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

rix

-0.65

ait

-0.64

 Sear

-0.63

ully

-0.63

sonian

-0.61

 subtitle

-0.60

 Blueprint

-0.60

ortium

-0.60

ulent

-0.59

 supra

-0.59

POSITIVE LOGITS

ãĥĻ

0.83

ework

0.71

=~

0.69

Ń·

0.68

 balloons

0.67

 Jagu

0.66

plings

0.64

poons

0.62

hots

0.62

 Swanson

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.