INDEX

Explanations

references to political positions and titles within government

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 electing

-0.50

KommentareTeilen

-0.49

 ELECTION

-0.47

 Tortoise

-0.46

Schedulers

-0.46

Soorten

-0.45

 dissection

-0.45

 voter

-0.45

 election

-0.44

 ballot

-0.44

POSITIVE LOGITS

 Minister

1.69

 ministry

1.63

 minister

1.62

 Ministry

1.56

Minister

1.49

 ministers

1.41

 ministries

1.40

Ministry

1.36

 Ministers

1.36

minist

1.36

Activations Density 0.193%

references to political positions and titles within government

No Comments

No Known Activations