INDEX

Explanations

names and titles related to characters or roles

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

Jasenovac, Koppenhaver, Borten, rosten

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

ſelf

-0.51

 Indah

-0.51

 neceff

-0.51

Pró

-0.49

 mosca

-0.49

期刊论文

-0.49

 disambiguazione

-0.47

 raiſ

-0.47

ünst

-0.46

 ſta

-0.45

POSITIVE LOGITS

<<-

0.42

ente

0.41

 userRepository

0.40

ChildScrollView

0.39

kitten

0.38

WebServlet

0.38

enti

0.37

 nahilalakip

0.37

enten

0.35

hence

0.35

Activations Density 0.093%

names and titles related to characters or roles

Jasenovac, Koppenhaver, Borten, rosten

No Comments

No Known Activations