INDEX

Explanations

the word "Not" at the beginning of sentences

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

kamp

-0.77

rift

-0.66

è¦ļéĨĴ

-0.64

stakes

-0.64

ç·

-0.63

creen

-0.61

 NETWORK

-0.60

ixel

-0.59

 avenues

-0.58

FI

-0.56

POSITIVE LOGITS

withstanding

1.42

eworthy

1.30

orious

1.28

ices

1.13

epad

1.12

icably

1.07

icing

1.05

ifications

1.02

 necessarily

1.01

ional

0.96

Activations Density 0.062%

the word "Not" at the beginning of sentences

No Comments

No Known Activations