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

 append

-0.70

 diets

-0.67

advertisement

-0.66

 wast

-0.64

 transit

-0.64

 menus

-0.62

 amen

-0.62

 autop

-0.62

 Upton

-0.62

 Blazers

-0.61

POSITIVE LOGITS

uces

0.88

leans

0.76

rophe

0.72

ange

0.69

anges

0.67

uate

0.66

uates

0.66

ament

0.65

antic

0.65

agher

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.