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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102558
    Prev
    Next
    INDEX
    Explanations

    life, role, experience, kill, save

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

    characters in narratives experiencing emotional or psychological states.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on words that denote social roles, titles, or statuses (e.g. Icon, Queen, Mother, Child, Orphan, God, Feminist, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     atento
    -1.36
    Discography
    -1.33
     السياس
    -1.32
    INERY
    -1.32
    gemeente
    -1.31
    ꒥
    -1.31
     prodot
    -1.30
    انظر
    -1.30
    𝒋
    -1.28
    limiento
    -1.26
    POSITIVE LOGITS
     something
    1.53
     about
    1.46
     anything
    1.44
     everything
    1.41
     to
    1.40
     (
    1.39
     básicamente
    1.38
     going
    1.37
     basically
    1.36
     being
    1.34
    Activations Density 0.057%

    No Known Activations