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

senal

-0.76

Downloadha

-0.71

ichick

-0.71

anamo

-0.70

 laund

-0.64

acs

-0.63

itsch

-0.62

âĹ¼

-0.62

 Bills

-0.62

 Papers

-0.61

POSITIVE LOGITS

CHO

0.69

gravity

0.64

0.63

EF

0.63

iber

0.62

frac

0.62

anged

0.61

lis

0.61

fram

0.61

animate

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.