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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

çīĪ

-0.90

VERTISEMENT

-0.80

LIA

-0.69

 Shinra

-0.68

æ©Ł

-0.66

Mae

-0.66

ONY

-0.65

--------------------------------------------------------

-0.64

Mia

-0.63

 Ùħ

-0.63

POSITIVE LOGITS

upiter

0.76

ropolitan

0.76

maxwell

0.74

earcher

0.74

folk

0.73

ederal

0.72

oult

0.72

ablishment

0.71

iscopal

0.71

ysc

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.