INDEX

Explanations

references to Fox News and related terms

oai_token-act-pair · gpt-3.5-turbo

references to Fox News

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

othing

-0.70

acterial

-0.68

ysis

-0.67

isu

-0.67

utic

-0.66

chest

-0.65

Ney

-0.65

 expressive

-0.65

val

-0.64

onement

-0.64

POSITIVE LOGITS

 Channel

1.25

Radio

0.95

 Radio

0.93

Channel

0.92

 Magazine

0.89

Talk

0.88

 pund

0.86

 Networks

0.85

News

0.84

 commentator

0.82

Activations Density 0.033%

references to Fox News and related terms

references to Fox News

No Comments

No Known Activations