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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abase

-0.71

Osw

-0.70

ramid

-0.69

 epid

-0.68

 mascul

-0.65

 convol

-0.64

aucuses

-0.64

 centr

-0.64

Cec

-0.63

icians

-0.63

POSITIVE LOGITS

wire

0.81

nikov

0.73

nik

0.69

Battery

0.67

angler

0.67

bug

0.66

////////

0.66

iterator

0.64

Nik

0.63

 Deng

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.