INDEX

Explanations

locations related to specific events or incidents

oai_token-act-pair · gpt-3.5-turbo

locations and events related to violence and tragedy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

obin

-0.84

xon

-0.80

zza

-0.77

ãĥĳ

-0.75

ÅŁ

-0.72

ullivan

-0.71

 ãĤµ

-0.71

rez

-0.68

ãĥĻ

-0.68

osexual

-0.66

POSITIVE LOGITS

 Colo

1.31

 Calif

1.31

Fla

1.14

 Tenn

1.14

Md

1.13

 Texas

1.06

Tex

1.03

TX

1.02

Ore

1.01

 Ohio

1.01

Activations Density 0.080%

locations related to specific events or incidents

locations and events related to violence and tragedy

No Comments

No Known Activations