INDEX

Explanations

titles or headings that are followed by some text

oai_token-act-pair · gpt-3.5-turbo

empty tokens or segment boundaries in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 destro

-0.85

 exha

-0.77

 rounding

-0.77

 shorth

-0.71

 Hitman

-0.70

å£«

-0.69

 hemor

-0.67

ĪĴ

-0.67

 tightening

-0.65

 grooming

-0.65

POSITIVE LOGITS

ribune

1.37

urtle

1.31

utorial

1.28

ournament

1.25

itled

1.24

itles

1.23

aylor

1.23

olkien

1.19

weet

1.19

ravis

1.19

Activations Density 0.030%

titles or headings that are followed by some text

empty tokens or segment boundaries in the text

No Comments

No Known Activations