INDEX

Explanations

code-related elements within programming functions and variable manipulations

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

mathematical operations or properties

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

 intptr

-0.50

 ComVisible

-0.39

titleMargin

-0.33

联系

-0.33

conci

-0.32

 bertanggung

-0.30

Hentet

-0.30

«<

-0.30

paravant

-0.30

httphttps

-0.30

POSITIVE LOGITS

 الحره

0.56

 AssemblyVersion

0.56

 autorytatywna

0.55

tonsoft

0.52

SBATCH

0.52

دانشنامهٔ

0.51

ContentAsync

0.47

saraba

0.46

 oprot

0.46

 saites

0.44

Activations Density 0.257%

code-related elements within programming functions and variable manipulations

mathematical operations or properties

No Comments

No Known Activations