INDEX

Explanations

references to the political group "Brotherhood"

oai_token-act-pair · gpt-3.5-turbo

references to the Muslim Brotherhood, particularly in the context of political events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

acly

-0.76

ãĤª

-0.73

 Glasgow

-0.70

abases

-0.69

amine

-0.66

EAR

-0.66

pport

-0.65

ãĥ¯ãĥ³

-0.65

othy

-0.65

ãĤ©

-0.65

POSITIVE LOGITS

 Brotherhood

0.85

 affiliates

0.83

ollah

0.75

 members

0.74

alions

0.71

 informants

0.71

undy

0.70

hemy

0.70

 infiltr

0.70

 affiliated

0.69

Activations Density 0.020%

references to the political group "Brotherhood"

references to the Muslim Brotherhood, particularly in the context of political events

No Comments

No Known Activations