INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

aldehyde

-0.80

amins

-0.80

imon

-0.73

 Drac

-0.70

agh

-0.68

claimer

-0.64

Dad

-0.64

 Builder

-0.62

draw

-0.61

abre

-0.59

POSITIVE LOGITS

 Nadu

0.77

 Chilean

0.71

 Chile

0.66

otom

0.66

 pepper

0.63

)=(

0.63

 Leone

0.62

SPONSORED

0.61

NCT

0.59

 Univ

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.