INDEX

Explanations

surface

np_max-act · gemini-2.0-flash

The neuron strongly activates on the token “surface,” i.e. it’s detecting mentions of “surface” in the text.

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

methods or techniques related to semiconductor manufacturing.

oai_token-act-pair · gpt-4o-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

 telegram

-0.06

 lazy

-0.06

 feeder

-0.06

_Width

-0.06

 misguided

-0.06

 Pamela

-0.06

.ColumnStyles

-0.06

 boilers

-0.06

 migrants

-0.06

ень

-0.06

POSITIVE LOGITS

 демон

0.06

πλα

0.06

法

0.06

StackTrace

0.06

bike

0.06

(Local

0.06

 外部リンク

0.06

/respond

0.06

 vết

0.06

.sales

0.06

Activations Density 0.005%

surface

The neuron strongly activates on the token “surface,” i.e. it’s detecting mentions of “surface” in the text.

methods or techniques related to semiconductor manufacturing.

No Comments

No Known Activations