INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Nem

-0.77

umbn

-0.74

 Acad

-0.73

 Painter

-0.72

Baz

-0.71

itsch

-0.69

bryce

-0.68

 Cosponsors

-0.67

Interstitial

-0.66

 rapp

-0.66

POSITIVE LOGITS

utable

0.78

factor

0.75

abyte

0.68

ersen

0.67

unch

0.66

ombo

0.64

 juicy

0.64

aneously

0.63

EVA

0.63

rarily

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.