INDEX

Explanations

instances of authorship or attribution in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

RegressionTest

-0.62

seamnă

-0.55

 ویکی‌پدی

-0.50

 podjela

-0.47

tagHelperRunner

-0.47

 estekak

-0.45

 operativos

-0.45

 aikaa

-0.44

 Allociné

-0.44

ReusableCell

-0.44

POSITIVE LOGITS

by

0.82

By

0.81

By

0.80

BY

0.72

by

0.67

BY

0.63

byn

0.53

CreatedBy

0.52

 createdBy

0.51

von

0.51

Activations Density 0.199%

instances of authorship or attribution in the text

No Comments

No Known Activations