INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 welf

-0.82

 disadvant

-0.75

 thous

-0.74

 hemor

-0.73

 polarization

-0.72

 hormone

-0.71

aucus

-0.69

 adolesc

-0.66

 legislator

-0.66

 constit

-0.65

POSITIVE LOGITS

Dud

0.75

 Shoot

0.71

ULTS

0.70

 Written

0.69

dash

0.68

 Autob

0.68

 Furious

0.68

Autom

0.68

Steven

0.67

Bat

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.