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

cash

-0.76

onga

-0.72

ezvous

-0.70

indust

-0.70

eers

-0.69

cart

-0.62

omics

-0.62

omic

-0.60

Tang

-0.60

eer

-0.59

POSITIVE LOGITS

 charism

0.85

lder

0.71

Pry

0.68

heast

0.65

 âĶľ

0.63

 Unsure

0.63

 sacrific

0.62

olated

0.61

 Presbyter

0.60

 Emin

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.