INDEX

Explanations

people's names and occupations

oai_token-act-pair · gpt-3.5-turbo

references to various professional roles and their affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

adelphia

-0.90

bucks

-0.84

CLOSE

-0.81

Alabama

-0.76

Kansas

-0.76

Seattle

-0.74

Oak

-0.74

isSpecialOrderable

-0.74

TPP

-0.72

lake

-0.71

POSITIVE LOGITS

 Alexand

1.42

 Nikol

1.42

 Andreas

1.40

 Igor

1.39

 Sergei

1.36

 Constantin

1.35

 Tomas

1.35

 Dmitry

1.34

 Giovanni

1.32

 Pavel

1.32

Activations Density 0.211%

people's names and occupations

references to various professional roles and their affiliations

No Comments

No Known Activations