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

 Sammy

-0.77

ĳå£«

-0.75

 Trayvon

-0.71

Mub

-0.70

 Straw

-0.68

 Leica

-0.66

 Anglo

-0.65

 flashbacks

-0.63

 Frenchman

-0.63

 butterflies

-0.63

POSITIVE LOGITS

trak

0.82

atche

0.77

catch

0.72

etz

0.72

ordinate

0.71

jong

0.70

 drainage

0.68

bite

0.67

ADVERTISEMENT

0.67

poon

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.