© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69926
    Prev
    Next
    INDEX
    Explanations

    purpose or target

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

    The neuron fires on abstract nouns that denote conceptual “vehicles” or mechanisms (e.g. pathway, ground, Trojan horse, mediator, showcase).

    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
     there
    -1.59
     klaar
    -1.56
     onderdeel
    -1.51
    取った
    -1.50
     enchanting
    -1.47
     inspirational
    -1.46
    nosť
    -1.46
     it
    -1.43
     nyelv
    -1.43
    amını
    -1.40
    POSITIVE LOGITS
    to
    1.85
    PLICATION
    1.68
    us
    1.59
     vna
    1.57
    フライパン
    1.56
    橚
    1.55
    me
    1.55
    //
    1.52
     campion
    1.52
    ik
    1.51
    Activations Density 0.092%

    No Known Activations