INDEX

Explanations

actions related to physical contact or friction, such as scratching, grazing, clawing, or scraping

oai_token-act-pair · gpt-3.5-turbo

words related to scratching or surface-level interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ettel

-0.74

inen

-0.73

amide

-0.71

gur

-0.68

onse

-0.67

FI

-0.67

 Worldwide

-0.64

berus

-0.64

 Wikimedia

-0.64

ItemImage

-0.63

POSITIVE LOGITS

 scratching

1.24

 scratched

1.10

 scratch

1.06

 scratches

1.01

bats

0.86

 nails

0.84

 itch

0.83

bones

0.80

staking

0.77

paw

0.77

Activations Density 0.005%

actions related to physical contact or friction, such as scratching, grazing, clawing, or scraping

words related to scratching or surface-level interactions

No Comments

No Known Activations