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

erence

-0.75

ity

-0.73

beh

-0.72

ople

-0.66

anton

-0.65

ess

-0.65

nature

-0.65

uers

-0.64

 implementations

-0.64

otin

-0.63

POSITIVE LOGITS

 Continental

0.65

oku

0.63

 Simpsons

0.60

 Wizard

0.59

 Gloria

0.59

ulas

0.59

perty

0.58

 hoops

0.57

 Guilty

0.57

 colorful

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.