INDEX

Explanations

mentions of the name "Kat" with various additional characters

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific person named Kat

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

cffff

-0.85

CLASSIFIED

-0.80

LEASE

-0.79

uploads

-0.72

 confir

-0.71

prise

-0.69

 contraceptives

-0.67

lict

-0.66

 margins

-0.65

interstitial

-0.64

POSITIVE LOGITS

anas

1.17

Kat

1.03

usha

0.98

kat

0.96

apesh

0.89

ney

0.88

Kat

0.87

zen

0.85

oton

0.84

ana

0.84

Activations Density 0.012%

mentions of the name "Kat" with various additional characters

mentions of a specific person named Kat

No Comments

No Known Activations