INDEX

Explanations

words associated with special features or characteristics

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

 виправивши

-0.52

 Biôgrafia

-0.49

 betweenstory

-0.46

mogat

-0.45

setVerticalGroup

-0.44

 inmigrantes

-0.40

 vlastní

-0.40

 risa

-0.39

地说道

-0.39

berdayakan

-0.39

POSITIVE LOGITS

:✨

0.90

ValueStyle

0.48

Hentet

0.45

 nahilalakip

0.41

NIP

0.40

 Oblig

0.39

 Nani

0.38

 AliExpress

0.38

amper

0.38

atify

0.38

Activations Density 0.000%

words associated with special features or characteristics

No Comments

No Known Activations