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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vier

-0.74

otomy

-0.73

 Kund

-0.73

 Karn

-0.68

 Shutdown

-0.66

iott

-0.65

esthes

-0.64

 Angola

-0.63

arding

-0.63

 Arkham

-0.62

POSITIVE LOGITS

ĸļ

0.76

efully

0.68

cdn

0.68

lain

0.67

mble

0.64

maxwell

0.63

chens

0.63

ropolitan

0.62

erve

0.62

ribute

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.