INDEX

Explanations

phrases indicating explanation or cause

oai_token-act-pair · gpt-3.5-turbo

phrases indicating reasons or justifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

chn

-0.85

OLOGY

-0.82

busters

-0.72

astics

-0.71

alsa

-0.70

thumbnails

-0.70

 ILCS

-0.70

framework

-0.70

forts

-0.70

tle

-0.69

POSITIVE LOGITS

why

1.22

 preferring

1.12

 wanting

1.02

 inaction

0.97

 optimism

0.97

 choosing

0.96

 dismissing

0.96

 excluding

0.95

 cance

0.95

 believing

0.95

Activations Density 0.082%

phrases indicating explanation or cause

phrases indicating reasons or justifications

No Comments

No Known Activations