INDEX

Explanations

specific references to legal cases and court rulings

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

later or final

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oredCriteria

-0.42

GeneratedValue

-0.35

SOUNDBITE

-0.35

mycompany

-0.35

Хьажоргаш

-0.35

scriptcase

-0.34

WorkItem

-0.33

getRuntime

-0.33

Jereo

-0.33

FormUrlEncoded

-0.33

POSITIVE LOGITS

 surla

0.63

 للمعارف

0.60

 betweenstory

0.56

transQ

0.55

 AssemblyCompany

0.54

AsUp

0.50

पया

0.50

 defaultstate

0.50

RegressionTest

0.49

SBATCH

0.46

Activations Density 0.295%

specific references to legal cases and court rulings

later or final

No Comments

No Known Activations