INDEX

Explanations

words related to political figures and international relations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating interactive components or features related to software or technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

angelo

-0.70

zona

-0.69

Cola

-0.66

 citiz

-0.63

ICAN

-0.55

CLASSIFIED

-0.55

 Veter

-0.55

ħĭ

-0.54

 ACTIONS

-0.53

ccording

-0.53

POSITIVE LOGITS

inki

0.61

Balt

0.53

acht

0.50

ittens

0.49

usercontent

0.48

 Berman

0.47

andise

0.47

iland

0.47

 overwhelm

0.46

espie

0.46

Activations Density 0.931%

words related to political figures and international relations

phrases indicating interactive components or features related to software or technology

No Comments

No Known Activations