INDEX

Explanations

references to body parts being touched or interacted with

oai_token-act-pair · gpt-3.5-turbo

possessive pronouns and related references to ownership or relationships

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

-0.85

ocument

-0.83

 Cosponsors

-0.79

ETF

-0.77

redits

-0.73

otin

-0.73

uph

-0.73

zsche

-0.72

inctions

-0.71

webkit

-0.70

POSITIVE LOGITS

own

1.48

 shoulders

1.29

 cheeks

1.28

 ears

1.28

 lips

1.27

 knees

1.21

 forehead

1.20

 chest

1.18

 throats

1.18

 throat

1.16

Activations Density 0.300%

references to body parts being touched or interacted with

possessive pronouns and related references to ownership or relationships

No Comments

No Known Activations