© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 7845
    Prev
    Next
    INDEX
    Explanations

    listing common words

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

    The neuron is primarily triggered by explicit numeric information (quantities, dates, years, counts) in the text.

    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
     formular
    0.79
     darkness
    0.78
     oversight
    0.77
    ah
    0.76
    a
    0.72
    キ
    0.71
    irik
    0.71
    /
    0.71
     Cultura
    0.70
     dold
    0.70
    POSITIVE LOGITS
    Когда
    0.93
    fänger
    0.86
    логи
    0.84
    Determine
    0.82
    single
    0.82
    nominated
    0.81
    regarded
    0.81
    nextToken
    0.80
    Define
    0.79
    terminated
    0.78
    Activations Density 0.001%

    No Known Activations