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

 trave

-0.73

iminary

-0.72

ecided

-0.72

cit

-0.71

NetMessage

-0.69

leans

-0.65

 Tsukuyomi

-0.65

irgin

-0.64

eton

-0.64

 exha

-0.63

POSITIVE LOGITS

0.67

backer

0.66

multipl

0.66

Ability

0.65

 isEnabled

0.65

MQ

0.63

FactoryReloaded

0.63

hack

0.63

Gameplay

0.63

Scope

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.