INDEX

Explanations

references to possession or ownership

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

possessive + noun

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

Rel

-0.33

 Стра

-0.31

manie

-0.31

ht

-0.29

readValue

-0.29

ps

-0.28

 jack

-0.28

irtschaft

-0.28

fie

-0.27

ya

-0.27

POSITIVE LOGITS

✨:

0.66

ConstraintMaker

0.64

 betweenstory

0.63

Искәрмәләр

0.61

featureID

0.59

 חיצוניים

0.59

NameInMap

0.57

opardy

0.57

ANTLR

0.56

 Мексичка

0.56

Activations Density 0.083%

references to possession or ownership

possessive + noun

No Comments

No Known Activations