INDEX

Explanations

specific terms and identifiers in a programming or technical context

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

Follows a word ending in "n" or contains "form"

np_acts-logits-general · gemini-2.0-flash

URL INET Publisher union form

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

 jangka

-0.33

-0.31

 geïsole

-0.31

collana

-0.31

 couvercle

-0.31

 COMMON

-0.29

 REPAIR

-0.29

 siir

-0.28

 aDecoder

-0.28

 pengusaha

-0.28

POSITIVE LOGITS

 autorytatywna

0.65

 defaultstate

0.57

 disambiguazione

0.57

StoryboardSegue

0.56

PDS

0.55

ScopeManager

0.53

الإنجليزية

0.53

Obrázky

0.52

pogon

0.52

تفصیلات

0.52

Activations Density 0.547%

specific terms and identifiers in a programming or technical context

Follows a word ending in "n" or contains "form"

URL INET Publisher union form

No Comments

No Known Activations

specific terms and identifiers in a programming or technical context

Follows a word ending in "n" or contains "form"

URL INET Publisher union form

No Comments

No Known Activations