INDEX

Explanations

elements related to enhancement and adding uniqueness or variety to experiences

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

before "to" or "into"

np_acts-logits-general · gemini-2.0-flash

to a / into both

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

Personensuche

-0.67

ttal

-0.57

ناية

-0.56

からです

-0.56

Personendaten

-0.56

SuppressMessage

-0.55

ільки

-0.55

帖最后由

-0.54

Datuak

-0.54

-0.53

POSITIVE LOGITS

 proceedings

0.68



0.58

otherwise

0.58

umns

0.55

TextAppearance

0.54

 dreary

0.54

 ohnehin

0.54

 mundane

0.53

Complement

0.53

 stressful

0.52

Activations Density 0.180%

elements related to enhancement and adding uniqueness or variety to experiences

before "to" or "into"

to a / into both

No Comments

No Known Activations