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

resses

-0.66

 holds

-0.64

aults

-0.63

ãĤº

-0.63

ãĥ«

-0.63

ival

-0.63

 Spaces

-0.62

ãĥ´ãĤ¡

-0.61

 Lies

-0.60

 supported

-0.60

POSITIVE LOGITS

umbn

0.83

Bey

0.82

DERR

0.82

ItemTracker

0.80

 Howell

0.78

FactoryReloaded

0.76

 deaf

0.73

alys

0.73

 disadvant

0.72

soDeliveryDate

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.