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

olor

-0.93

 Fargo

-0.72

 Chero

-0.71

aq

-0.68

 Romo

-0.61

aucas

-0.61

Ã³

-0.61

 Amos

-0.60

Ez

-0.60

 Boyd

-0.58

POSITIVE LOGITS

RANT

0.86

udic

0.77

 BALL

0.75

ffen

0.74

chwitz

0.73

hung

0.73

 Abyssal

0.72

wered

0.70

zsche

0.69

userc

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.