INDEX

Explanations

sequences of numerical data and identifiers

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

bitwise flags

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

 ویکی‌پدی

-0.48

脚注の使い方

-0.46

UserScript

-0.43

 ioutil

-0.40

Portály

-0.38

ChildScrollView

-0.38

RepeatedField

-0.38

VYMaps

-0.38

CURLOPT

-0.38

webElement

-0.37

POSITIVE LOGITS

protoimpl

0.57

 AttributeSet

0.46

baliknya

0.46

 gyhoeddwyd

0.44

FunctionFlags

0.42

出版年

0.41

writeFieldEnd

0.40

 történ

0.40

:+:

0.40

 arşivlendi

0.40

Activations Density 0.115%

sequences of numerical data and identifiers

bitwise flags

No Comments

No Known Activations