INDEX

Explanations

references to future plans or intentions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Cause

-0.70

 ILCS

-0.68

inas

-0.67

osi

-0.65

INTON

-0.64

inion

-0.64

avery

-0.64

 Stain

-0.62

 Trophy

-0.62

weed

-0.61

POSITIVE LOGITS

emaker

0.84

etary

0.83

etting

0.81

isphere

0.79

 Parenthood

0.76

obs

0.76

 ahead

0.75

etter

0.75

 horizon

0.73

for

0.72

Activations Density 0.054%

references to future plans or intentions

No Comments

No Known Activations