INDEX

Explanations

references to specific journals and publications in scientific writing

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

journal articles and publications

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

principalColumn

-0.86

 للمعارف

-0.77

RTEX

-0.75

GEBURTSDATUM

-0.75

MessageOf

-0.74

SharedDtor

-0.73

IntoConstraints

-0.73

 kaarangay

-0.73

 MainAxisSize

-0.73

 AssemblyCompany

-0.72

POSITIVE LOGITS

 magazine

0.34

 journal

0.33

 manuscript

0.32

 publication

0.30

 Magazine

0.29

 series

0.29

magazine

0.28

 murs

0.28

COMMA

0.27

 published

0.27

Activations Density 0.074%

references to specific journals and publications in scientific writing

journal articles and publications

No Comments

No Known Activations