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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ï¸

-0.67

EStreamFrame

-0.67

00200000

-0.65

acked

-0.64

umble

-0.64

 Change

-0.64

ystem

-0.63

sbm

-0.62

EV

-0.62

 wors

-0.61

POSITIVE LOGITS

stadt

0.86

eport

0.80

ogan

0.79

ogly

0.77

 Cohen

0.68

ardy

0.67

akin

0.64

ard

0.64

ulum

0.64

Ireland

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.