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

)=(

-0.81

~~~~~~~~

-0.79

elaide

-0.72

Rect

-0.71

Feed

-0.68

ulations

-0.67

~~~~

-0.66

ä½¿

-0.64

WER

-0.64

Buyable

-0.62

POSITIVE LOGITS

ukemia

0.74

 Trem

0.73

ettle

0.66

stood

0.66

ocr

0.65

vit

0.64

edom

0.63

 Plat

0.63

 Pledge

0.62

ovie

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.