INDEX

Explanations

legal terms and references associated with defendants and defenses in court cases

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

starting with "Def" and variations

np_acts-logits-general · gemini-2.0-flash

defending, defendants, defunct, defamatory

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

Autoritní

-0.74

 financière

-0.71

Демо

-0.66

 AttributeSet

-0.60

 nahilalakip

-0.59

 resourceCulture

-0.59

AnchorStyles

-0.58

 препратки

-0.58

Externé

-0.58

EndContext

-0.58

POSITIVE LOGITS

tely

0.86

teros

0.54

lated

0.51

 terms

0.50

roster

0.49

 quoi

0.49

mechanism

0.48

 atas

0.46

lating

0.45

tively

0.45

Activations Density 0.144%

legal terms and references associated with defendants and defenses in court cases

starting with "Def" and variations

defending, defendants, defunct, defamatory

No Comments

No Known Activations