INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

soDeliveryDate

-0.88

 Mahjong

-0.79

iversary

-0.78

osher

-0.73

 Robots

-0.72

ecause

-0.70

omon

-0.68

aker

-0.68

uesday

-0.67

lucent

-0.67

POSITIVE LOGITS

Sit

0.65

erial

0.63

Physical

0.63

Ale

0.61

Psy

0.60

Ni

0.58

 personalities

0.58

 equilibrium

0.58

pb

0.58

caster

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.