INDEX

Explanations

names of individuals or entities with high activations for various terms

oai_token-act-pair · gpt-3.5-turbo

proper nouns and technical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

PB

-0.70

 DIRECT

-0.67

 calendars

-0.66

DAY

-0.65

 microphones

-0.65

 Entered

-0.65

 fitness

-0.65

 ACTION

-0.65

 IMAGES

-0.65

 PRODUCT

-0.64

POSITIVE LOGITS

adish

1.00

opa

0.90

ek

0.88

ioch

0.87

eal

0.87

oad

0.87

inse

0.87

izoph

0.86

alm

0.84

obic

0.83

Activations Density 0.347%

names of individuals or entities with high activations for various terms

proper nouns and technical terms

No Comments

No Known Activations