INDEX

Explanations

specific numerical data or references related to metrics and scientific variables

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

citation markers

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

DockStyle

-0.79

الدراسه

-0.68

OGND

-0.63

 AssemblyCulture

-0.63

FunctionFlags

-0.63

 noDo

-0.62

HasAnnotation

-0.61

CloseOperation

-0.61

 تضيفلها

-0.60

цездатний

-0.60

POSITIVE LOGITS

 tierras

0.36

 térmico

0.35

 integrante

0.35

 diamantes

0.34

 évêque

0.34

 enfans

0.33

others

0.33

 diseños

0.33

Fish

0.33

 realice

0.33

Activations Density 0.112%

specific numerical data or references related to metrics and scientific variables

citation markers

No Comments

No Known Activations