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

stals

-0.79

Ãī

-0.76

Lis

-0.74

 Hutch

-0.73

 Ames

-0.72

CTV

-0.72

stones

-0.71

urch

-0.71

 Spiegel

-0.68

Ireland

-0.66

POSITIVE LOGITS

ĵĺ

0.68

iquid

0.67

 implicit

0.65

ego

0.63

 negligent

0.63

reme

0.63

 deficient

0.63

adr

0.63

suc

0.63

 hippocampus

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.