INDEX

Explanations

entities related to healthcare and education across various contexts

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

health and professional roles

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

ViewFeatures

-0.52

 intptr

-0.46

diğini

-0.46

 feroit

-0.46

occasione

-0.45

 Depending

-0.43

diğinde

-0.43

 człowie

-0.43

 tekem

-0.43

 keber

-0.42

POSITIVE LOGITS

 authorities

0.58

 traders

0.54

 advisors

0.53

 sceptre

0.51

 insurers

0.50

 courtiers

0.50

 retailers

0.49

 experts

0.49

 dealers

0.49

 officials

0.48

Activations Density 0.085%

entities related to healthcare and education across various contexts

health and professional roles

No Comments

No Known Activations