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

ogie

-0.81

achus

-0.68

 Categories

-0.68

chwitz

-0.68

antage

-0.68

 Generic

-0.67

 Thumbnails

-0.66

 Archdemon

-0.64

phasis

-0.63

aic

-0.63

POSITIVE LOGITS

fin

0.69

icipated

0.67

 sworn

0.66

staking

0.64

 lent

0.64

aunts

0.61

ominated

0.60

ful

0.60

 invested

0.60

erent

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.