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

ainer

-0.70

Kiw

-0.66

road

-0.63

 diesel

-0.62

builders

-0.60

 HTTPS

-0.60

 UKIP

-0.60

 Reloaded

-0.60

born

-0.59

 surf

-0.58

POSITIVE LOGITS

Keefe

0.73

Pwr

0.73

emis

0.71

Ops

0.70

 Boyle

0.69

ivo

0.69

 Styles

0.69

buquerque

0.69

agara

0.68

 Foley

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.