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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yss

-0.81

nesota

-0.72

lege

-0.72

TAG

-0.71

="#

-0.68

LG

-0.66

cing

-0.65

ipop

-0.65

adoes

-0.65

ade

-0.65

POSITIVE LOGITS

 redacted

0.69

omething

0.66

 Saudis

0.66

 committees

0.65

 adul

0.64

lia

0.64

 kitchens

0.64

erous

0.62

 dissatisf

0.61

 Rack

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.