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

Paper

-0.73

ãĥª

-0.69

Rocket

-0.68

OME

-0.67

Jen

-0.67

ãĤŃ

-0.66

unn

-0.65

Pain

-0.64

ãĥĻ

-0.64

Grab

-0.64

POSITIVE LOGITS

hovah

0.70

ggle

0.68

cellence

0.68

jri

0.66

umenthal

0.65

erate

0.65

 diversity

0.65

 antitrust

0.64

MSG

0.64

mbuds

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.