INDEX

Explanations

mentions of knowledge and understanding in various contexts

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

she says

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

Personensuche

-0.78

transQ

-0.70

 ویکی‌پدی

-0.69

 noDo

-0.68

 kasarigan

-0.68

NameInMap

-0.67

BufferException

-0.65

 ModelExpression

-0.63

 ligiloj

-0.63

 Winaray

-0.62

POSITIVE LOGITS

 however

0.33

 plaisir

0.31

öz

0.30

 dég

0.29

 funny

0.29

 hingegen

0.28

 Pace

0.28

 pleasure

0.27

 amounts

0.27

Pace

0.27

Activations Density 0.052%

mentions of knowledge and understanding in various contexts

she says

No Comments

No Known Activations