INDEX

Explanations

references to "Stanford" University

oai_token-act-pair · gpt-3.5-turbo

mentions of "Stanford"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

alez

-0.77

unal

-0.75

cedented

-0.75

usher

-0.72

merce

-0.69

hops

-0.67

 substant

-0.66

unct

-0.66

olitan

-0.66

ramid

-0.66

POSITIVE LOGITS

 Stanford

0.98

 Institution

0.96

 University

0.89

 Cardinal

0.85

thal

0.84

sburg

0.80

 Alto

0.77

 Haram

0.76

 Olson

0.74

 Hills

0.73

Activations Density 0.009%

references to "Stanford" University

mentions of "Stanford"

No Comments

No Known Activations