INDEX

Explanations

database-related terms and operations

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

after "Role" or abbreviations "RoleBranch"

np_acts-logits-general · gemini-2.0-flash

UserRoleBranch

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

 BorderRadius

-0.67

Hentet

-0.66

saraba

-0.66

webElementXpaths

-0.65

MLLoader

-0.63

 otomatig

-0.63

ſſung

-0.62

 հղումներ

-0.60

ſicht

-0.60

 ſind

-0.60

POSITIVE LOGITS

 silla

0.27

ing

0.25

 Näch

0.24

 thereof

0.24

 fungerar

0.24

 tiens

0.24

JoinTable

0.23

 recently

0.23

ɜ

0.23

 quanti

0.23

Activations Density 0.428%

database-related terms and operations

after "Role" or abbreviations "RoleBranch"

UserRoleBranch

No Comments

No Known Activations