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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĨĴ

-0.67

 Commodore

-0.67

Center

-0.64

 Polaris

-0.63

 Chapters

-0.61

 Grade

-0.60

rix

-0.60

 Scarborough

-0.60

 Coliseum

-0.60

Union

-0.60

POSITIVE LOGITS

hw

0.77

etheus

0.76

phrine

0.75

hammer

0.73

houn

0.68

clad

0.68

ãģ®ç

0.68

isman

0.67

knife

0.66

azard

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.