INDEX

Explanations

references to court systems and judicial terms

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

court of

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Negative Logits

 disambiguazione

-0.85

 bureauc

-0.68

MemoryWarning

-0.67

testify

-0.66

<unused79>

-0.66

[@BOS@]

-0.66

<unused80>

-0.66

 للاسماء

-0.66

<unused74>

-0.66

<unused14>

-0.66

POSITIVE LOGITS

 integrada

0.35

intéress

0.33

 señores

0.33

 integrado

0.32

 właści

0.31

sterreich

0.31

相应的

0.31

 irré

0.31

 mediante

0.31

 trouverez

0.30

Activations Density 0.182%

references to court systems and judicial terms

court of

No Known Activations