INDEX

Explanations

acronyms or specific terms related to a technology, organization, or product called "SN."

oai_token-act-pair · gpt-3.5-turbo

references to the television network "SN," likely related to show listings or network content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Archdemon

-0.72

tons

-0.64

headed

-0.62

Reviewer

-0.62

 Weinstein

-0.60

 unnamed

-0.60

 Reincarnated

-0.60

chart

-0.60

 fictional

-0.59

Pv

-0.59

POSITIVE LOGITS

SN

1.47

BOOK

0.90

ãĤ§

0.90

TY

0.88

ASH

0.87

ZI

0.84

ACK

0.83

KR

0.83

ERG

0.83

SN

0.81

Activations Density 0.004%

acronyms or specific terms related to a technology, organization, or product called "SN."

references to the television network "SN," likely related to show listings or network content

No Comments

No Known Activations