INDEX

Explanations

words related to darkness or negative connotations

oai_token-act-pair · gpt-3.5-turbo

references to darkness and negative connotations

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

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

agine

-0.76

utic

-0.75

Dur

-0.71

yip

-0.71

PsyNetMessage

-0.71

cule

-0.71

ovo

-0.69

utable

-0.69

raltar

-0.68

kson

-0.66

POSITIVE LOGITS

ening

1.31

ened

1.13

horse

0.96

eners

0.92

room

0.92

haired

0.89

 clouds

0.89

moon

0.87

hound

0.87

ener

0.85

Activations Density 0.024%

words related to darkness or negative connotations

references to darkness and negative connotations

No Comments

No Known Activations