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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ´ãĥ³

-0.81

vous

-0.73

NetMessage

-0.72

 condu

-0.68

eat

-0.68

ewitness

-0.67

assadors

-0.65

gemony

-0.64

cham

-0.64

angular

-0.63

POSITIVE LOGITS

 PROGRAM

0.67

0.66

 Jaguar

0.64

0.62

chnology

0.61

0.60

 YEAR

0.60

nikov

0.59

 JUSTICE

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.