INDEX

Explanations

keywords related to announcements, confirmations, and teasers in the context of product releases and events

oai_token-act-pair · gpt-3.5-turbo

words and phrases associated with announcements and revealing information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bard

-0.90

mad

-0.89

pet

-0.75

mite

-0.73

ma

-0.73

avery

-0.72

jud

-0.71

pocket

-0.71

avis

-0.71

river

-0.71

POSITIVE LOGITS

 specs

1.16

 unveil

1.07

 teaser

1.05

 specifications

0.90

 announcements

0.90

 hints

0.88

TBA

0.87

 previews

0.87

 leaks

0.86

 wraps

0.83

Activations Density 0.291%

keywords related to announcements, confirmations, and teasers in the context of product releases and events

words and phrases associated with announcements and revealing information

No Comments

No Known Activations