INDEX

Explanations

elements related to computer programming and API functionality

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

code and programming 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

+#+#

-0.70

Erstellt

-0.54

WebElementEntity

-0.46

InSection

-0.45

KommentareTeilen

-0.45

lapsingToolbar

-0.45

#+#

-0.42

SBATCH

-0.42

+#+

-0.41

 <=",

-0.41

POSITIVE LOGITS

 ویکی‌پدی

0.52

 autorytatywna

0.49



0.48

省市镇

0.48

 مشين

0.46

 мәкал

0.46

ulemon

0.45

 Roskov

0.42

ſelf

0.41

adalajara

0.40

Activations Density 0.171%

elements related to computer programming and API functionality

code and programming terms

No Comments

No Known Activations