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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 104613
    Prev
    Next
    INDEX
    Explanations

    department, teams, identity

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

    The neuron fires on numeric literals (tokens that are numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     nhat
    -1.05
    修为
    -1.02
    ǹ
    -1.02
     mortes
    -1.02
     gema
    -1.00
    gradable
    -0.99
     territoriale
    -0.98
     elegante
    -0.98
     multiuso
    -0.97
    ℅
    -0.96
    POSITIVE LOGITS
     gender
    1.05
     religion
    0.98
     സ്
    0.91
     mikä
    0.91
     уже
    0.88
    嘣
    0.88
     kysy
    0.87
    ῆ
    0.87
     ethnicity
    0.86
     bluzka
    0.86
    Activations Density 0.008%

    No Known Activations