INDEX

Explanations

phrases related to decision-making and agency

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

survive scrutiny

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

 disambiguazione

-0.70

rungsseite

-0.69

abestanden

-0.68

 '\\;'

-0.64

 للمعارف

-0.62

 Audiodateien

-0.61

 snippetHide

-0.58

GEBURTSDATUM

-0.58

GraphicsUnit

-0.57

RenderAtEndOf

-0.57

POSITIVE LOGITS

 venner

0.35

friend

0.32

Friend

0.31

gatsby

0.31

Ha

0.30

erv

0.30

ern

0.30

 clienti

0.29

 serve

0.28

clientWidth

0.28

Activations Density 0.271%

phrases related to decision-making and agency

survive scrutiny

No Comments

No Known Activations