© 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. 71343
    Prev
    Next
    INDEX
    Explanations

    HMAC, file paths, articles, study pilots

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

    The neuron activates on numeric literal tokens (multi‐digit numbers) in code.

    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
     tornillo
    -0.91
    标签
    -0.85
     Saat
    -0.84
     Etat
    -0.82
     Kø
    -0.81
    ᅠ
    -0.81
     cuivre
    -0.81
     jūs
    -0.80
     coussin
    -0.80
    rédients
    -0.79
    POSITIVE LOGITS
    主婦
    0.85
     někter
    0.82
    KURZBESCHREIBUNG
    0.78
    chita
    0.77
     пі
    0.76
    transitions
    0.73
     kebanyakan
    0.73
     desejo
    0.73
    おもちゃ
    0.72
    . 
    0.70
    Activations Density 0.026%

    No Known Activations