INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĤª

-0.77

OTA

-0.77

 Winter

-0.74

ORPG

-0.72

 Forth

-0.72

 Rite

-0.71

-+

-0.71

ģĸ

-0.71

days

-0.70

Winter

-0.70

POSITIVE LOGITS

uana

0.83

 pane

0.75

agine

0.71

rit

0.68

lig

0.65

neum

0.65

ugi

0.65

 impover

0.65

acca

0.64

amel

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.