INDEX

Explanations

references to legal principles and judicial processes

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

followed by adjectives

np_acts-logits-general · gemini-2.0-flash

very limited

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

engagent

-0.75

 kasarigan

-0.69

 Kontext

-0.65

 épaules

-0.64

 контек

-0.62

 collaboratively

-0.59

########.

-0.58

 genoux

-0.58

تقاوى

-0.58

 contexto

-0.57

POSITIVE LOGITS

pi

0.54

KURZBESCHREIBUNG

0.49

celer

0.47

getItemId

0.46

 serv

0.45

 marked

0.44

 diff

0.43

Württemberg

0.43

cheinend

0.43

tires

0.42

Activations Density 0.480%

references to legal principles and judicial processes

followed by adjectives

very limited

No Comments

No Known Activations