© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 11194
    Prev
    Next
    INDEX
    Explanations

    mentions of male subjects and their identifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Masculinity and male pronouns

    np_acts-logits-general · gemini-2.0-flash

    he and his

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    :✨
    -0.45
    Groetjes
    -0.38
     שוליים
    -0.36
    CORBA
    -0.36
    setVerticalGroup
    -0.33
    stateProvider
    -0.32
    userdetails
    -0.32
    jsdelivr
    -0.32
    velocityY
    -0.31
     kasarigan
    -0.30
    POSITIVE LOGITS
     manly
    0.82
     męski
    0.81
     masculino
    0.78
     manhood
    0.78
     masculina
    0.77
     masculinity
    0.77
    Males
    0.75
     mascul
    0.75
     boyhood
    0.75
     masculinos
    0.74
    Activations Density 0.764%

    No Known Activations