INDEX

Explanations

words related to inappropriate behavior or content

oai_token-act-pair · gpt-3.5-turbo

terms related to inappropriate or improper behavior and conduct

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ership

-0.96

utra

-0.86

reens

-0.78

thood

-0.77

glas

-0.77

grass

-0.77

planes

-0.76

winner

-0.76

stone

-0.75

abiding

-0.75

POSITIVE LOGITS

ities

0.88

 interference

0.83

 undermin

0.82

 behaviour

0.79

 interfering

0.76

 inappropriate

0.75

 improper

0.75

 behavior

0.74

 Parenthood

0.73

 misuse

0.72

Activations Density 0.049%

words related to inappropriate behavior or content

terms related to inappropriate or improper behavior and conduct

No Comments

No Known Activations