INDEX

Explanations

and flag instances of toxic terms or descriptions

oai_token-act-pair · gpt-3.5-turbo

references to toxic substances and their effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

gain

-0.83

hung

-0.76

FORE

-0.74

AUT

-0.73

bler

-0.71

ploma

-0.70

quart

-0.70

BO

-0.70

stand

-0.70

Month

-0.69

POSITIVE LOGITS

 poisoning

1.00

ologist

0.99

 toxic

0.95

ologically

0.95

 substances

0.91

ological

0.90

 fumes

0.89

 masculinity

0.89

ologists

0.88

ology

0.86

Activations Density 0.008%

and flag instances of toxic terms or descriptions

references to toxic substances and their effects

No Comments

No Known Activations