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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

dream

-0.69

 imperson

-0.67

edin

-0.67

tein

-0.66

 payroll

-0.65

 droid

-0.64

erto

-0.63

dylib

-0.63

steel

-0.62

ten

-0.62

POSITIVE LOGITS

ijah

0.81

iru

0.74

 Nanto

0.69

icent

0.68

ãĥ©ãĥ³

0.67

yl

0.66

ureau

0.64

airy

0.64

gie

0.64

iae

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.