INDEX

Explanations

references to judicial proceedings and legal terminology

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

abbreviations and specific terms

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

Numéro

-0.35

 меры

-0.34

oda

-0.32

 mejorado

-0.32

 إي

-0.32

gesteld

-0.31

 financial

-0.31

<eos>

-0.31

 speranza

-0.31

 szá

-0.30

POSITIVE LOGITS

GTCX

0.65

脚注の使い方

0.63

AnimationsModule

0.63

 <=",

0.60

msgTypes

0.60

 jScrollPane

0.59

awaiter

0.59

GEBURTSDATUM

0.59

 Wicidata

0.58

 linkovi

0.58

Activations Density 0.082%

references to judicial proceedings and legal terminology

abbreviations and specific terms

No Comments

No Known Activations