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

ynski

-0.84

ttle

-0.70

be

-0.68

asses

-0.67

leep

-0.67

gat

-0.66

atch

-0.64

opes

-0.63

ype

-0.63

ublic

-0.63

POSITIVE LOGITS

Ĥİ

0.75

eport

0.70

 è£ıè

0.68

 behavi

0.67

nels

0.66

Ò

0.65

 hemor

0.64

 Ripple

0.62

iversal

0.60

 nails

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.