INDEX

Explanations

words related to urban areas and cities

oai_token-act-pair · gpt-3.5-turbo

references to various types of "cities" or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

netflix

-0.74

syn

-0.73

isco

-0.73

empt

-0.72

uran

-0.71

coni

-0.70

thia

-0.69

×©

-0.69

aptic

-0.67

neck

-0.67

POSITIVE LOGITS

etter

0.98

hare

0.90

hell

0.75

hower

0.74

 Reserved

0.73

 Matters

0.70

aurus

0.70

afety

0.64

Peb

0.63

ities

0.63

Activations Density 0.019%

words related to urban areas and cities

references to various types of "cities" or locations

No Comments

No Known Activations