INDEX

Explanations

proper nouns or names mentioned in news articles

oai_token-act-pair · gpt-3.5-turbo

references to notable events or accomplishments

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

 citiz

-0.93

 tiss

-0.91

 tram

-0.83

 screwed

-0.82

 proport

-0.82

 casc

-0.82

 stray

-0.80

rul

-0.80

 crunch

-0.79

 hydrogen

-0.78

POSITIVE LOGITS

His

2.26

Born

2.07

He

1.99

During

1.66

Prior

1.59

Known

1.58

Contents

1.54

Recently

1.53

Years

1.53

Since

1.52

Activations Density 0.355%

proper nouns or names mentioned in news articles

references to notable events or accomplishments

No Comments

No Known Activations