© 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. 8972
    Prev
    Next
    INDEX
    Explanations

    modelurban mobility, power, models

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

    This neuron is basically a “position” detector: its activation grows roughly in proportion to how far a token is from the start of the input, i.e. it flags tokens by their absolute position in the sequence.

    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
    はや
    0.80
    ма
    0.73
    (\
    0.70
    ları
    0.69
    жу
    0.67
     Gann
    0.66
    専
    0.65
     staked
    0.65
    شو
    0.64
     underlies
    0.64
    POSITIVE LOGITS
     ativos
    1.00
     debido
    0.88
     alegria
    0.85
     металли
    0.84
     fevereiro
    0.82
    ковь
    0.81
     соответствии
    0.80
     adultos
    0.80
    চৈতন্যের
    0.80
    _',
    0.80
    Activations Density 0.000%

    No Known Activations