INDEX

Explanations

references to subcategories or identifiers, particularly those that are prefixed with "Sub"

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

"Sub" or "subsection"

np_acts-logits-general · gemini-2.0-flash

"sub" prefix words

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

ValueStyle

-0.62

 Meksiku

-0.62

 jsPsych

-0.60

̀ng

-0.59

AxisAlignment

-0.58

yii

-0.57

 &___

-0.57

 Docs

-0.56

esm

-0.56

WindowConstants

-0.56

POSITIVE LOGITS

 OnDestroy

1.00

 getSub

0.67

tance

0.61

 Subs

0.58

Beneath

0.58

 виправивши

0.58

 giù

0.58

woofer

0.57

Билгалдахарш

0.57

 Sotto

0.57

Activations Density 0.211%

references to subcategories or identifiers, particularly those that are prefixed with "Sub"

"Sub" or "subsection"

"sub" prefix words

No Comments

No Known Activations