INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

llah

-0.73

etheless

-0.69

cour

-0.67

Virgin

-0.67

 Prosecut

-0.67

 Furious

-0.66

selection

-0.66

 screenings

-0.64

Prosecut

-0.64

uph

-0.63

POSITIVE LOGITS

izen

0.73

 GOODMAN

0.68

unts

0.66

eret

0.64

pite

0.63

rawdownloadcloneembedreportprint

0.63

ierre

0.62

iency

0.62

emon

0.61

itches

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.