INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Tycoon

-0.74

allion

-0.64

arth

-0.63

EEK

-0.58

ald

-0.58

icka

-0.58

chio

-0.56

eus

-0.56

 CASE

-0.55

iety

-0.54

POSITIVE LOGITS

ounces

0.73

velop

0.68

bats

0.67

ensen

0.67

paces

0.67

ographically

0.64

cms

0.64

runs

0.64

prints

0.63

ouncing

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.