INDEX

Explanations

hyphenated phrases used to negate or contrast a concept

oai_token-act-pair · gpt-3.5-turbo

negative and contradictory phrases

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

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

ãĤ¼ãĤ¦ãĤ¹

-0.68

isine

-0.65

Ĥİ

-0.63

ogy

-0.59

 TAMADRA

-0.59

ixel

-0.58

illon

-0.58

 showc

-0.57

assetsadobe

-0.55

ellar

-0.55

POSITIVE LOGITS

yet

1.02

orious

1.00

quite

0.99

withstanding

0.88

icably

0.87

icable

0.86

 terribly

0.85

eworthy

0.84

 Quite

0.81

 necessarily

0.78

Activations Density 0.086%

hyphenated phrases used to negate or contrast a concept

negative and contradictory phrases

No Comments

No Known Activations