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

 tiss

-0.90

eatures

-0.89

 destro

-0.86

 challeng

-0.84

 millenn

-0.84

 DEFENSE

-0.80

 tremend

-0.79

entric

-0.79

enium

-0.76

osi

-0.75

POSITIVE LOGITS

 Boot

0.70

 Wolves

0.67

åİ

0.63

 Paste

0.62

me

0.62

 Colour

0.61

Fro

0.61

 Gret

0.59

 Wolf

0.59

shared

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.