INDEX

Explanations

mentions of political or government positions, specifically the title "Minister"

oai_token-act-pair · gpt-3.5-turbo

mentions of government officials or ministers

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

vent

-0.84

feat

-0.80

eros

-0.68

ever

-0.68

eat

-0.67

 lightsaber

-0.65

gers

-0.64

 rive

-0.64

ilers

-0.64

DIS

-0.63

POSITIVE LOGITS

ial

0.99

ially

0.98

arians

0.83

onse

0.79

ials

0.77

 Minister

0.77

iture

0.77

arian

0.75

aceutical

0.73

itol

0.73

Activations Density 0.025%

mentions of political or government positions, specifically the title "Minister"

mentions of government officials or ministers

No Comments

No Known Activations