INDEX

Explanations

references to betrayal and themes of loyalty

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

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

 autorytatywna

-0.74

Autoritní

-0.71

 disambiguazione

-0.67

:✨

-0.65

 تانيه

-0.64

 informée

-0.63

存于互联网档案馆

-0.60

دانشنامهٔ

-0.60

 wikipagina

-0.58

CppMethod

-0.58

POSITIVE LOGITS

 redemption

0.34

0.31

TagMode

0.31

0.29

 protect

0.29

BoxShadow

0.29

########.

0.28

arrêter

0.28

Activations Density 0.469%

references to betrayal and themes of loyalty

No Comments

No Known Activations