INDEX

Explanations

numbers, particularly those related to scientific data or citations

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

journal volume numbers

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

 AssemblyVersion

-0.84

 Normdatei

-0.75

 للمعارف

-0.66

IsMutable

-0.65

 Administrativna

-0.64

fromnode

-0.63

UrlResolution

-0.60

 تكبرها

-0.60

 مشين

-0.59

 poffe

-0.57

POSITIVE LOGITS

 ochron

0.37

wnątrz

0.36

 productivo

0.36

 Wirkung

0.36

 głos

0.35

 táctil

0.34

 universidades

0.34

Perché

0.34

的问题

0.33

 wiedzy

0.33

Activations Density 0.258%

numbers, particularly those related to scientific data or citations

journal volume numbers

No Comments

No Known Activations