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

winner

-0.80

rist

-0.71

 Shant

-0.69

pecting

-0.68

ENG

-0.67

elcome

-0.66

arters

-0.65

 Shal

-0.65

 Antar

-0.64

arma

-0.62

POSITIVE LOGITS

 burns

0.85

 residues

0.79

 recess

0.72

 soil

0.69

iane

0.67

 injections

0.65

 pupils

0.65

 ripple

0.63

 plantations

0.63

 soils

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.