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

 unchecked

-0.85

 intu

-0.74

tyr

-0.73

idable

-0.73

rypted

-0.72

 unsuspecting

-0.72

awa

-0.69

 Azerb

-0.69

 tresp

-0.68

 mathemat

-0.68

POSITIVE LOGITS

MET

0.85

atoon

0.74

inence

0.72

cipline

0.70

 attendance

0.70

rology

0.70

cal

0.66

NESS

0.65

Ã§

0.65

ECTION

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.