INDEX

Explanations

sorcerer

np_max-act · gemini-2.0-flash

The neuron strongly activates on the subword tokens composing “sorceress,” i.e. it detects occurrences of that word.

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

 blowjob

-0.07

 Journalism

-0.07

 opat

-0.07

')}}"

-0.06

//"

-0.06

 polite

-0.06

'T

-0.06

 RELEASE

-0.06

ينه

-0.06

IPT

-0.06

POSITIVE LOGITS

COND

0.07

_ACC

0.06

 mage

0.06

bri

0.06

posing

0.06

abis

0.06

 witches

0.06

 purge

0.06

(equalTo

0.06

(+

0.06

Activations Density 0.042%

sorcerer

The neuron strongly activates on the subword tokens composing “sorceress,” i.e. it detects occurrences of that word.

No Comments

No Known Activations