INDEX

Explanations

terms associated with scientific studies and their findings regarding protective effects or impacts on health

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

notification judgment logging composition

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

GEBURTS

-0.51

Hentet

-0.39

delwed

-0.36

 miniaturka

-0.36

AnchorTagHelper

-0.35

 gelungen

-0.35

DBNull

-0.34

olev

-0.33

 geschehen

-0.33

 $('<

-0.33

POSITIVE LOGITS

 Italijanski

0.73

:✨

0.63

complexContent

0.60

 Administrativna

0.57

 Wikimedijinoj

0.54

 صوتيه

0.54

 beginnetje

0.53

 Италијани

0.52

 Chwiliwch

0.50

AnchorStyles

0.49

Activations Density 0.027%

terms associated with scientific studies and their findings regarding protective effects or impacts on health

notification judgment logging composition

No Comments

No Known Activations