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

 neigh

-0.70

north

-0.65

 darts

-0.64

à¼

-0.63

Ele

-0.62

comings

-0.61

>:

-0.61

aspberry

-0.61

///

-0.59

 neighbours

-0.59

POSITIVE LOGITS

soDeliveryDate

0.83

displayText

0.79

uz

0.75

 Brune

0.71

uchin

0.70

 ãĤµ

0.67

ĪĴ

0.67

ardless

0.65

edited

0.64

lite

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.