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

æĹ

-0.70

hetic

-0.69

cliffe

-0.68

wit

-0.66

 gloom

-0.65

ulative

-0.65

Â·Â·

-0.64

ources

-0.63

generated

-0.63

eting

-0.62

POSITIVE LOGITS

orie

0.79

 Nurs

0.79

zman

0.65

onson

0.64

 Suite

0.63

 Swanson

0.63

emies

0.62

 satell

0.62

GoldMagikarp

0.62

 Twins

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.