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

lly

-0.72

lees

-0.71

 Bailey

-0.71

 Primal

-0.68

 Cassidy

-0.68

 Casey

-0.67

 Clancy

-0.67

 Armory

-0.67

 ceilings

-0.66

 Lucy

-0.65

POSITIVE LOGITS

ersen

0.84

ongyang

0.75

)].

0.72

minist

0.68

osph

0.68

irteen

0.67

hedon

0.67

)]

0.66

utsch

0.66

æĺ

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.