INDEX

Explanations

proper nouns or brand names

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

German or Germanic names

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 InputDecoration

-0.71

IntoConstraints

-0.66

 disambiguazione

-0.63

OCCURRED

-0.62

RegressionTest

-0.61

Personendaten

-0.61

NameInMap

-0.61

 يتيمه

-0.61

stuhl

-0.60

 Tasse

-0.59

POSITIVE LOGITS

deutig

0.44

geführten

0.31

 Handeln

0.30

 Савезне

0.30

jenige

0.28

 umgeben

0.28

 Hinsicht

0.28

 zagran

0.27

 czasie

0.27

 indipendente

0.26

Activations Density 0.183%

proper nouns or brand names

German or Germanic names

No Comments

No Known Activations