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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

details

-0.71

 entitlement

-0.67

 blues

-0.63

 Byrne

-0.60

 Pryor

-0.59

 understatement

-0.58

 apprehension

-0.58

Kam

-0.58

 Situation

-0.57

 referrals

-0.57

POSITIVE LOGITS

afort

0.90

elsius

0.83

odiac

0.77

spection

0.76

ividual

0.76

brance

0.75

 Canaver

0.75

zag

0.74

soDeliveryDate

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.