INDEX

Explanations

references to results, figures, and data in scientific contexts

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

presents findings and data

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

الحياه

-0.68

AndEndTag

-0.49

期刊论文

-0.46

ंदीखरीदारी

-0.46

 Taktlose

-0.46

uxxxx

-0.45

Personendaten

-0.45

Хьажоргаш

-0.44

 jsPsych

-0.44

المكان

-0.43

POSITIVE LOGITS

WebVitals

0.63

MigrationBuilder

0.54

ConstraintMaker

0.49

queryInterface

0.43

 defaultstate

0.41

istoitu

0.39

 initComponents

0.36

 vermelhas

0.36

 plateado

0.35

InsertCommand

0.35

Activations Density 0.586%

references to results, figures, and data in scientific contexts

presents findings and data

No Comments

No Known Activations