yslu

0.07

nb

0.06

что

0.06

Sz

0.06

Activations Density 0.017%

ological

This neuron detects the word “methodological” (especially as part of section headings like “Methodological considerations”).

No Comments

No Known Activations