INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

soType

-0.74

ority

-0.73

Pand

-0.72

âķĲâķĲ

-0.72

çİĭ

-0.71

articles

-0.70

BuyableInstoreAndOnline

-0.69

ciples

-0.67

erver

-0.63

 Alone

-0.63

POSITIVE LOGITS

 Complex

0.74

?,

0.67

berra

0.66

 spor

0.61

aneous

0.60

ird

0.60

 Rober

0.59

region

0.59

trans

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.