INDEX

Explanations

phrases indicating useful tools or services for specific tasks

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

exceptional quality and distinction

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

 AssemblyCulture

-0.51

 gynhyrchwyd

-0.47

 Wicidata

-0.45

 wireType

-0.44

pushFollow

-0.40

extAlignment

-0.40

 Skin

-0.39

 &___

-0.39

лтемелер

-0.38

 Bioaccumulative

-0.38

POSITIVE LOGITS

 незавершена

0.51

Personendaten

0.43

 excellent

0.42

ConstraintMaker

0.41

 especially

0.41

 EXCELL

0.41

 excellence

0.41

 outstanding

0.41

jątk

0.41

 exceptionally

0.40

Activations Density 0.010%

phrases indicating useful tools or services for specific tasks

exceptional quality and distinction

No Comments

No Known Activations