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

herer

-0.78

genic

-0.68

flix

-0.68

psy

-0.67

 Hein

-0.67

 centrif

-0.67

 Dian

-0.65

âĺħâĺħ

-0.62

 Angela

-0.62

 GREEN

-0.62

POSITIVE LOGITS

 condem

0.78

ovember

0.75

olo

0.72

 scrut

0.70

 warr

0.70

 plur

0.68

 Pill

0.65

 encount

0.65

uscript

0.64

ockets

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.