INDEX

Explanations

mentions of specific organizations, authorities, and their roles in various contexts

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

Appears before nouns

np_acts-logits-general · gemini-2.0-flash

specific professions or entities

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-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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tinyos

-0.47

ніципалі

-0.46

MOQ

-0.46

 Empfindung

-0.46

ðsíða

-0.44

Vidite

-0.43

Revenir

-0.42

 podjela

-0.42

therlands

-0.41

ficulty

-0.41

POSITIVE LOGITS

SystemColors

0.40

出版年

0.36

 byteArray

0.36

Décès

0.36

 通販

0.35

hafter

0.35

们

0.35

Autoritní

0.35

ByUser

0.34

 initComponents

0.34

Activations Density 0.603%

mentions of specific organizations, authorities, and their roles in various contexts

Appears before nouns

specific professions or entities

No Comments

No Known Activations