INDEX

Explanations

the word "absent" or phrases related to absence

oai_token-act-pair · gpt-3.5-turbo

the presence and frequency of the word "absent"

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ipel

-0.91

Ocean

-0.81

ãĥī

-0.76

emetery

-0.75

ombs

-0.74

ricanes

-0.72

zees

-0.71

flo

-0.71

inical

-0.70

ounters

-0.70

POSITIVE LOGITS

minded

0.90

 minded

0.83

soever

0.73

wich

0.71

lihood

0.71

alus

0.70

volent

0.65

 Mondays

0.64

etheless

0.64

lust

0.64

Activations Density 0.027%

the word "absent" or phrases related to absence

the presence and frequency of the word "absent"

No Known Activations