INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

aughs

-0.85

icter

-0.83

imation

-0.77

acca

-0.77

enthal

-0.68

ombies

-0.68

aden

-0.66

artney

-0.65

coni

-0.64

aleb

-0.64

POSITIVE LOGITS

 Institution

0.75

Phi

0.68

dp

0.62

ente

0.60

 watering

0.58

istor

0.57

Nun

0.57

gd

0.57

 Univ

0.57

 Glad

0.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.