INDEX

Explanations

terms related to decision-making and the consequences of choices

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

states or conditions

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

rzost

-0.50

 informée

-0.43

AndEndTag

-0.43

 विश्वसनीयता

-0.42

 Puedes

-0.34

 Ajust

-0.32

 editor

-0.32

sendMail

-0.32

 binding

-0.32

expires

-0.31

POSITIVE LOGITS

ChildScrollView

0.56

JvmStatic

0.54

SequentialGroup

0.52

Заключение

0.49

forChild

0.48

хьтан

0.47

Kjelder

0.47

Kesimpulan

0.47

istoitu

0.47

 PeEnEo

0.47

Activations Density 0.702%

terms related to decision-making and the consequences of choices

states or conditions

No Comments

No Known Activations