INDEX

Explanations

strong affirmations of authenticity or truthfulness

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

truth or genuineness

np_acts-logits-general · gemini-2.0-flash

truth and genuineness

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

üb

-0.39

ไว้

-0.38

 previous

-0.37

ไว

-0.37

 foresee

-0.37

 vors

-0.36

 пом

-0.36

 previously

-0.36

 предпо

-0.35

はじめに

-0.34

POSITIVE LOGITS

truly

1.16

 truly

1.13

Truly

1.01

 echten

0.96

 Truly

0.96

genuine

0.96

 realmente

0.96

 verdadero

0.95

 verdadeira

0.95

 genuine

0.94

Activations Density 0.222%

strong affirmations of authenticity or truthfulness

truth or genuineness

truth and genuineness

No Comments

No Known Activations