© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14353
    Prev
    Next
    INDEX
    Explanations

    detection and prevention systems

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

    The neuron spikes on tokens that denote procedural actions or steps (e.g. “step,” “perform,” etc.).

    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
    yv
    -0.83
     квітня
    -0.81
    lsz
    -0.79
     kuliah
    -0.78
    opathie
    -0.78
     considère
    -0.78
     Pública
    -0.78
    therefore
    -0.78
     cetak
    -0.77
     spør
    -0.76
    POSITIVE LOGITS
    inib
    0.87
    Millan
    0.84
    🧪
    0.83
     diagonal
    0.83
     uniquement
    0.82
     Diagonal
    0.81
    plass
    0.81
    diagonal
    0.80
    I
    0.80
     Tonga
    0.79
    Activations Density 0.003%

    No Known Activations