INDEX

Explanations

mentions of prominent individuals or figures

oai_token-act-pair · gpt-3.5-turbo

instances of the word "prominent" used to describe individuals or entities with significant influence or recognition

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yrinth

-0.77

eries

-0.73

ope

-0.73

Phones

-0.72

ework

-0.70

ceans

-0.68

ilk

-0.68

thur

-0.67

claimer

-0.66

ectar

-0.66

POSITIVE LOGITS

 personalities

0.86

 prominent

0.74

 roles

0.73

 role

0.72

 figures

0.72

 commentator

0.70

 endors

0.68

 personality

0.68

 landmarks

0.67

 contributor

0.67

Activations Density 0.029%

mentions of prominent individuals or figures

instances of the word "prominent" used to describe individuals or entities with significant influence or recognition

No Comments

No Known Activations