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

vec

-0.77

elta

-0.75

iaz

-0.68

ifax

-0.67

kowski

-0.67

 fide

-0.65

VIDIA

-0.65

 isot

-0.63

iegel

-0.62

gex

-0.60

POSITIVE LOGITS

Rush

0.72

Kings

0.68

ãĤ©

0.68

PokÃ©

0.67

lov

0.67

 Hobby

0.64

ITNESS

0.63

Winged

0.62

 Advice

0.62

Limited

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.