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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

photos

-0.95

edit

-0.75

zees

-0.72

 quotas

-0.71

nda

-0.69

âĺĨ

-0.69

 Tweet

-0.68

FK

-0.67

sy

-0.67

lov

-0.67

POSITIVE LOGITS

annabin

0.68

 safest

0.68

 coron

0.68

erald

0.66

 culmination

0.66

 sidelines

0.66

inois

0.65

regor

0.65

 heartbeat

0.65

 beauty

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.