INDEX

Explanations

places or locations mentioned in a text, specifically those with "Street" in their name

oai_token-act-pair · gpt-3.5-turbo

references to locations or specific streets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

CVE

-0.78

CLASSIFIED

-0.68

BIL

-0.65

uning

-0.64

trak

-0.63

icho

-0.62

Anonymous

-0.62

VW

-0.62

Charges

-0.61

nery

-0.60

POSITIVE LOGITS

 globe

0.91

 entire

0.91

 plains

0.89

 aisle

0.87

 breadth

0.87

 river

0.85

 entirety

0.84

 country

0.84

 pond

0.83

 board

0.83

Activations Density 0.070%

places or locations mentioned in a text, specifically those with "Street" in their name

references to locations or specific streets

No Comments

No Known Activations