INDEX

Explanations

guidelines and tips for social etiquette and professional behavior

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Autoritní

-0.45

 CanadaChoose

-0.42

-0.41

'@/

-0.40

 concept

-0.38

inv

-0.38

TintMode

-0.38

openzeppelin

-0.38

┻

-0.37

fäl

-0.37

POSITIVE LOGITS

 yourself

0.68

 الحره

0.60

yourself

0.56

 yourselves

0.52

 jangan

0.51

 auffi

0.51

 your

0.49

 mümkün

0.49

 Yourself

0.47

 YOURSELF

0.46

Activations Density 0.325%

guidelines and tips for social etiquette and professional behavior

No Comments

No Known Activations