INDEX

Explanations

phrases related to news, especially fake news

oai_token-act-pair · gpt-3.5-turbo

references to news and discussions about fake news and its impact on society

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ayne

-0.75

Ĥª

-0.73

 Roses

-0.73

ptoms

-0.66

asio

-0.65

ription

-0.65

cair

-0.64

irlf

-0.64

rimp

-0.63

izo

-0.63

POSITIVE LOGITS

 dissemin

0.97

worthy

0.92

 dissemination

0.91

orial

0.86

 feeds

0.86

telling

0.84

fulness

0.83

feed

0.83

 spreads

0.82

rooms

0.78

Activations Density 0.187%

phrases related to news, especially fake news

references to news and discussions about fake news and its impact on society

No Comments

No Known Activations