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

 Speedway

-0.81

gew

-0.72

 Kills

-0.71

 Amin

-0.70

vind

-0.68

Wak

-0.66

 Tuls

-0.65

 resulting

-0.64

 Sorceress

-0.64

Fem

-0.64

POSITIVE LOGITS

pole

0.80

ocrine

0.73

iblical

0.73

urat

0.71

GD

0.70

Letter

0.70

advertising

0.67

legged

0.66

Europe

0.66

enance

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.