INDEX

Explanations

references to legal terminology and processes

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

legal context or specific actions

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ainfi

-0.57

 للمعارف

-0.57

 Biôgrafia

-0.53

orsese

-0.48

 istore

-0.47

 vectorielles

-0.45

 idéal

-0.44

 errno

-0.44

doub

-0.44

 rêver

-0.44

POSITIVE LOGITS

Vaya

0.43

Демографія

0.39

 Réponses

0.38

 تانيه

0.37

 StatelessWidget

0.37

PyTuple

0.36

djangoproject

0.35

Tazama

0.35

WindowConstants

0.33

cerne

0.33

Activations Density 0.256%

references to legal terminology and processes

legal context or specific actions

No Comments

No Known Activations