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

BN

-0.85

vid

-0.67

weet

-0.67

bour

-0.63

pell

-0.63

pent

-0.63

fram

-0.60

pelling

-0.60

price

-0.60

 Century

-0.59

POSITIVE LOGITS

ertodd

0.92

anooga

0.77

zai

0.75

 triv

0.74

 srfAttach

0.73

aeus

0.69

 Adinida

0.68

iculty

0.67

ouf

0.67

emi

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.