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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rarity

-0.82

©¶æ

-0.71

taboola

-0.68

 Stern

-0.67

icles

-0.64

ghai

-0.64

 pregn

-0.63

 defective

-0.61

igslist

-0.61

advertisement

-0.60

POSITIVE LOGITS

Ts

0.68

illary

0.67

redo

0.66

anches

0.64

Had

0.63

<-

0.63

velt

0.62

shows

0.62

Bal

0.61

ahime

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.