INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĳå£«

-0.91

enei

-0.70

istant

-0.70

IBLE

-0.69

vised

-0.69

cffff

-0.68

wered

-0.68

ancial

-0.67

Offline

-0.65

ible

-0.65

POSITIVE LOGITS

 Gavin

0.71

Sv

0.70

annis

0.65

 charism

0.65

 Luther

0.65

marks

0.63

Kate

0.62

Eva

0.61

GD

0.61

 Johann

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.