INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

âĢ¢âĢ¢

-0.70

ĳå£«

-0.68

 funeral

-0.66

âĢ¢âĢ¢âĢ¢âĢ¢

-0.65

 Icar

-0.65

Ń·

-0.64

Hes

-0.64

IPS

-0.64

 Rumble

-0.64

 farewell

-0.63

POSITIVE LOGITS

vey

0.80

cheat

0.72

uan

0.65

ahime

0.64

ancial

0.64

ams

0.61

ricanes

0.61

sei

0.61

senal

0.61

earcher

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.