INDEX

Explanations

proper nouns related to locations or organizations

oai_token-act-pair · gpt-3.5-turbo

references to a specific type of automobile or car-related topics

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

ymes

-0.71

umbnail

-0.69

eki

-0.68

krit

-0.66

ĨĴ

-0.65

ures

-0.65

uania

-0.65

iciency

-0.63

slow

-0.63

 clubhouse

-0.62

POSITIVE LOGITS

olina

1.20

negie

1.10

ousel

0.95

bons

0.92

ohydrate

0.89

oline

0.86

oled

0.86

acter

0.85

olyn

0.83

riers

0.83

Activations Density 0.025%

proper nouns related to locations or organizations

references to a specific type of automobile or car-related topics

No Comments

No Known Activations