INDEX

Explanations

references to executive roles and positions within organizations

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fromnode

-0.54

acamata

-0.53

 COLORS

-0.52

Datei

-0.50

 Parkplatz

-0.50

COLORS

-0.49

 Polda

-0.49

Kelas

-0.49

 nerds

-0.49

❹

-0.48

POSITIVE LOGITS

 executive

0.72

 useNavigate

0.62

 Executive

0.60

executive

0.60

ecutive

0.56

 EXECUTIVE

0.53

存于互联网档案馆

0.51

',

0.51

 Chwiliwch

0.50

 Exec

0.49

Activations Density 0.001%

references to executive roles and positions within organizations

No Comments

No Known Activations