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

APD

-0.80

enegger

-0.69

Osw

-0.67

 particulars

-0.66

edIn

-0.65

emort

-0.65

çİĭ

-0.65

 ACTIONS

-0.64

 Adin

-0.64

addon

-0.62

POSITIVE LOGITS

kB

0.66

ij

0.64

amar

0.63

 Collider

0.63

 Chou

0.61

rica

0.60

onym

0.59

jar

0.59

 Karma

0.58

eous

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.