INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Hardy

-0.76

orem

-0.66

 Ones

-0.64

 Sessions

-0.63

Oak

-0.63

Warren

-0.61

 Bars

-0.61

ī

-0.58

meg

-0.57

ãĥ¤

-0.57

POSITIVE LOGITS

 preceded

0.80

 sugg

0.75

osponsors

0.72

anship

0.71

 correl

0.70

 predec

0.69

 governed

0.69

 subord

0.67

DragonMagazine

0.67

essage

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.