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

 embod

-0.71

ogle

-0.70

 prototyp

-0.69

 challeng

-0.68

 ornament

-0.65

 embell

-0.65

 expressive

-0.63

amorph

-0.63

 showc

-0.62

 pearl

-0.62

POSITIVE LOGITS

xes

0.68

 Michaels

0.66

zee

0.65

Stream

0.65

ONSORED

0.65

thia

0.65

Mich

0.64

uces

0.63

iquid

0.63

 urine

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.