INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to individuals, particularly names

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

arist

-0.68

bg

-0.67

quer

-0.66

chal

-0.63

ggle

-0.63

 dear

-0.61

eals

-0.61

semble

-0.61

UAL

-0.60

POSITIVE LOGITS

 Fowler

1.33

herty

0.81

ocity

0.81

Osw

0.80

reys

0.79

shire

0.78

asma

0.76

enhagen

0.75

owler

0.74

GF

0.72

Activations Density 0.007%

names of individuals

proper nouns related to individuals, particularly names

No Comments

No Known Activations