INDEX

Explanations

references to specific media titles, franchises, and properties

oai_token-act-pair · gpt-3.5-turbo

specific names and titles related to entertainment and media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

roman

-0.85

ord

-0.80

hai

-0.78

adem

-0.77

angel

-0.75

opol

-0.72

avier

-0.72

gaard

-0.72

ia

-0.71

atom

-0.71

POSITIVE LOGITS

FORM

1.33

LET

1.29

OUT

1.28

NING

1.27

ISM

1.27

ABLE

1.27

TERN

1.26

IST

1.25

FOR

1.24

OUS

1.24

Activations Density 0.255%

references to specific media titles, franchises, and properties

specific names and titles related to entertainment and media

No Comments

No Known Activations