INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

yip

-0.70

portion

-0.67

Pok

-0.64

dp

-0.61

Quantity

-0.61

 Grizz

-0.61

blance

-0.60

falls

-0.60

poons

-0.59

ħ

-0.59

POSITIVE LOGITS

 hypoc

0.76

enegger

0.68

emale

0.67

 philos

0.66

 photoc

0.66

 ********************************

0.64

ãĤ´ãĥ³

0.63

 hypocr

0.61

utenberg

0.61

 Manip

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.