INDEX

Explanations

questions and prompts for audience engagement in discussions or feedback

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

Code, numbers, or symbols

np_acts-logits-general · gemini-2.0-flash

code syntax and numbers

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

PropertyChanging

-0.54

 Administrativna

-0.52

تقاوى

-0.51

mbggenerated

-0.51

GenerationType

-0.48

ArgsConstructor

-0.48

 resourceCulture

-0.48

 PRES

-0.46

RTGC

-0.45

 NavController

-0.45

POSITIVE LOGITS

 telefónica

0.43

новништво

0.41

 Italij

0.40

 navideña

0.39

AccessorTable

0.38

EndContext

0.38

 popolare

0.37

 allemaal

0.37

Последние

0.37

 perfeito

0.36

Activations Density 0.738%

questions and prompts for audience engagement in discussions or feedback

Code, numbers, or symbols

code syntax and numbers

No Comments

No Known Activations