INDEX

Explanations

elements that represent emphasis or importance in text

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

Caret (^) followed by a number or symbol

np_acts-logits-general · gemini-2.0-flash

^ marked citations and special notation

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

 Wikimedijinoj

-0.50

<?

-0.49

Билгалдахарш

-0.42

national

-0.42

rizona

-0.42

documentclass

-0.42

zungs

-0.41

PhysRevD

-0.41

 funda

-0.41

oplasma

-0.41

POSITIVE LOGITS

帖最后由

0.71

 оригіналу

0.69

 للاسماء

0.69

WriteAttribute

0.68

AddTagHelper

0.68

@"/

0.68

CloseOperation

0.66

ValueStyle

0.63

 Wiktionnaire

0.63

 Савезне

0.62

Activations Density 0.035%

elements that represent emphasis or importance in text

Caret (^) followed by a number or symbol

^ marked citations and special notation

No Comments

No Known Activations