© 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. 57395
    Prev
    Next
    INDEX
    Explanations

    impact statement

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

    This neuron reliably picks out formal document headings, titles, and other high-level metadata (e.g. section names, key headings, dates).

    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
    fotografie
    -0.73
     匿名
    -0.71
    Ὁ
    -0.70
     advised
    -0.69
    हाय
    -0.68
     fresas
    -0.68
     Strategie
    -0.68
    CreatedAt
    -0.67
     divina
    -0.66
     custos
    -0.66
    POSITIVE LOGITS
     EIR
    0.91
     recirc
    0.89
     circulation
    0.80
     Responses
    0.78
    responses
    0.78
    poxy
    0.76
     Eugenia
    0.76
     fibrillation
    0.75
     Jagger
    0.73
     responses
    0.73
    Activations Density 0.017%

    No Known Activations