© 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. 34-GEMMASCOPE-RES-131K
    7. 102667
    Prev
    Next
    INDEX
    Explanations

    Shipping, Symptoms, Pitch, FOR

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

    The neuron is tuned to spot numeric tokens (digits and number sequences) in structured or code‐like text (e.g. version numbers, IDs, counts in YAML).

    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
    QM
    -0.87
    endpush
    -0.85
    ṓ
    -0.77
     Wass
    -0.77
    وا
    -0.76
    ciless
    -0.76
    ünden
    -0.74
     Mano
    -0.73
    fif
    -0.72
    лів
    -0.72
    POSITIVE LOGITS
    notifier
    0.87
    salud
    0.82
    দ্
    0.79
    grandes
    0.77
    dacia
    0.76
    downloaded
    0.75
    jobb
    0.75
    erati
    0.75
    lantern
    0.75
    cyjny
    0.74
    Activations Density 0.033%

    No Known Activations