INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

inventoryQuantity

-0.89

é¾įåĸļå£«

-0.81

çĶ°

-0.77

natureconservancy

-0.74

éĥ

-0.74

Ļ½

-0.72

»

-0.72

AppData

-0.70

åŃĲ

-0.70

çī

-0.70

POSITIVE LOGITS

eday

0.86

perty

0.81

irtual

0.75

ical

0.75

iami

0.71

 Gamble

0.67

 aggregate

0.66

etheless

0.63

icipated

0.63

 emanc

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.