INDEX

Explanations

locations mentioned in news articles

oai_token-act-pair · gpt-3.5-turbo

locations, particularly cities and states, mentioned in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

-0.71

oton

-0.71

ocial

-0.61

 lucky

-0.58

¦

-0.56

bringer

-0.56

aiman

-0.56

lust

-0.56

vig

-0.56

igen

-0.55

POSITIVE LOGITS

 respectively

0.80

 Dept

0.78

etc

0.77

umbn

0.75

agara

0.62

ukong

0.61

 Springer

0.61

 Devon

0.60

 Leban

0.59

ilation

0.58

Activations Density 0.189%

locations mentioned in news articles

locations, particularly cities and states, mentioned in text

No Comments

No Known Activations