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

onite

-0.80

owitz

-0.75

 Morales

-0.72

 Rivera

-0.70

andon

-0.68

 Macron

-0.67

arro

-0.65

ano

-0.65

Rac

-0.64

 Morocco

-0.62

POSITIVE LOGITS

liction

0.66

pak

0.66

emis

0.66

vironment

0.66

 Aval

0.65

utenant

0.65

glomer

0.65

terness

0.64

hots

0.64

assetsadobe

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.