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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

asse

-0.63

Ç

-0.63

 Garc

-0.61

 Holo

-0.60

 Eclipse

-0.60

Hib

-0.59

 defends

-0.59

ãĥĥãĥī

-0.58

Hik

-0.58

ebook

-0.58

POSITIVE LOGITS

 Flavoring

0.73

omon

0.72

gencies

0.68

ocation

0.68

ificent

0.67

 destro

0.67

DAY

0.67

 Cause

0.66

onday

0.66

yssey

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.