INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 filings

-0.75

ħĭ

-0.72

isner

-0.69

urai

-0.66

uctor

-0.66

bernatorial

-0.66

Copyright

-0.65

phrine

-0.63

 fabricated

-0.63

)--

-0.61

POSITIVE LOGITS

idious

0.73

addons

0.69

 neut

0.68

iq

0.68

ku

0.66

alys

0.63

lain

0.63

ront

0.63

ween

0.62

oos

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.