INDEX

Explanations

people's last names ending in "ber" and related words

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the name "Heber."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

URA

-0.69

qqa

-0.68

 pity

-0.65

govtrack

-0.63

Ã¡s

-0.63

ENDED

-0.62

uras

-0.60

ourced

-0.59

ured

-0.59

 Cruel

-0.59

POSITIVE LOGITS

culosis

1.31

keley

1.24

deen

1.14

netic

1.07

ding

1.01

wald

0.98

punk

0.95

geon

0.89

wolves

0.87

lin

0.84

Activations Density 0.017%

people's last names ending in "ber" and related words

the repeated use of the name "Heber."

No Comments

No Known Activations