INDEX

Explanations

people or things that are prominent or well-known

oai_token-act-pair · gpt-3.5-turbo

the term "prominent" used in various contexts to describe notable individuals or figures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ceans

-0.75

eries

-0.75

ope

-0.74

yrinth

-0.74

ework

-0.71

aves

-0.68

claimer

-0.68

atson

-0.67

ilk

-0.66

zees

-0.66

POSITIVE LOGITS

 personalities

0.85

 prominent

0.81

 commentator

0.79

 contributor

0.74

 role

0.73

 roles

0.73

 predec

0.72

 figures

0.71

 finan

0.70

 prominently

0.69

Activations Density 0.016%

people or things that are prominent or well-known

the term "prominent" used in various contexts to describe notable individuals or figures

No Comments

No Known Activations