INDEX

Explanations

news-related terms and phrases

oai_token-act-pair · gpt-3.5-turbo

references to local news and headlines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lyak

-0.77

Ã¨re

-0.77

Äĩ

-0.74

enario

-0.67

lihood

-0.67

rers

-0.66

lain

-0.65

Ã©n

-0.65

uchi

-0.65

 Enix

-0.64

POSITIVE LOGITS

 podcasts

0.91

 journalism

0.90

Subscribe

0.83

 Subscribe

0.83

 celeb

0.82

podcast

0.82

 weekday

0.78

 subscribe

0.78

 newsletter

0.77

scribe

0.76

Activations Density 0.396%

news-related terms and phrases

references to local news and headlines

No Comments

No Known Activations