INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 largeDownload

-0.75

ams

-0.74

woman

-0.73

odo

-0.72

eker

-0.71

 redes

-0.67

aret

-0.65

etics

-0.65

mite

-0.63

xp

-0.61

POSITIVE LOGITS

 Faust

0.69

CTV

0.66

 Seym

0.66

 Archie

0.64

************

0.62

icago

0.62

Fel

0.61

EngineDebug

0.59

Els

0.57

Mrs

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.