INDEX

Explanations

political and governmental terms or entities

oai_token-act-pair · gpt-3.5-turbo

political or governmental terms and concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 oneself

-0.91

SON

-0.64

 Topic

-0.61

 Palin

-0.60

ussions

-0.59

 Reincarn

-0.59

 myself

-0.57

 Crew

-0.57

 Anon

-0.57

 Singer

-0.57

POSITIVE LOGITS

iest

0.95

 woes

0.89

 population

0.80

 heritage

0.78

 legislature

0.75

 monarchy

0.73

 rulers

0.71

 destiny

0.71

 economy

0.70

 workforce

0.70

Activations Density 0.465%

political and governmental terms or entities

political or governmental terms and concepts

No Comments

No Known Activations