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

ustomed

-0.74

ufact

-0.71

 Fors

-0.69

~~~~

-0.67

ongh

-0.66

HL

-0.66

CN

-0.64

yrinth

-0.64

Hu

-0.64

 meaning

-0.63

POSITIVE LOGITS

 elector

0.67

raq

0.66

ando

0.62

 Diplom

0.60

wright

0.60

Offic

0.60

bridge

0.60

tip

0.60

rape

0.59

anus

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.