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

Kent

-0.78

Sund

-0.72

Harris

-0.67

GAN

-0.65

Gro

-0.65

 Tribune

-0.65

 metic

-0.64

Pitt

-0.64

 Chains

-0.62

GF

-0.62

POSITIVE LOGITS

rahim

0.77

ibles

0.75

orsi

0.71

chwitz

0.68

ĳå£«

0.67

gio

0.67

ecause

0.66

ulkan

0.66

illas

0.65

reau

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.