INDEX

Explanations

proper nouns, likely focusing on names of people

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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.83

atically

-0.66

-0.65

DonaldTrump

-0.64

etheless

-0.63

 Italians

-0.63

cknowled

-0.63

respective

-0.62

ãĤ¸

-0.62

stood

-0.62

POSITIVE LOGITS

III

0.89

Jr

0.88

strom

0.83

aka

0.81

zinski

0.79

Sr

0.79

gren

0.76

hurst

0.75

(@

0.74

wine

0.74

Activations Density 0.290%

proper nouns, likely focusing on names of people

proper nouns, specifically names of individuals

No Comments

No Known Activations