INDEX

Explanations

No Explanations Found

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

izable

-0.70

 reviewers

-0.64

rist

-0.63

LIN

-0.63

 crim

-0.63

Mass

-0.61

IGF

-0.60

annel

-0.60

 ANGEL

-0.59

 mamm

-0.59

POSITIVE LOGITS

PLIED

0.79

vertisement

0.72

ertodd

0.71

inator

0.71

uckland

0.68

Lago

0.68

aeda

0.67

 Pref

0.65

vette

0.65

pure

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.