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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vertisement

-0.70

owicz

-0.69

dyn

-0.69

 reluct

-0.64

 millenn

-0.64

raph

-0.61

uer

-0.61

Dyn

-0.60

iosyn

-0.60

gement

-0.59

POSITIVE LOGITS

lihood

0.79

SourceFile

0.74

£ı

0.72

ulic

0.71

 Entered

0.68

hei

0.67

 Flavoring

0.66

itzerland

0.66

Pwr

0.66

natureconservancy

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.