INDEX

Explanations

references to legal terms and conditions

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

surrounding anatomy

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

 surla

-0.64

UnitTesting

-0.49

 ModelRenderer

-0.48

 ✭✭

-0.45

 BrowserModule

-0.44

➟

-0.43

estros

-0.41

opsida

-0.41

 desertcart

-0.40

CppCodeGen

-0.40

POSITIVE LOGITS

__*/

0.45

الدراسه

0.44

StoreMessageInfo

0.43

addContainerGap

0.43

 schmerz

0.42

ThroughAttribute

0.39

recated

0.39

 ouder

0.39

BeginInit

0.38

脚注の使い方

0.37

Activations Density 0.163%

references to legal terms and conditions

surrounding anatomy

No Comments

No Known Activations