INDEX

Explanations

the name "Lorraine."

oai_token-act-pair · gpt-3.5-turbo

the name "Lor" or related variations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ħĭ

-0.78

-0.77

STEM

-0.76

Cand

-0.71

growth

-0.71

 Dating

-0.68

manship

-0.66

eded

-0.66

sense

-0.64

Beat

-0.64

POSITIVE LOGITS

raine

1.13

ussia

1.06

ijn

0.96

wyn

0.90

adier

0.88

illard

0.87

adian

0.82

avin

0.79

 Lann

0.76

ideon

0.76

Activations Density 0.007%

the name "Lorraine."

the name "Lor" or related variations

No Comments

No Known Activations