INDEX

Explanations

phrases starting with "To put" followed by an explanation or statement

oai_token-act-pair · gpt-3.5-turbo

phrases related to summarizing or rephrasing information

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

externalActionCode

-0.73

iller

-0.67

cius

-0.66

vill

-0.63

è¦ļéĨĴ

-0.63

COL

-0.63

KEN

-0.62

KY

-0.62

 violated

-0.60

Developer

-0.59

POSITIVE LOGITS

 aside

1.01

 together

0.99

 bluntly

0.93

ogether

0.86

 succinct

0.85

it

0.84

atively

0.80

 things

0.79

 plainly

0.76

hetically

0.75

Activations Density 0.055%

phrases starting with "To put" followed by an explanation or statement

phrases related to summarizing or rephrasing information

No Comments

No Known Activations