INDEX

Explanations

names of people, places, and organizations

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to organizations, teams, or governmental bodies

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

untarily

-0.69

 consent

-0.68

onym

-0.65

 whisper

-0.64

uphem

-0.64

ezvous

-0.63

xtap

-0.62

 Consent

-0.61

heid

-0.60

akespeare

-0.60

POSITIVE LOGITS

 sake

0.97

 purposes

0.81

 coffers

0.78

 lovers

0.74

!/

0.73

ãĤ®

0.72

 considering

0.71

é¾įåĸļå£«

0.70

"}],"

0.67

 morale

0.67

Activations Density 0.499%

names of people, places, and organizations

proper nouns related to organizations, teams, or governmental bodies

No Comments

No Known Activations