INDEX

Explanations

words related to obscuring or concealing something

oai_token-act-pair · gpt-3.5-turbo

terms related to obscenity and obscured elements in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

GD

-0.73

DF

-0.71

DS

-0.66

WARD

-0.66

 Clover

-0.64

DIT

-0.64

 Berk

-0.64

EMOTE

-0.64

DR

-0.64

CTV

-0.63

POSITIVE LOGITS

ĸļ

1.19

Û

1.16

 obsc

0.97

ured

0.91

urities

0.87

eki

0.87

uring

0.84

ures

0.82

uration

0.81

 offend

0.78

Activations Density 0.011%

words related to obscuring or concealing something

terms related to obscenity and obscured elements in the text

No Comments

No Known Activations