INDEX

Explanations

proper nouns and names

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

WindowConstants

-0.52

 Савезне

-0.49

 autorytatywna

-0.49

LookAnd

-0.48

 esternos

-0.47

yntaxException

-0.47

GEBURTSDATUM

-0.46

expandindo

-0.44

 ویکی‌پدی

-0.43

 nakalista

-0.43

POSITIVE LOGITS

bootstrapcdn

0.58

탈

0.44

MediaStore

0.41

anonymous

0.40

 pseud

0.40

artist

0.39

เอง

0.39

makeText

0.39

님

0.39

 anonymous

0.38

Activations Density 0.413%

proper nouns and names

No Comments

No Known Activations