INDEX

Explanations

phrases indicating novelty or recent developments

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

newly or new followed by noun

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

 Genehmigung

-0.36

 alqu

-0.33

atyw

-0.32

 Bestand

-0.32

atorul

-0.31

 suena

-0.31

 Artikel

-0.31

lardır

-0.30

SuppressLint

-0.30

了出去

-0.30

POSITIVE LOGITS

 kaarangay

0.79

 newfound

0.69

UserScript

0.68

 المعيارى

0.63

RectangleBorder

0.63

IndentedString

0.62

featureID

0.62

 Wikimedijinoj

0.61

afficheront

0.60

RepeatedField

0.60

Activations Density 0.060%

phrases indicating novelty or recent developments

newly or new followed by noun

No Comments

No Known Activations