INDEX

Explanations

fraternity-related terms

oai_token-act-pair · gpt-3.5-turbo

references to fraternities

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

stroke

-0.86

 Kers

-0.74

over

-0.71

 Nanto

-0.70

esm

-0.70

 Homs

-0.69

Alz

-0.69

ibrary

-0.68

stro

-0.66

 Overs

-0.66

POSITIVE LOGITS

 fraternity

1.42

Phi

1.31

ority

1.07

aternity

1.04

 alumni

1.03

 Sigma

1.03

 Kappa

0.98

sor

0.94

atern

0.91

 dues

0.90

Activations Density 0.056%

fraternity-related terms

references to fraternities

No Comments

No Known Activations