INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Hemisphere

-0.75

 Tend

-0.74

ĸļ

-0.73

 Editors

-0.72

amy

-0.71

 Associ

-0.69

 Retrieved

-0.65

mare

-0.64

imum

-0.64

Rio

-0.63

POSITIVE LOGITS

plementation

0.75

 disarm

0.73

stro

0.71

endum

0.68

endor

0.68

agos

0.67

emo

0.66

arb

0.66

lance

0.65

capital

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.