INDEX

Explanations

elements of deception and infiltration in narratives

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

suggesting ignorance or unawareness

np_acts-logits-general · gemini-2.0-flash

unaware of the truth

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sério

-0.46

mogat

-0.46

AndEndTag

-0.45

XmlRootElement

-0.45

Organisateur

-0.44

 Administrativna

-0.44

 vejo

-0.43

 creativos

-0.43

AndroidJUnit

-0.42

 Codable

-0.42

POSITIVE LOGITS

 oblivious

0.49

 unaware

0.44

AccessFile

0.42

 blind

0.42

 mistakenly

0.42

 unsuspecting

0.42

 unsus

0.41

 Dete

0.40

rub

0.40

 innocently

0.39

Activations Density 0.465%

elements of deception and infiltration in narratives

suggesting ignorance or unawareness

unaware of the truth

No Comments

No Known Activations