INDEX

Explanations

mentions or references to cats

oai_token-act-pair · gpt-3.5-turbo

mentions of cats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

wana

-0.79

 Goddard

-0.79

omatic

-0.76

Roy

-0.75

 Transparency

-0.74

OOD

-0.73

alties

-0.72

Roy

-0.71

izers

-0.70

 Slovenia

-0.70

POSITIVE LOGITS

cat

3.65

 cats

2.46

Cat

2.34

Cat

2.30

cat

2.28

 Cats

1.92

cats

1.79

 kitten

1.75

CAT

1.72

dog

1.53

Activations Density 0.013%

mentions or references to cats

mentions of cats

No Comments

No Known Activations