INDEX

Explanations

proper nouns, specifically names of places, characters, or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

 مشين

-0.74

SharedCtor

-0.64

ysty

-0.63

 varandra

-0.60

avigation

-0.60

neſs

-0.60

 ujednoznacz

-0.59

'{@

-0.59

ednes

-0.58

 محفوظة

-0.58

POSITIVE LOGITS

<bos>

0.59

0.56

RefNanny

0.54

en

0.47

als

0.46

 brand

0.46

 bazie

0.46

WarningLevel

0.46

iddhartha

0.45

 présent

0.45

Activations Density 0.076%

proper nouns, specifically names of places, characters, or organizations

No Comments

No Known Activations