INDEX

Explanations

terms related to institutional or structural concepts within various contexts

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

numbers and specific nouns

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

principalTable

-0.74

ViewFeatures

-0.58

Cyfeiriadau

-0.56

قایناق‌لار

-0.55

 Réponses

-0.53

 خارجية

-0.53

ViewImports

-0.52

PreExecute

-0.49

 onPostExecute

-0.48

 estekak

-0.47

POSITIVE LOGITS

 Савезне

0.41

 certificación

0.40

 coûts

0.39

 Komunikasi

0.38

 Biôgrafia

0.37

 mußten

0.37

agissait

0.35

⤹

0.35

nezeu

0.34

 ospiti

0.34

Activations Density 0.031%

terms related to institutional or structural concepts within various contexts

numbers and specific nouns

No Comments

No Known Activations