INDEX

Explanations

mentions of news outlets

oai_token-act-pair · gpt-3.5-turbo

references to news media outlets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Magnetic

-0.70

ston

-0.66

 saline

-0.65

 Magnet

-0.64

genic

-0.63

vol

-0.62

onics

-0.62

manuel

-0.59

potion

-0.57

pathic

-0.57

POSITIVE LOGITS

 outlets

0.99

etter

0.81

£ı

0.77

peak

0.76

 outlet

0.75

 ende

0.74

 pundits

0.74

 amplify

0.70

 inaccur

0.69

NPR

0.69

Activations Density 0.041%

mentions of news outlets

references to news media outlets

No Comments

No Known Activations