INDEX

Explanations

phrases mentioning specific names, possibly related to leadership or prominence

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'e'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ategory

-0.79

anova

-0.73

 sidx

-0.71

rooms

-0.71

irtual

-0.70

iets

-0.70

NESS

-0.68

milo

-0.68

 glim

-0.68

antha

-0.67

POSITIVE LOGITS

lements

1.37

gger

1.11

cki

1.06

gypt

1.04

zza

1.03

ck

1.03

agle

1.03

lev

1.02

agles

1.00

ld

1.00

Activations Density 0.035%

phrases mentioning specific names, possibly related to leadership or prominence

occurrences of the letter 'e'

No Comments

No Known Activations