INDEX

Explanations

references to public policy and public opinion

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "public" and related concepts associated with public sentiment or opinion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

uana

-0.74

imov

-0.73

ï¸

-0.71

wered

-0.71

RANT

-0.70

xual

-0.70

Pg

-0.69

Scroll

-0.67

aundering

-0.67

AZ

-0.67

POSITIVE LOGITS

 relations

0.89

 opinion

0.89

 servants

0.88

 outcry

0.87

ised

0.85

relations

0.82

 appetite

0.82

 sector

0.80

izing

0.79

 treasury

0.79

Activations Density 0.044%

references to public policy and public opinion

occurrences of the word "public" and related concepts associated with public sentiment or opinion

No Comments

No Known Activations