INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

vised

-0.76

PS

-0.75

 Dungeons

-0.75

insert

-0.74

keyes

-0.72

byter

-0.72

gar

-0.70

pots

-0.68

expl

-0.67

eps

-0.67

POSITIVE LOGITS

 navy

0.71

jew

0.65

 tariff

0.64

rush

0.63

rity

0.62

 coast

0.62

 levy

0.62

apo

0.61

WTO

0.61

 export

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.