INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Asuka

-0.68

 Drift

-0.68

 Arabs

-0.66

 Ceres

-0.65

quet

-0.65

 Crusher

-0.64

 Bahrain

-0.63

meter

-0.63

 Cooperation

-0.62

 Eaton

-0.62

POSITIVE LOGITS

terday

0.77

pire

0.73

yss

0.72

ieu

0.69

utral

0.68

essential

0.67

ilk

0.67

 thoughts

0.67

othal

0.64

inous

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.