INDEX

Explanations

references to facial features and expressions

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

"face" or "faces"

np_acts-logits-general · gemini-2.0-flash

face and faces

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 kasarigan

-0.56

 Albany

-0.55

HasForeignKey

-0.53

ReusableCell

-0.52

ftagPool

-0.50

хьтан

-0.49

expandindo

-0.48

dataProvider

-0.48

 pétrole

-0.47

 المعيارى

-0.47

POSITIVE LOGITS

 face

2.25

Face

2.09

 Face

2.09

face

2.06

 FACE

1.97

 faces

1.81

FACE

1.81

 Faces

1.63

Faces

1.54

faces

1.53

Activations Density 0.090%

references to facial features and expressions

"face" or "faces"

face and faces

No Comments

No Known Activations