INDEX

Explanations

prominent figures or entities mentioned in a document

oai_token-act-pair · gpt-3.5-turbo

terms related to high-profile individuals and their significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

uel

-0.87

©¶æ

-0.83

amar

-0.78

ilk

-0.75

aton

-0.74

prep

-0.74

ueller

-0.73

ramid

-0.73

onz

-0.72

yrinth

-0.71

POSITIVE LOGITS

 exceptions

1.02

 endors

1.02

 bearer

1.01

 alumni

1.00

 examples

0.98

 exception

0.96

 personalities

0.96

 landmarks

0.94

 occasions

0.89

 names

0.89

Activations Density 0.076%

prominent figures or entities mentioned in a document

terms related to high-profile individuals and their significance

No Comments

No Known Activations