INDEX

Explanations

images or captions that are meant to be enlarged or toggled

oai_token-act-pair · gpt-3.5-turbo

commands related to image manipulation features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

esses

-0.67

eday

-0.64

orkshire

-0.64

ivism

-0.60

ĪĴ

-0.58

nesia

-0.57

nov

-0.57

metic

-0.57

cellence

-0.57

uality

-0.55

POSITIVE LOGITS

 toggle

1.02

 caption

0.99

Enlarge

0.85

Pinterest

0.79

 captcha

0.75

 WATCHED

0.75

taboola

0.75

 "$:/

0.74

ãĤ´

0.71

[[

0.68

Activations Density 0.008%

images or captions that are meant to be enlarged or toggled

commands related to image manipulation features

No Comments

No Known Activations