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

 Tours

-0.71

 Tour

-0.70

vous

-0.67

UTC

-0.67

VIS

-0.67

 Planetary

-0.66

hack

-0.64

 Anniversary

-0.64

 Worlds

-0.64

 Forever

-0.63

POSITIVE LOGITS

flix

0.88

crim

0.78

 combust

0.73

 cured

0.71

Ļ½

0.64

 reversible

0.64

terness

0.64

seek

0.63

enf

0.62

hari

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.