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

rush

-0.72

soDeliveryDate

-0.71

arma

-0.65

ploma

-0.65

 invitations

-0.63

psey

-0.62

ternity

-0.62

ovych

-0.62

 Marketable

-0.61

 shortages

-0.59

POSITIVE LOGITS

aido

0.81

trak

0.63

 Rapp

0.63

ameda

0.63

 Brun

0.62

 Perform

0.61

Cru

0.61

Û

0.60

ECK

0.59

Ear

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.