INDEX

Explanations

words related to neutrality or staying impartial

oai_token-act-pair · gpt-3.5-turbo

language related to neutrality and non-bias

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

Hop

-0.87

Mill

-0.81

soDeliveryDate

-0.80

BOOK

-0.77

PER

-0.77

RESULTS

-0.76

URES

-0.76

URE

-0.75

PE

-0.74

MA

-0.74

POSITIVE LOGITS

 neutral

1.05

utral

1.03

izes

0.91

ization

0.88

izing

0.88

izers

0.88

 neutrality

0.87

itÃ©

0.84

ifax

0.83

 buoy

0.83

Activations Density 0.010%

words related to neutrality or staying impartial

language related to neutrality and non-bias

No Comments

No Known Activations