INDEX

Explanations

occurrences of non-zero activation values indicating distinct elements in quantitative data

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

Code, math, or unusual characters

np_acts-logits-general · gemini-2.0-flash

tongue diagnosis, variables were, origins

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

extAlignment

-0.72

expandindo

-0.70

QMetaType

-0.68

 EconPapers

-0.67

 المعيارى

-0.64

>",

-0.61

AntiForgeryToken

-0.61

NameInMap

-0.59

parsedMessage

-0.59

 esternos

-0.59

POSITIVE LOGITS

ówczas

0.50

 télécommande

0.49

 entonces

0.49

 noastră

0.48

paddingRight

0.48

ő

0.48

出版年

0.47

 origini

0.46

StackNavigator

0.46

Hozzáférés

0.45

Activations Density 0.067%

occurrences of non-zero activation values indicating distinct elements in quantitative data

Code, math, or unusual characters

tongue diagnosis, variables were, origins

No Comments

No Known Activations