INDEX

Explanations

allest

np_max-act · gemini-2.0-flash

This neuron activates on words that indicate quantitative changes, such as “increased,” “decreased,” and their variants (e.g., “increased by,” “decreased from”).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 futile

-0.08

ImageData

-0.06

expiry

-0.06

guns

-0.06

 ánh

-0.06

 horr

-0.06

ListItemText

-0.06

흔

-0.06

 Frances

-0.06

StdString

-0.06

POSITIVE LOGITS

пос

0.07

sui

0.07

)+"

0.07

个

0.07

 portfolios

0.07

 innovative

0.07

 полит

0.06

改

0.06

_DATA

0.06

_rights

0.06

Activations Density 0.004%

allest

This neuron activates on words that indicate quantitative changes, such as “increased,” “decreased,” and their variants (e.g., “increased by,” “decreased from”).

No Comments

No Known Activations