INDEX

Explanations

requests for help or information

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

questions to anyone

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 оригіналу

-0.49

feinander

-0.39

APORE

-0.39

 Chwiliwch

-0.38

ILDREN

-0.36

ndice

-0.36

bership

-0.34

NameInMap

-0.34

nô

-0.33

andafter

-0.33

POSITIVE LOGITS

 ویکی‌پدیا

0.55

AnchorStyles

0.49

urlpatterns

0.45

 useNavigate

0.45

guém

0.44

 président

0.44

Παραπομπές

0.43

 BaseActivity

0.43

 ainfi

0.43

 referenties

0.42

Activations Density 0.029%

requests for help or information

questions to anyone

No Comments

No Known Activations