INDEX

Explanations

key terms and variables in scientific discussions or reports

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

this variant or setting

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

 الحره

-0.41

sizePolicy

-0.37

HasForeignKey

-0.36

大さじ

-0.35

 elong

-0.34

 Carthag

-0.32

เปล

-0.32

 touch

-0.32

sol

-0.32

 IBOutlet

-0.31

POSITIVE LOGITS

 Biôgrafia

0.65

UserScript

0.63

 otomatig

0.59

Билгалдахарш

0.58

EndGlobalSection

0.58

ьаж

0.55

LookAnd

0.54

 InputDecoration

0.52

oneofs

0.51

 autorytatywna

0.50

Activations Density 0.028%

key terms and variables in scientific discussions or reports

this variant or setting

No Comments

No Known Activations