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

exclusive

-0.72

FTWARE

-0.63

æ©Ł

-0.63

absolute

-0.62

ipes

-0.62

bows

-0.62

unal

-0.61

iery

-0.60

agher

-0.60

à

-0.59

POSITIVE LOGITS

 helicop

0.82

amera

0.69

 Carth

0.65

creen

0.63

apan

0.61

ktop

0.60

rior

0.60

 texted

0.59

 sailed

0.59

 withd

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.