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

ĸļ

-0.69

 dominating

-0.65

Sm

-0.63

¶ħ

-0.63

INAL

-0.61

Sut

-0.61

 underlying

-0.60

ut

-0.60

Ut

-0.60

 Bund

-0.59

POSITIVE LOGITS

ryu

0.93

 Chocobo

0.84

 Presbyterian

0.82

ooters

0.74

xon

0.71

 Presbyter

0.71

alion

0.68

shire

0.67

olicy

0.67

arth

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.