INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names or titles

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

naires

-0.88

rons

-0.83

naire

-0.79

rade

-0.70

DOWN

-0.66

owl

-0.61

ties

-0.61

ropolitan

-0.60

lace

-0.59

 hindsight

-0.59

POSITIVE LOGITS

hler

1.44

hner

1.38

ppel

1.31

lect

1.29

cker

1.26

ffer

1.20

cean

1.07

ger

1.03

hl

1.02

lyn

1.01

Activations Density 0.061%

proper nouns

proper nouns, specifically names or titles

No Comments

No Known Activations