INDEX

Explanations

phrases related to shared characteristics or beliefs among a group of individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals or groups characterized as "like-minded."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ILA

-0.71

Recomm

-0.69

adium

-0.69

PsyNetMessage

-0.69

osponsors

-0.69

adier

-0.69

IVER

-0.67

ç«

-0.66

ARDS

-0.65

CHA

-0.64

POSITIVE LOGITS

ness

1.39

edly

1.23

nesses

1.08

ly

1.03

minded

0.98

edIn

0.83

 minded

0.79

quartered

0.78

cipled

0.77

smanship

0.77

Activations Density 0.025%

phrases related to shared characteristics or beliefs among a group of individuals

references to individuals or groups characterized as "like-minded."

No Comments

No Known Activations