INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

BAT

-0.73

ONT

-0.72

OLD

-0.70

itar

-0.69

Lat

-0.68

SourceFile

-0.68

IRO

-0.68

 overdue

-0.67

ISION

-0.67

IOR

-0.67

POSITIVE LOGITS

ãĥĥãĥī

0.67

Ic

0.66

oshi

0.65

 Demons

0.65

 Invaders

0.64

 Reneg

0.64

 redes

0.63

querade

0.63

 Forth

0.60

 Scor

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.