INDEX

Explanations

mentions of dogs

oai_token-act-pair · gpt-3.5-turbo

references to dogs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

oulos

-0.78

 Sins

-0.74

ayers

-0.71

Alz

-0.71

 Occupations

-0.70

éĹĺ

-0.70

ORN

-0.67

aukee

-0.65

farious

-0.65

tical

-0.65

POSITIVE LOGITS

dog

1.14

Dog

1.08

 dogs

1.05

meat

0.99

patch

0.98

pup

0.97

dog

0.95

 barking

0.92

 puppy

0.88

fighter

0.87

Activations Density 0.014%

mentions of dogs

references to dogs

No Comments

No Known Activations