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

pmwiki

-0.85

Kem

-0.76

æ©

-0.68

ãĤ½

-0.67

ugen

-0.67

 Fraz

-0.67

igne

-0.64

 Saras

-0.63

nown

-0.63

 Route

-0.62

POSITIVE LOGITS

ornia

0.79

 livest

0.76

zona

0.74

 constitu

0.73

iations

0.72

ierrez

0.71

puter

0.68

iera

0.68

otypes

0.68

 subscriptions

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.