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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Confederation

-0.88

è¦ļéĨĴ

-0.78

ãĥ´ãĤ¡

-0.78

uner

-0.73

ãĤ´ãĥ³

-0.69

Secret

-0.68

efe

-0.68

PsyNetMessage

-0.67

 Grande

-0.66

ãĥį

-0.64

POSITIVE LOGITS

adult

0.65

dale

0.64

sets

0.64

iller

0.64

busters

0.63

entry

0.63

thinkable

0.62

 anatomy

0.62

comes

0.62

 wallpaper

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.