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

zzle

-0.80

nell

-0.77

witz

-0.77

 Scrib

-0.75

ledge

-0.75

eph

-0.74

wich

-0.68

yright

-0.67

ourgeois

-0.67

bye

-0.65

POSITIVE LOGITS

ItemImage

0.70

respected

0.66

 ORIG

0.66

ivities

0.66

intent

0.64

overs

0.64

inches

0.60

Management

0.59

OVER

0.59

Const

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.