INDEX

Explanations

references to art, artists, and literary works

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

artist or author names and works

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

IRQn

-0.57

:✨

-0.54

ChildScrollView

-0.53

ConstraintMaker

-0.52

expandindo

-0.51

 ujednoznacz

-0.49

 Euer

-0.49

SwitchCompat

-0.49

InsertCommand

-0.48

 electrolux

-0.48

POSITIVE LOGITS

 connus

0.45

 famous

0.42

 known

0.40

 famosos

0.39

KNOWN

0.39

 famosa

0.38

 terkenal

0.38

famous

0.36

Known

0.35

Famous

0.35

Activations Density 0.025%

references to art, artists, and literary works

artist or author names and works

No Comments

No Known Activations