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

yi

-0.67

 Civilization

-0.67

uxe

-0.67

steen

-0.65

BP

-0.64

 Deutsche

-0.64

iage

-0.64

WC

-0.63

Book

-0.63

Bey

-0.62

POSITIVE LOGITS

cedented

0.79

 VIDEOS

0.72

 sinks

0.64

 spat

0.63

ciating

0.62

 nutrit

0.62

 lung

0.62

chio

0.61

documented

0.59

 suscept

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.