© 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. 34-GEMMASCOPE-RES-131K
    7. 92810
    Prev
    Next
    INDEX
    Explanations

    LICENSE binary form

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

    The neuron fires on the big “heading” tokens that introduce a new code or content section—i.e. prominent domain-specific keywords (like import paths, CSS functions, license identifiers, section titles) that label the start of a block.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     all
    -0.88
     stainless
    -0.84
    visor
    -0.84
     Tieren
    -0.84
    uxx
    -0.83
     there
    -0.82
    isation
    -0.81
    万人
    -0.79
     includes
    -0.78
     it
    -0.77
    POSITIVE LOGITS
    Kau
    0.92
    乜
    0.91
    athion
    0.88
    菊
    0.84
    ーマ
    0.84
    genossen
    0.83
    Embrace
    0.82
    礡
    0.82
    Nasc
    0.82
     محت
    0.81
    Activations Density 0.001%

    No Known Activations