INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

etting

-0.76

idation

-0.72

anyahu

-0.68

ipping

-0.67

jriwal

-0.66

ongyang

-0.66

tering

-0.63

uling

-0.63

inent

-0.62

 earthqu

-0.62

POSITIVE LOGITS

Medic

0.78

php

0.73

DragonMagazine

0.72

workshop

0.69

Compat

0.68

ITNESS

0.68

GW

0.67

 Adren

0.66

Reviewer

0.65

KNOWN

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.