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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ILCS

-0.85

elist

-0.71

ģ«

-0.67

hack

-0.66

DragonMagazine

-0.63

-+-+

-0.63

 thous

-0.62

bly

-0.61

aleigh

-0.60

ĵĺ

-0.60

POSITIVE LOGITS

iddles

0.64

omsday

0.63

iverpool

0.63

wo

0.63

ernal

0.62

 regret

0.61

Gat

0.60

 Majesty

0.59

erv

0.58

onder

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.