INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Offline

-0.80

 Minute

-0.70

Geo

-0.67

inker

-0.66

 WATCHED

-0.66

 Cato

-0.65

Paw

-0.64

 Towns

-0.62

Â·Â·

-0.60

Jeb

-0.60

POSITIVE LOGITS

packing

0.90

skirts

0.78

quartered

0.73

 contrace

0.70

conserv

0.69

ãĥĺãĥ©

0.66

issance

0.66

isites

0.63

EStream

0.63

cknow

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.