INDEX

Explanations

content related to author attribution and manuscript-related phrases

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

the or a followed by a specific 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

 BorderSide

-0.56

InSection

-0.48

 Normdatei

-0.44

 Schi

-0.44

Synonymes

-0.42

illier

-0.42

AddHtmlAttribute

-0.42

angliski

-0.41

protoimpl

-0.40

 minimizar

-0.40

POSITIVE LOGITS

 noDo

0.48

 LIRE

0.48

 <<<<<<<<<<<<<<

0.47

 Paglinawan

0.47

adaptiveStyles

0.46

 JADX

0.46

 queſta

0.46

 preſent

0.45

 ſtate

0.45

 pleaſure

0.45

Activations Density 0.027%

content related to author attribution and manuscript-related phrases

the or a followed by a specific noun

No Comments

No Known Activations