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

APTER

-0.76

"]=>

-0.73

 ILCS

-0.69

SCP

-0.68

cases

-0.64

 traveller

-0.62

fty

-0.60

bill

-0.59

erity

-0.58

ESV

-0.57

POSITIVE LOGITS

obbies

0.77

ogun

0.75

ocard

0.70

ichick

0.69

rait

0.69

 guiActiveUn

0.68

eson

0.68

eton

0.64

 imperson

0.64

Ã§a

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.