INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

lar

-0.76

ma

-0.70

ãĥ¤

-0.68

dar

-0.66

oe

-0.65

eston

-0.64

interstitial

-0.64

phas

-0.64

found

-0.63

hur

-0.61

POSITIVE LOGITS

 livest

0.83

 conclud

0.73

 advoc

0.71

HCR

0.70

 Beir

0.69

 eleph

0.68

ppe

0.65

accompan

0.64

fty

0.64

 exorc

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.