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

sth

-0.72

bestos

-0.65

ihad

-0.65

leground

-0.62

eteenth

-0.60

uary

-0.59

 invented

-0.59

 resorted

-0.59

jan

-0.59

amara

-0.58

POSITIVE LOGITS

incinn

0.77

 Prospect

0.70

 Balance

0.69

 Atmospheric

0.68

 Sector

0.67

redo

0.67

obal

0.66

 âĨ

0.65

 Loch

0.64

 Atmosp

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.