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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

achi

-0.78

cffffcc

-0.69

ilaterally

-0.69

rontal

-0.68

hani

-0.66

 facilit

-0.65

dayName

-0.63

ãĤ´ãĥ³

-0.63

hran

-0.62

quickShipAvailable

-0.62

POSITIVE LOGITS

ippery

0.68

apo

0.67

onite

0.66

coni

0.63

brook

0.62

orthy

0.59

tical

0.59

 Vehicle

0.58

peria

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.