INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

uries

-0.68

thora

-0.68

iciency

-0.62

zees

-0.61

 omega

-0.61

 casualty

-0.60

berman

-0.60

surv

-0.59

upiter

-0.58

Princ

-0.57

POSITIVE LOGITS

 Brach

0.73

aceae

0.71

 exting

0.69

âĿ

0.69

 Rebell

0.68

 toget

0.66

iation

0.65

 Kemp

0.64

 surpr

0.63

Kub

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.