© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9812
    Prev
    Next
    INDEX
    Explanations

    Internal wallet

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

    The neuron is highly sensitive to numeric‐ or technical‐notation tokens—especially slashes used in fractions or measurements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     confused
    0.92
     manslaughter
    0.91
     frustrated
    0.88
     Վ
    0.88
     strangle
    0.86
     superfluous
    0.85
     Έ
    0.84
     vlá
    0.83
    щення
    0.82
     Ե
    0.82
    POSITIVE LOGITS
    eye
    0.74
    ln
    0.71
    egg
    0.70
    ringa
    0.69
    hil
    0.69
    erd
    0.68
    eyed
    0.67
    ema
    0.66
    त्तीस
    0.66
    forge
    0.65
    Activations Density 0.000%

    No Known Activations