INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

âĹ¼

-0.69

udes

-0.68

isons

-0.67

 Metatron

-0.66

cia

-0.65

 Emin

-0.65

oses

-0.64

 GOODMAN

-0.64

aza

-0.63

ows

-0.63

POSITIVE LOGITS

 Disclosure

0.74

fare

0.70

ttle

0.64

HCR

0.64

etheless

0.63

 Pengu

0.59

dule

0.59

 Socket

0.58

 threshold

0.58

 GEAR

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.