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

cens

-0.80

Topics

-0.73

glomer

-0.70

 Afgh

-0.68

Published

-0.66

Topic

-0.65

ãĥ¬

-0.64

ollah

-0.64

etimes

-0.64

ertodd

-0.63

POSITIVE LOGITS

ourn

0.67

umber

0.66

illard

0.65

 Â®

0.64

cker

0.64

 Sail

0.64

oute

0.62

TT

0.62

bery

0.61

NAS

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.