INDEX

Explanations

references to legal petitions and associated terminology

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

legal petition context

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

Autowired

-0.48

 Blanc

-0.47

Usual

-0.46

∞

-0.44

ChildScrollView

-0.43

بالإنجليزية

-0.43

awful

-0.42

 Clara

-0.41

Horrible

-0.41

williams

-0.41

POSITIVE LOGITS

 petition

1.23

 Petition

1.14

 petitions

1.07

 PETITION

0.98

Petition

0.97

 petitioned

0.94

petition

0.91

titions

0.88

 Peti

0.72

 petitioner

0.69

Activations Density 0.023%

references to legal petitions and associated terminology

legal petition context

No Comments

No Known Activations