INDEX

Explanations

mentions of a specific name: "McCormick"

oai_token-act-pair · gpt-3.5-turbo

references to the name "McCorm" and its variations in context

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

holes

-0.83

arist

-0.80

hole

-0.78

terior

-0.75

gypt

-0.74

tymology

-0.73

izons

-0.71

uyomi

-0.71

 nomine

-0.71

galitarian

-0.71

POSITIVE LOGITS

artney

0.91

 McCorm

0.86

atcher

0.85

inelli

0.83

 McCabe

0.83

onduct

0.82

riages

0.73

ONEY

0.73

onductor

0.72

urdy

0.71

Activations Density 0.021%

mentions of a specific name: "McCormick"

references to the name "McCorm" and its variations in context

No Comments

No Known Activations