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

acca

-0.89

eus

-0.88

ipel

-0.82

aceous

-0.76

ÏĤ

-0.75

velt

-0.75

è¦ļéĨĴ

-0.71

Äĩ

-0.70

nell

-0.69

ainer

-0.69

POSITIVE LOGITS

 Pledge

0.77

 Shack

0.75

 Published

0.70

 Transparency

0.67

TPP

0.66

conduct

0.66

URRENT

0.64

NPR

0.64

 RELEASE

0.64

 Kimber

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.