INDEX

Explanations

references to power dynamics and influence within political or social contexts

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

power, authority, and control

np_acts-logits-general · gemini-2.0-flash

wielding influence and authority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 leren

-0.42

 بهد

-0.36

ChildIndex

-0.35

 gericht

-0.35

 kebutuhan

-0.35

 pidana

-0.34

 caracteriza

-0.34

 sasaran

-0.34

 gagal

-0.33

ellini

-0.33

POSITIVE LOGITS

 influence

1.59

 influential

1.41

 power

1.41

influence

1.39

 Influence

1.37

Influence

1.33

 authority

1.29

influ

1.24

 INFLUENCE

1.21

 powerful

1.20

Activations Density 0.427%

references to power dynamics and influence within political or social contexts

power, authority, and control

wielding influence and authority

No Comments

No Known Activations