INDEX

Explanations

the word "ey" with varying activation values

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ey" in text

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

mint

-0.77

ACTED

-0.75

istani

-0.73

dayName

-0.71

atoon

-0.70

å¥³

-0.70

ãĥķãĤ©

-0.64

CPC

-0.63

IFA

-0.63

å°Ĩ

-0.62

POSITIVE LOGITS

ey

1.04

ewitness

0.95

ield

0.89

esy

0.88

oshi

0.87

terness

0.86

estinal

0.86

ipel

0.82

er

0.80

outube

0.79

Activations Density 0.009%

the word "ey" with varying activation values

occurrences of the substring "ey" in text

No Comments

No Known Activations