INDEX

Explanations

references to specific social media interactions and metrics

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

Comes after "user" at the start of turn

np_acts-logits-general · gemini-2.0-flash

graphical interface elements

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

Kariera

-0.40

ecution

-0.39

})`

-0.38

NextPage

-0.37

 especiales

-0.37

########.

-0.37

icated

-0.34

 sages

-0.34

 Pagina

-0.34

mistry

-0.33

POSITIVE LOGITS

twimg

0.84

httphttps

0.82

 snippetHide

0.65

UserScript

0.59

setupUi

0.53

 NSCoder

0.49

ModelSerializer

0.47

providedIn

0.46

 OnInit

0.46

ToFit

0.45

Activations Density 0.098%

references to specific social media interactions and metrics

Comes after "user" at the start of turn

graphical interface elements

No Comments

No Known Activations