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

ISM

-0.68

isSpecialOrderable

-0.67

iculty

-0.64

ograp

-0.64

minist

-0.63

 encour

-0.63

elist

-0.63

ipop

-0.61

MRI

-0.59

ometers

-0.59

POSITIVE LOGITS

rip

0.73

agan

0.71

icks

0.70

uld

0.69

antha

0.69

Sul

0.69

 Gallup

0.66

 Roots

0.66

alla

0.66

nton

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.