INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

amily

-0.84

ertodd

-0.83

scene

-0.74

agame

-0.73

rities

-0.72

ride

-0.71

 lobby

-0.71

 quake

-0.69

ARP

-0.69

oshop

-0.66

POSITIVE LOGITS

áµ

0.82

sel

0.80

BuyableInstoreAndOnline

0.75

éŃĶ

0.69

 Negro

0.69

Sel

0.68

VEL

0.66

Merc

0.65

Sel

0.63

Raz

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.