INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ModLoader

-0.75

DonaldTrump

-0.59

 retina

-0.59

 Perception

-0.55

Jose

-0.52

 advertisers

-0.52

 grat

-0.52

muc

-0.50

ILA

-0.50

LEASE

-0.50

POSITIVE LOGITS

ovich

0.76

sson

0.76

quez

0.76

monds

0.71

plin

0.71

ensen

0.70

uez

0.69

aka

0.68

kson

0.68

zinski

0.68

Activations Density 0.304%

names of individuals

proper nouns, specifically names of individuals

No Comments

No Known Activations