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

ool

-0.69

place

-0.67

yles

-0.65

 scrimmage

-0.63

effic

-0.63

arming

-0.62

olis

-0.59

ools

-0.58

aint

-0.58

OOL

-0.58

POSITIVE LOGITS

 conflic

0.75

 captcha

0.74

Interstitial

0.73

 tremend

0.70

 Citiz

0.69

enance

0.68

etheless

0.68

 certified

0.67

AVG

0.66

gio

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.