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

aida

-0.80

 derail

-0.69

hiba

-0.68

akespeare

-0.67

ixie

-0.66

 directives

-0.65

 derailed

-0.64

hovah

-0.64

spection

-0.63

 indoctr

-0.63

POSITIVE LOGITS

Pool

0.82

CAR

0.78

FREE

0.76

Offline

0.73

eff

0.72

CM

0.69

tackle

0.69

à

0.68

RM

0.67

Ö¼

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.